Composition Software...?

Hey everyone...

Just curious, what do people use nowadays to compose? I've been doing
things using Sonar 7 in 72-tet by using 6 MIDI channels detuned by the
right amount. As you might imagine, this is a huge pain to have to
deal with, and I might imagine there's an easier way that people are
working with.

I did some searching of the archives and I found some notation
software, but nothing stuck out at me for composition. Can anyone
point me in a better direction?

Thanks,
Mike

I use blue[1], but it is specific to Csound. It is graphically-oriented and doesn't utilize notation (which I don't need or use for my electronic music anyway). There is direct support for microtonal scales (Scala format) in the Piano Roll tool. And of course Csound can do anything pitch-wise that you can possibly imagine, but it's certainly not everyone's cup of tea.

When I want to use notation, I use Noteworthy Composer[2], which is very inexpensive ($39) and can play or generate MIDI files. Nowhere near Sibelius or Finale in terms or power or flexibility, but very good. No direct support for microtonal scales, but you can use Scala or Fractal Tune Smithy (among others) to retune the MIDI.

- Dave

[1] http://www.csounds.com/stevenyi/blue/
[2] http://www.noteworthysoftware.com/

Mike Battaglia wrote:
> Hey everyone...
> > Just curious, what do people use nowadays to compose? I've been doing
> things using Sonar 7 in 72-tet by using 6 MIDI channels detuned by the
> right amount. As you might imagine, this is a huge pain to have to
> deal with, and I might imagine there's an easier way that people are
> working with.
> > I did some searching of the archives and I found some notation
> software, but nothing stuck out at me for composition. Can anyone
> point me in a better direction?
> > Thanks,
> Mike

--- In MakeMicroMusic@yahoogroups.com, "Mike Battaglia"
<battaglia01@...> wrote:
>
> Hey everyone...
>
> Just curious, what do people use nowadays to compose? I've been doing
> things using Sonar 7 in 72-tet by using 6 MIDI channels detuned by the
> right amount. As you might imagine, this is a huge pain to have to
> deal with, and I might imagine there's an easier way that people are
> working with.
>

I use a keyboard that can be run in "split" mode (upper half and lower
half of the keyboard send to different Midi channels) and a synth that
can detune each Midi channel differently. This enables me to play in
realtime up to 24 notes per octave with not too much pain. I know
there are keyboards that support up to 4 split areas (such as the
M-Audio Keystation Pro 88), so the same procedure could be used for up
to 48 notes pro octave - but with increasing pain, I imagine - and
still not 72-tet.

> I did some searching of the archives and I found some notation
> software, but nothing stuck out at me for composition. Can anyone
> point me in a better direction?
>

I think Band in a Box (http://www.band-in-a-box.com) might by
something in this direction. I don't know it myself, though, and dunno
whether it supports microtonality.

I currently work as a beta-tester for a commercial composition
software that is called "Music Prototyping Studio"
(http://www.cognitone.com). Very powerful, definitely - but, as you
might suspect, not really made for microtonality. It can be brought to
work in microtonal contexts, though - if you need not more than 12
tones per octave.
--
Hans Straub

At 02:25 AM 6/14/2008, you wrote:
>Hey everyone...
>
>Just curious, what do people use nowadays to compose? I've been doing
>things using Sonar 7 in 72-tet by using 6 MIDI channels detuned by the
>right amount. As you might imagine, this is a huge pain to have to
>deal with, and I might imagine there's an easier way that people are
>working with.
>
>I did some searching of the archives and I found some notation
>software, but nothing stuck out at me for composition. Can anyone
>point me in a better direction?
>
>Thanks,
>Mike

Hiya Mike,

If you want a score editor, some members here have said very
good things about:

http://www.myriad-online.com/en/products/harmony.htm

But I haven't tried it myself.

-Carl

Mike,

--- In MakeMicroMusic@yahoogroups.com, "Mike Battaglia"
<battaglia01@...> wrote:
> Just curious, what do people use nowadays to compose? I've been doing
> things using Sonar 7 in 72-tet by using 6 MIDI channels detuned by the
> right amount. As you might imagine, this is a huge pain...

Ow. That makes my head hurt.

I've been using Sonar since V3, and now up to V7. I utilize a number
of softsynths and softsamplers, virtually all of which can load Scala
tuning files, so I simply retune the instruments, and don't have to
muck about with channelization and the rest of the workarounds.

It doesn't help you if there is an instrument you *really* like that
can't be microtuned, but I've either come close or samples notes and
then put them into some form of sample format that could be used by a
microtunable instrument.

The other thing is getting used to using a standard 7-5 keyboard for
input, but I just don't worry about it too much. If I was either a
better keyboardist, or a more methodical/analytical composer, it might
matter more. Fortunately I'm neither.

HTH,
Jon

Hello everyone, and thanks for the insights. I'll check that out.

Jon:
> I've been using Sonar since V3, and now up to V7. I utilize a number
> of softsynths and softsamplers, virtually all of which can load Scala
> tuning files, so I simply retune the instruments, and don't have to
> muck about with channelization and the rest of the workarounds.

What samplers are you using? Right now I'm using East West with Native
Instruments' Kompakt sampler, although I do have a version of Kontakt
around somewhere if that would work. Do these samplers work with more
than 12 notes to the octave? Or can they only use 12-note subsets of
extended tunings?

> It doesn't help you if there is an instrument you *really* like that
> can't be microtuned, but I've either come close or samples notes and
> then put them into some form of sample format that could be used by a
> microtunable instrument.

Honestly, all I really need is something that will let me just write
in 72-tet or 53-tet or 19-tet or BP if I want. If I can do that, and
actually hear what it would sound like (by playing in what I write -
do these programs have some sort of staff notation component? because
that's usually how I write), then if I can just save it as a standard
GM file then I'll have no problem loading it into EastWest to get some
high quality sounds into it.

> The other thing is getting used to using a standard 7-5 keyboard for
> input, but I just don't worry about it too much. If I was either a
> better keyboardist, or a more methodical/analytical composer, it might
> matter more. Fortunately I'm neither.

Ah, screw it. Someday I'll get my AXiS and live happily ever after.
For now, I'm happy if I can use a mouse.

-Mike

Mike wrote:
>What samplers are you using? Right now I'm using East West with Native
>Instruments' Kompakt sampler, although I do have a version of Kontakt
>around somewhere if that would work.

If you have Kontakt then you need:
http://12equalboresme.com/Scala2Kontakt

>Ah, screw it. Someday I'll get my AXiS and live happily ever after.
>For now, I'm happy if I can use a mouse.

Actually, extended keyboards are further along than extended
notation systems.

Many microtonalists have at least tried using the "piano roll"
view in their MIDI sequencer as a direct composing tool, and
several have had good results. The other main options seems to
be Myriad Harmony Assistant (see my prev. message) and Finale.
Lsat I heard, Sibelius was still a pretty bad hack, but that
might have changed.

Oh, and there is typing in a text-based format directly, like
microabc.
http://br.geocities.com/hfmlacerda/abc/microabc-about.html

-Carl

--- In MakeMicroMusic@yahoogroups.com, "Mike Battaglia"
<battaglia01@...> wrote:
>
> Hey everyone...
>
> Just curious, what do people use nowadays to compose? I've been doing
> things using Sonar 7 in 72-tet by using 6 MIDI channels detuned by the
> right amount. As you might imagine, this is a huge pain to have to
> deal with, and I might imagine there's an easier way that people are
> working with.
>
> I did some searching of the archives and I found some notation
> software, but nothing stuck out at me for composition. Can anyone
> point me in a better direction?
>
> Thanks,
> Mike
>

Hi Mike. I'm currently programming a free, open-source sequencer for
unlimited extended just intonation. I haven't uploaded a version yet,
but the page is here:
http://www.sourceforge.net/projects/rationale
I'll put an announcement here once I get a fairly complete version up.

It's probably not too useful for 72-tet in this form, but if there's
enough interest, I could probably program a sequencer for ET tunings.
In my experience MIDI is useless for microtuning. Far better to set
something up with OSC, which is more customizable. Alas, the options
for non-programmers to use OSC are limited right now.
One solid option is to use/create frontends to Csound, like Blue,
which Dave mentioned. Csound can load soundfonts and micro-tune them
without the one-pitchbend-value-per-channel limitation. It is a dream
come true.

If anyone has a micro-tunable sequencer that can send OSC, or that is
open-source, I'd be interested in helping to set it up to play
soundfonts using Csound.

-Chuckk

--- In MakeMicroMusic@yahoogroups.com, "Mike Battaglia"
<battaglia01@...> wrote:
>
> Hey everyone...
>
> Just curious, what do people use nowadays to compose? I've been doing
> things using Sonar 7 in 72-tet by using 6 MIDI channels detuned by the
> right amount. As you might imagine, this is a huge pain to have to
> deal with, and I might imagine there's an easier way that people are
> working with.
>
> I did some searching of the archives and I found some notation
> software, but nothing stuck out at me for composition. Can anyone
> point me in a better direction?
>
> Thanks,
> Mike

Hi Mike,

I did a good chunk of my work in custom software I now call
'micro_composer'. It used to be called 'et_compose', which in a way is
a more accurate description, b/c I designed it for n-edos. However,
b/c one can simulate JI with large edos, I liked the general term.

One enters an ascii pitch and rhythmic notation, and some other
parameters are possible, like dynamics and tempi, and you can also
humanize the timing and dynamics using gaussian noise functions. The
program (a Python script) parses the input file, changing it to an
intermediate text representation of a MIDI file, and uses 't2mf' togo
the next step of giving you the MIDI file output, which has the same
basename as the input file with a '.mid' on the end instead of '.mc'
(for 'micro composer', an arbitrary file-type suffix I made up)

The limitations are:

1) It's only for equal temperaments, and octave-based ones at that.
Note that JI can be approximated arbitrarily closely using high
n-edos, however. The octave based-restriction is not a deal-breaker
for me, but may be for some (the nonoctave.org type crowd). If I had
enough interest, I'd code non-octave tunings into the things, but
right now, I'm letting it stand as is.

2) It's polyphonic, but a 'voice' cannot be anything other than a
single melodic strand per voice, i.e. one cannot do a 3-voice chord
without actually using 3 voices.

3) It's not as expressive as playing an instrument, and probably never
will be. What it gains in flexibility, it loses in corporeality.
However, I managed to do some good work with it, including "The
Juggler", "Melancholic", "Insect Ballet", "Adagio for Margo", etc.

The pros:

1) One is not constrained by a keyboard. (a plus AND minus)

2) Any pure-octave tuning is theoretically possible. Even unequal
octave based ones can be approximated using something like 1200-edo
for instance.

3) For most tunings, one can use a standard notation-like syntax. You
can even edit the script (it's Python) to define your own syntax is
you are adventurous...one also can use a pitch-class numerical based
syntax for any tuning, but especially this is helpful for things like
11-edo when note names are not helpful.

I haven't uploaded the latest version to my site yet, but email if
you're interested in checking it out.

-A.

Very interesting, but I notice that there is nothing up there yet.
Could you make an announcement again once you have some code committed to the CVS?

Magnus Jonsson

On Mon, 16 Jun 2008, Chuckk Hubbard wrote:

> Hi Mike. I'm currently programming a free, open-source sequencer for
> unlimited extended just intonation. I haven't uploaded a version yet,
> but the page is here:
> http://www.sourceforge.net/projects/rationale
> I'll put an announcement here once I get a fairly complete version up.
>
> It's probably not too useful for 72-tet in this form, but if there's
> enough interest, I could probably program a sequencer for ET tunings.
> In my experience MIDI is useless for microtuning. Far better to set
> something up with OSC, which is more customizable. Alas, the options
> for non-programmers to use OSC are limited right now.
> One solid option is to use/create frontends to Csound, like Blue,
> which Dave mentioned. Csound can load soundfonts and micro-tune them
> without the one-pitchbend-value-per-channel limitation. It is a dream
> come true.
>
> If anyone has a micro-tunable sequencer that can send OSC, or that is
> open-source, I'd be interested in helping to set it up to play
> soundfonts using Csound.
>
> -Chuckk
>
>

>>What samplers are you using? Right now I'm using East West with Native
>>Instruments' Kompakt sampler, although I do have a version of Kontakt
>>around somewhere if that would work.
>
> If you have Kontakt then you need:
> http://12equalboresme.com/Scala2Kontakt

Wow. That's amazing. I'm going to buy that immediately.

>>Ah, screw it. Someday I'll get my AXiS and live happily ever after.
>>For now, I'm happy if I can use a mouse.
>
> Actually, extended keyboards are further along than extended
> notation systems.

So I've noticed.

> Many microtonalists have at least tried using the "piano roll"
> view in their MIDI sequencer as a direct composing tool, and
> several have had good results. The other main options seems to
> be Myriad Harmony Assistant (see my prev. message) and Finale.
> Lsat I heard, Sibelius was still a pretty bad hack, but that
> might have changed.

Yeah. Right now I'm using the "Staff view" view in SONAR 7 to input
the notes. I have 6 channels - +3, +2, +1, 0 -1, -2, with each channel
representing a "degree" of 72-tet. So if I want to play 4:5:6:7:9:11,
I put the 4, 6, and 9 on the "0" channel, the 5 on the "-1" channel,
the 7 on the "-2" channel, and the 11 on the "+3" channel (transposed
of course). So it's actually pretty useful. I just long to see the day
when I can just put sims-maneri or sagittal notation into some
software directly and be gone with it. As you can imagine, having
separate tracks to handle different voices becomes the most difficult
thing in the world. And if I'm wasting 6 eastwest channels on one
sound (where 1 channel can vary from 50-400 MB), then I run out of ram
pretty quickly, even with my 3GB stash right now.

So if there's any microtonal notation software in which I can hear the
sounds that I'm putting in in realtime with some decent sample library
(VST instruments are preferred, in reality GM will suffice), that's
what I need. I looked at the Myriad Harmony Assistant, I'll have to
check that out.

Honestly, if I can get any program that can just save the results as a
MIDI file, even if it uses 5 ports worth of 16 MIDI channels to
accomplish this, I'm set - I can just independently render the results
in an Kontakt like program, and I'm good to go.

You mentioned Finale - can Finale actually play microtonal inflections
that are entered into it? I thought not, though I might be wrong.

> Oh, and there is typing in a text-based format directly, like
> microabc.
> http://br.geocities.com/hfmlacerda/abc/microabc-about.html

I'll check it out. I'm not really used to that format for composition,
but if it proves to be more useful than my current rig with SONAR, I'm
all for it.

Thanks,
Mike

> Hi Mike. I'm currently programming a free, open-source sequencer for
> unlimited extended just intonation. I haven't uploaded a version yet,
> but the page is here:
> http://www.sourceforge.net/projects/rationale
> I'll put an announcement here once I get a fairly complete version up.

Wow. That sounds great. I'll be looking closely at that. What are you
programming it in? I am fairly good with C/C++ (mainly C++, some C),
and if you're looking for some help on the project, I'd be glad to
contribute.

> It's probably not too useful for 72-tet in this form, but if there's
> enough interest, I could probably program a sequencer for ET tunings.

I think it would be incredibly useful if there were an ET component to
it. In general, the ability to import scala files would prove to be
invaluable. Temperament, whether equal or not, is musically useful in
and of itself. People do like their miracle and magic temperaments,
you know :P

Actually, if your sequencer could switch between different tunings on
the fly, in the middle of a composition, that would be EXTREMELY
useful. I don't know of any such software that can do that (unless I'm
missing something). I dream of the day when I can be in 13-limit just
and move in chord modulations that go up in repeating patterns of 1/4
or 1/5 of an octave without having to resort to using some ridiculous
high-number equal temperament. Also would be really useful for the
exploration of some of the "non-rational" harmonies involving "noble
mediants" and phi-weighted ratios that are currently all the rage on
the tuning group.

> In my experience MIDI is useless for microtuning. Far better to set
> something up with OSC, which is more customizable. Alas, the options
> for non-programmers to use OSC are limited right now.
> One solid option is to use/create frontends to Csound, like Blue,
> which Dave mentioned. Csound can load soundfonts and micro-tune them
> without the one-pitchbend-value-per-channel limitation. It is a dream
> come true.

I'm starting to realize that about MIDI. The "mu-units" or something I
remember reading about on the tonalsoft encyclopedia seemed to be a
clever way to get around those limitations within a certain margin of
error. I don't know much about OSC and CSound, only rough notions of
what they are. You say that CSound can load soundfonts... Is there any
work being in trying to connect CSound to traditional VST or DXi based
sample libraries or synths? I'm talking about Kontakt, Gigasampler,
Pianoteq, and the like. If so, problem solved.

> If anyone has a micro-tunable sequencer that can send OSC, or that is
> open-source, I'd be interested in helping to set it up to play
> soundfonts using Csound.

I'll talk to one of my professors at school about it - he composes
everything in CSound, and he might have some sort of useful
connection. The software that AKJ just talked about would probably be
useful as well - his is only equal temperament, and yours is only JI,
so there could be potential.

Plus, there are quite a few microtonal notation programs being talked
about, and if any of them are open source, then it might be easy to
route those into your sequencer.

Again, these are all just some ideas, take them for what they're worth.

-Mike

> Hi Mike,
>
> I did a good chunk of my work in custom software I now call
> 'micro_composer'. It used to be called 'et_compose', which in a way is
> a more accurate description, b/c I designed it for n-edos. However,
> b/c one can simulate JI with large edos, I liked the general term.
>
> One enters an ascii pitch and rhythmic notation, and some other
> parameters are possible, like dynamics and tempi, and you can also
> humanize the timing and dynamics using gaussian noise functions. The
> program (a Python script) parses the input file, changing it to an
> intermediate text representation of a MIDI file, and uses 't2mf' togo
> the next step of giving you the MIDI file output, which has the same
> basename as the input file with a '.mid' on the end instead of '.mc'
> (for 'micro composer', an arbitrary file-type suffix I made up)

Wow. That sounds great.

> The limitations are:
>
> 1) It's only for equal temperaments, and octave-based ones at that.
> Note that JI can be approximated arbitrarily closely using high
> n-edos, however. The octave based-restriction is not a deal-breaker
> for me, but may be for some (the nonoctave.org type crowd). If I had
> enough interest, I'd code non-octave tunings into the things, but
> right now, I'm letting it stand as is.

To be honest, I've always been pretty biased in favor of equal
temperaments anyway. In strict composition, anything goes, but when it
comes to recreating potential performances later, I always find it
much easier mentally to think in terms of things like 72-tet or 31-tet
or 34-tet or 19-tet. So no problems here.

> 2) It's polyphonic, but a 'voice' cannot be anything other than a
> single melodic strand per voice, i.e. one cannot do a 3-voice chord
> without actually using 3 voices.

Fine.

> 3) It's not as expressive as playing an instrument, and probably never
> will be. What it gains in flexibility, it loses in corporeality.
> However, I managed to do some good work with it, including "The
> Juggler", "Melancholic", "Insect Ballet", "Adagio for Margo", etc.

I heard both your Insect Ballet and Adagio for Margo recordings just
recently... How did you get such realistic sounds of it? Did you just
save the input as a midi file and run it through a softsynth/sampler?
Or was the Insect Ballet one live? I see it's credited to Jacob Barton
on your site.

> The pros:
>
> 1) One is not constrained by a keyboard. (a plus AND minus)
>
> 2) Any pure-octave tuning is theoretically possible. Even unequal
> octave based ones can be approximated using something like 1200-edo
> for instance.
>
> 3) For most tunings, one can use a standard notation-like syntax. You
> can even edit the script (it's Python) to define your own syntax is
> you are adventurous...one also can use a pitch-class numerical based
> syntax for any tuning, but especially this is helpful for things like
> 11-edo when note names are not helpful.
>
> I haven't uploaded the latest version to my site yet, but email if
> you're interested in checking it out.
>
> -A.

Definitely interested. I'll be emailing you about the details shortly.

-Mike

I was just thinking,

It would be great to see if we could somehow align all of this
software that people are making into one solid, cohesive unit. If we
could take Aaron's ET-based program and mix it with Chuck's JI-based
sequencer, integrate it with a staff-based Finale-esque input
mechanism (of which I am sure a thousand open source programs exist),
give it the capability to to read Scala files (which might be useful
mainly as a way to easily to define note names, as Chuck's program
already has an infinite set of pitches available), and then be able to
output the whole thing to a MIDI file as is already implemented in
Aaron's program, we'd have the perfect microtonal sequencer. Really
all that would be needed to make it the "perfect" microtonal
all-in-one package would be the ability to record MIDI or whatever
format from an external keyboard.

It seems like all of the individual parts are already worked out, so
all that remains to be done would be have them all communicate with
one another and create a GUI interface to round the whole thing out.
The most difficult aspect of this I think would be the staff-based
component, although depending on what's out there already, it might
not be that hard.

Just some thoughts, and if anyone else is willing to contribute to
such an objective, I think it's very feasible. Certainly it would
facilitate the creation of a lot of microtonal music from a lot of
people who might not have the technological know-how to work in Csound
and such.

-Mike

On Tue, Jun 17, 2008 at 5:11 AM, Mike Battaglia <battaglia01@...> wrote:
>> Hi Mike,
>>
>> I did a good chunk of my work in custom software I now call
>> 'micro_composer'. It used to be called 'et_compose', which in a way is
>> a more accurate description, b/c I designed it for n-edos. However,
>> b/c one can simulate JI with large edos, I liked the general term.
>>
>> One enters an ascii pitch and rhythmic notation, and some other
>> parameters are possible, like dynamics and tempi, and you can also
>> humanize the timing and dynamics using gaussian noise functions. The
>> program (a Python script) parses the input file, changing it to an
>> intermediate text representation of a MIDI file, and uses 't2mf' togo
>> the next step of giving you the MIDI file output, which has the same
>> basename as the input file with a '.mid' on the end instead of '.mc'
>> (for 'micro composer', an arbitrary file-type suffix I made up)
>
> Wow. That sounds great.
>
>> The limitations are:
>>
>> 1) It's only for equal temperaments, and octave-based ones at that.
>> Note that JI can be approximated arbitrarily closely using high
>> n-edos, however. The octave based-restriction is not a deal-breaker
>> for me, but may be for some (the nonoctave.org type crowd). If I had
>> enough interest, I'd code non-octave tunings into the things, but
>> right now, I'm letting it stand as is.
>
> To be honest, I've always been pretty biased in favor of equal
> temperaments anyway. In strict composition, anything goes, but when it
> comes to recreating potential performances later, I always find it
> much easier mentally to think in terms of things like 72-tet or 31-tet
> or 34-tet or 19-tet. So no problems here.
>
>> 2) It's polyphonic, but a 'voice' cannot be anything other than a
>> single melodic strand per voice, i.e. one cannot do a 3-voice chord
>> without actually using 3 voices.
>
> Fine.
>
>> 3) It's not as expressive as playing an instrument, and probably never
>> will be. What it gains in flexibility, it loses in corporeality.
>> However, I managed to do some good work with it, including "The
>> Juggler", "Melancholic", "Insect Ballet", "Adagio for Margo", etc.
>
> I heard both your Insect Ballet and Adagio for Margo recordings just
> recently... How did you get such realistic sounds of it? Did you just
> save the input as a midi file and run it through a softsynth/sampler?
> Or was the Insect Ballet one live? I see it's credited to Jacob Barton
> on your site.
>
>> The pros:
>>
>> 1) One is not constrained by a keyboard. (a plus AND minus)
>>
>> 2) Any pure-octave tuning is theoretically possible. Even unequal
>> octave based ones can be approximated using something like 1200-edo
>> for instance.
>>
>> 3) For most tunings, one can use a standard notation-like syntax. You
>> can even edit the script (it's Python) to define your own syntax is
>> you are adventurous...one also can use a pitch-class numerical based
>> syntax for any tuning, but especially this is helpful for things like
>> 11-edo when note names are not helpful.
>>
>> I haven't uploaded the latest version to my site yet, but email if
>> you're interested in checking it out.
>>
>> -A.
>
> Definitely interested. I'll be emailing you about the details shortly.
>
> -Mike
>

Mike Battaglia wrote:
>> Hi Mike. I'm currently programming a free, open-source sequencer for
>> unlimited extended just intonation. I haven't uploaded a version yet,
>> but the page is here:
>> http://www.sourceforge.net/projects/rationale
>> I'll put an announcement here once I get a fairly complete version up.
> > Wow. That sounds great. I'll be looking closely at that. What are you
> programming it in? I am fairly good with C/C++ (mainly C++, some C),
> and if you're looking for some help on the project, I'd be glad to
> contribute.

It's in Python, isn't it? I know C, C++, and Python but I have to be careful about committing myself to projects because there are so many other things to do...

>> It's probably not too useful for 72-tet in this form, but if there's
>> enough interest, I could probably program a sequencer for ET tunings.
> > I think it would be incredibly useful if there were an ET component to
> it. In general, the ability to import scala files would prove to be
> invaluable. Temperament, whether equal or not, is musically useful in
> and of itself. People do like their miracle and magic temperaments,
> you know :P

If it works with just intonation, it surely can't be difficult to re-map to a temperament. The only problem would be that you have to look at it from a harmonic perspective.

> Actually, if your sequencer could switch between different tunings on
> the fly, in the middle of a composition, that would be EXTREMELY
> useful. I don't know of any such software that can do that (unless I'm
> missing something). I dream of the day when I can be in 13-limit just
> and move in chord modulations that go up in repeating patterns of 1/4
> or 1/5 of an octave without having to resort to using some ridiculous
> high-number equal temperament. Also would be really useful for the
> exploration of some of the "non-rational" harmonies involving "noble
> mediants" and phi-weighted ratios that are currently all the rage on
> the tuning group.

My MIDI Relay could switch between tunings by loading Scala files, but it's obsolete now. Can't Scala do it? Fractal Tune Smithy also had support for cross-fading between tunings. I've had that working in Kyma and Csound as well. You can set a knob to control the tuning provided you have a knob spare. It depends on knowing what tuning space you want to explore in advance. I've also had Kyma changing tunings with a pedal. That should work better with Csound because I know how to stop already sounding notes from changing -- all I need is a pedal!

I now have a waveguide clarinet running in Csound with magic temperament. (As it happens the current tuning is with a noble mediant on the approximate 11:8.) I don't think it's that much like a clarinet but it sounds good. If you're interested I could easily adapt it for other tunings with your MIDI control of choice. It'll almost run out of the box once you install Csound. It took me a long time to get the tuning correct (this is a problem with waveguides) and it still isn't JI-perfect.

>> In my experience MIDI is useless for microtuning. Far better to set
>> something up with OSC, which is more customizable. Alas, the options
>> for non-programmers to use OSC are limited right now.
>> One solid option is to use/create frontends to Csound, like Blue,
>> which Dave mentioned. Csound can load soundfonts and micro-tune them
>> without the one-pitchbend-value-per-channel limitation. It is a dream
>> come true.
> > I'm starting to realize that about MIDI. The "mu-units" or something I
> remember reading about on the tonalsoft encyclopedia seemed to be a
> clever way to get around those limitations within a certain margin of
> error. I don't know much about OSC and CSound, only rough notions of
> what they are. You say that CSound can load soundfonts... Is there any
> work being in trying to connect CSound to traditional VST or DXi based
> sample libraries or synths? I'm talking about Kontakt, Gigasampler,
> Pianoteq, and the like. If so, problem solved.

Last I heard the SoundFont support wasn't complete. You could use samples from SoundFonts but the envelopes were ignored. Or you could use FluidSynth. The problem there is that it's an existing synth engine like you suggest. But the interface to it is all geared to MIDI -- and therefore 12 notes to the octave. The only way to retune it is by sending pitch bends, like you would for MIDI.

In general, connecting Csound to an existing synthesizer by VST won't magically improve the tuning capabilities.

If you're going to use MIDI I found it's best to use the piano roll. You should be able to set it up so that the pitch comes from the keyboard and you don't have to remember what note's on which position. That all depends on having a good keyboard mapping. There are plenty of microtonal soft-synths -- isn't Kontakt one?

>> If anyone has a micro-tunable sequencer that can send OSC, or that is
>> open-source, I'd be interested in helping to set it up to play
>> soundfonts using Csound.
> > I'll talk to one of my professors at school about it - he composes
> everything in CSound, and he might have some sort of useful
> connection. The software that AKJ just talked about would probably be
> useful as well - his is only equal temperament, and yours is only JI,
> so there could be potential.
> > Plus, there are quite a few microtonal notation programs being talked
> about, and if any of them are open source, then it might be easy to
> route those into your sequencer.

Maybe Lime. I never worked it out:

http://www.cerlsoundgroup.org/cgi-bin/Lime/Windows.html

The best spoken of open source notation/sequencer program out there is Rosegarden. It's Linux only which is one reason I didn't look at it before. There's a team in Scotland that were hot-wiring it for microtonal use. Unfortunately, they said the core developers weren't interested in microtonal support.

http://www.rosegardenmusic.com/

Graham

--- In MakeMicroMusic@yahoogroups.com, Graham Breed <gbreed@...> wrote:
>
> Mike Battaglia wrote:
> >> Hi Mike. I'm currently programming a free, open-source sequencer for
> >> unlimited extended just intonation. I haven't uploaded a version yet,
> >> but the page is here:
> >> http://www.sourceforge.net/projects/rationale
> >> I'll put an announcement here once I get a fairly complete
version up.
> >
> > Wow. That sounds great. I'll be looking closely at that. What are you
> > programming it in? I am fairly good with C/C++ (mainly C++, some C),
> > and if you're looking for some help on the project, I'd be glad to
> > contribute.
>
> It's in Python, isn't it? I know C, C++, and Python but I
> have to be careful about committing myself to projects
> because there are so many other things to do...

Yup, Python it is. I've thought about perhaps a C++ version, but I
don't know the language well enough, and this program is my main
reason for wanting to program at all...

>
> >> It's probably not too useful for 72-tet in this form, but if there's
> >> enough interest, I could probably program a sequencer for ET tunings.
> >
> > I think it would be incredibly useful if there were an ET component to
> > it. In general, the ability to import scala files would prove to be
> > invaluable. Temperament, whether equal or not, is musically useful in
> > and of itself. People do like their miracle and magic temperaments,
> > you know :P
>
> If it works with just intonation, it surely can't be
> difficult to re-map to a temperament. The only problem
> would be that you have to look at it from a harmonic
> perspective.

Well, one feature is that one may recenter the entire score (or just
parts of it) by changing, globally or locally, where 1/1 is. Thus no
"scale" is necessary; anything within the 13-limit is possible by
changing 1/1 enough times. No reason this couldn't work with tempered
tunings, but for most people it isn't necessary.

> > what they are. You say that CSound can load soundfonts... Is there any
> > work being in trying to connect CSound to traditional VST or DXi based
> > sample libraries or synths? I'm talking about Kontakt, Gigasampler,
> > Pianoteq, and the like. If so, problem solved.
>
> Last I heard the SoundFont support wasn't complete. You
> could use samples from SoundFonts but the envelopes were
> ignored. Or you could use FluidSynth. The problem there is
> that it's an existing synth engine like you suggest. But
> the interface to it is all geared to MIDI -- and therefore
> 12 notes to the octave. The only way to retune it is by
> sending pitch bends, like you would for MIDI.

The advantage, as I understand it, is that one can have any number of
FluidSynth engines, so any number of microtunings simultaneously. I
asked about it a while back on the Csound list, maybe it's time to do
some more research.

> >> If anyone has a micro-tunable sequencer that can send OSC, or that is
> >> open-source, I'd be interested in helping to set it up to play
> >> soundfonts using Csound.
> >
> > I'll talk to one of my professors at school about it - he composes
> > everything in CSound, and he might have some sort of useful
> > connection. The software that AKJ just talked about would probably be
> > useful as well - his is only equal temperament, and yours is only JI,
> > so there could be potential.
> >
> > Plus, there are quite a few microtonal notation programs being talked
> > about, and if any of them are open source, then it might be easy to
> > route those into your sequencer.
>
> Maybe Lime. I never worked it out:
>
> http://www.cerlsoundgroup.org/cgi-bin/Lime/Windows.html
>
> The best spoken of open source notation/sequencer program
> out there is Rosegarden. It's Linux only which is one
> reason I didn't look at it before. There's a team in
> Scotland that were hot-wiring it for microtonal use.
> Unfortunately, they said the core developers weren't
> interested in microtonal support.

I'm a devoted Linux user myself, but I see that as Rosegarden's main
limitation; if anyone cares about influencing musicians or broadening
the microtonal community, cross-platform software is a must.

-Chuckk

--- In MakeMicroMusic@yahoogroups.com, "Mike Battaglia"
<battaglia01@...> wrote:
>
> > Hi Mike. I'm currently programming a free, open-source sequencer for
> > unlimited extended just intonation. I haven't uploaded a version yet,
> > but the page is here:
> > http://www.sourceforge.net/projects/rationale
> > I'll put an announcement here once I get a fairly complete version up.
>
> Wow. That sounds great. I'll be looking closely at that. What are you
> programming it in? I am fairly good with C/C++ (mainly C++, some C),
> and if you're looking for some help on the project, I'd be glad to
> contribute.

Well I've been working in Python...
I'd considered C++, which I don't know so well, as it has extensions
for the same GUI library (Tk), but I've been working on it for a while
in Python, so it'd be a bear to switch.

>
> > It's probably not too useful for 72-tet in this form, but if there's
> > enough interest, I could probably program a sequencer for ET tunings.
>
> I think it would be incredibly useful if there were an ET component to
> it. In general, the ability to import scala files would prove to be

It might work to set it up to import scala .seq files. I got the idea
for the moving tonal center from that format.

> invaluable. Temperament, whether equal or not, is musically useful in
> and of itself. People do like their miracle and magic temperaments,
> you know :P
>
> Actually, if your sequencer could switch between different tunings on
> the fly, in the middle of a composition, that would be EXTREMELY
> useful. I don't know of any such software that can do that (unless I'm
> missing something). I dream of the day when I can be in 13-limit just
> and move in chord modulations that go up in repeating patterns of 1/4
> or 1/5 of an octave without having to resort to using some ridiculous
> high-number equal temperament. Also would be really useful for the
> exploration of some of the "non-rational" harmonies involving "noble
> mediants" and phi-weighted ratios that are currently all the rage on
> the tuning group.

Hmm... that's definitely a little farther down the road right now.
But a good idea.

>
> > In my experience MIDI is useless for microtuning. Far better to set
> > something up with OSC, which is more customizable. Alas, the options
> > for non-programmers to use OSC are limited right now.
> > One solid option is to use/create frontends to Csound, like Blue,
> > which Dave mentioned. Csound can load soundfonts and micro-tune them
> > without the one-pitchbend-value-per-channel limitation. It is a dream
> > come true.
>
> I'm starting to realize that about MIDI. The "mu-units" or something I
> remember reading about on the tonalsoft encyclopedia seemed to be a
> clever way to get around those limitations within a certain margin of
> error. I don't know much about OSC and CSound, only rough notions of
> what they are. You say that CSound can load soundfonts... Is there any
> work being in trying to connect CSound to traditional VST or DXi based
> sample libraries or synths? I'm talking about Kontakt, Gigasampler,
> Pianoteq, and the like. If so, problem solved.

There is a VST component for Csound, both for using Csound as a VST
plugin, and for loading VST plugins. Me, I do my synthesis work in
Csound; it's a little more involved, but more flexible than most.

> > If anyone has a micro-tunable sequencer that can send OSC, or that is
> > open-source, I'd be interested in helping to set it up to play
> > soundfonts using Csound.
>
> I'll talk to one of my professors at school about it - he composes
> everything in CSound, and he might have some sort of useful
> connection. The software that AKJ just talked about would probably be
> useful as well - his is only equal temperament, and yours is only JI,
> so there could be potential.
>
> Plus, there are quite a few microtonal notation programs being talked
> about, and if any of them are open source, then it might be easy to
> route those into your sequencer.

As soon as I get a day off, I'll get a prototype posted on the
Sourceforge page. It runs, it changes tonality, and it plays through
Csound, but I've yet to finish two major features: changing which
instrument one is adding (all instruments are on one big page, rather
than separate staves, which could cause a problem with overlapping
notes), and controlling which sounds each instrument makes. I haven't
been able to justify putting it up without those features in place.

> Again, these are all just some ideas, take them for what they're worth.

Good stuff, I'm glad I'm not the only one around interested in these
things. I have been away from the community for a while, since I
realized the software I wanted didn't exist, and since I went
fanatical on the JI thing.

-Chuckk

Absolutely.

--- In MakeMicroMusic@yahoogroups.com, Magnus Jonsson <magnus@...> wrote:
>
> Very interesting, but I notice that there is nothing up there yet.
> Could you make an announcement again once you have some code
committed to
> the CVS?
>
> Magnus Jonsson
>
> On Mon, 16 Jun 2008, Chuckk Hubbard wrote:
>
> > Hi Mike. I'm currently programming a free, open-source sequencer for
> > unlimited extended just intonation. I haven't uploaded a version yet,
> > but the page is here:
> > http://www.sourceforge.net/projects/rationale
> > I'll put an announcement here once I get a fairly complete version up.
> >
> > It's probably not too useful for 72-tet in this form, but if there's
> > enough interest, I could probably program a sequencer for ET tunings.
> > In my experience MIDI is useless for microtuning. Far better to set
> > something up with OSC, which is more customizable. Alas, the options
> > for non-programmers to use OSC are limited right now.
> > One solid option is to use/create frontends to Csound, like Blue,
> > which Dave mentioned. Csound can load soundfonts and micro-tune them
> > without the one-pitchbend-value-per-channel limitation. It is a dream
> > come true.
> >
> > If anyone has a micro-tunable sequencer that can send OSC, or that is
> > open-source, I'd be interested in helping to set it up to play
> > soundfonts using Csound.
> >
> > -Chuckk
> >
> >
>

--- In MakeMicroMusic@yahoogroups.com, "Mike Battaglia"
<battaglia01@...> wrote:
>
> I was just thinking,
>
> It would be great to see if we could somehow align all of this
> software that people are making into one solid, cohesive unit. If we
> could take Aaron's ET-based program and mix it with Chuck's JI-based
> sequencer, integrate it with a staff-based Finale-esque input
> mechanism (of which I am sure a thousand open source programs
> exist), give it the capability to to read Scala files (which might
> be useful mainly as a way to easily to define note names, as
> Chuck's program already has an infinite set of pitches available),
> and then be able to output the whole thing to a MIDI file as is
> already implemented in Aaron's program, we'd have the perfect
> microtonal sequencer. Really all that would be needed to make it
> the "perfect" microtonal all-in-one package would be the ability to
> record MIDI or whatever format from an external keyboard.
>
> It seems like all of the individual parts are already worked out, so
> all that remains to be done would be have them all communicate with
> one another and create a GUI interface to round the whole thing out.
> The most difficult aspect of this I think would be the staff-based
> component, although depending on what's out there already, it might
> not be that hard.
>
> Just some thoughts, and if anyone else is willing to contribute to
> such an objective, I think it's very feasible. Certainly it would
> facilitate the creation of a lot of microtonal music from a lot of
> people who might not have the technological know-how to work in
> Csound and such.
>

Well, I suspect an all-in-one device suitable for every purpose is
not realistic... But actually, I have done some thoughts about
microtonal sequencers, too. Here are them:

First: there is only a problem if we need more than 12 pitches per
octave. In other cases, standard sequencers/notation programs and the
standard midi file format work perfectly, together with sound devices
that support retuning and/or scala which can retune standard midi
files by inserting pitch-bends.

Second: I would let the sequencer take and produce standard midi,
keeping tuning details out of the sequencer as much as possible.
After that, the same as above for devices that support retuning
and/or scala support.

What is to be done, in my eyes, would be:

Take a usual standard sequencer, with a piano-roll e.g, and add a
layer that converts between the paino roll and the actual midi
events - concrete: notes of one "virtual" microtonal channel would
have to be distributed to two or more standard midi channels,
according to a definable pattern. In the case of 72tet, this would
be, e.g.:

"virtual" note 1 to standard midi note 1 of channel 1
"virtual" note 2 to standard midi note 1 of channel 2
"virtual" note 3 to standard midi note 1 of channel 3
"virtual" note 4 to standard midi note 1 of channel 4
"virtual" note 5 to standard midi note 1 of channel 5
"virtual" note 6 to standard midi note 1 of channel 6
"virtual" note 7 to standard midi note 2 of channel 1
"virtual" note 8 to standard midi note 2 of channel 2

etc.

For real-time playing, the converting layer would go into both
directions (combining notes of several standard midi channels into
one "virtual" microtonal channel).

Usual sequencers often have staff-notation view and piano-roll views
as alternatives - for our microtonal sequencer, this would be, I
suppose, a staff-notation view with sagittal notation.
For this, some details about the tuning in the sequencer could not be
avoided (definition which note of the staff notation view maps to
which row in the piano roll view).

I have been thinking of trying to do something like that - but I
haven't fountd the time til now :-( In case a project like this is
started, I might be interested to contribute, in C++ preferably, for
example with the wxWidgets-framework, which would be cross-platform.
Or as an extension to Rosegarden...
--
Hans Straub

Hiya Mike,

>> Many microtonalists have at least tried using the "piano roll"
>> view in their MIDI sequencer as a direct composing tool, and
>> several have had good results. The other main options seems to
>> be Myriad Harmony Assistant (see my prev. message) and Finale.
>> Lsat I heard, Sibelius was still a pretty bad hack, but that
>> might have changed.
>
>Yeah. Right now I'm using the "Staff view" view in SONAR 7 to input
>the notes. I have 6 channels - +3, +2, +1, 0 -1, -2, with each channel
>representing a "degree" of 72-tet. So if I want to play 4:5:6:7:9:11,
>I put the 4, 6, and 9 on the "0" channel, the 5 on the "-1" channel,
>the 7 on the "-2" channel, and the 11 on the "+3" channel (transposed
>of course). So it's actually pretty useful. I just long to see the day
>when I can just put sims-maneri or sagittal notation into some
>software directly and be gone with it.

Supposedly, Finale *can* do this. You can use any ttf file for
accidentals or something, and then you assign MIDI note number
movements to each one. Or something.

>So if there's any microtonal notation software in which I can hear the
>sounds that I'm putting in in realtime with some decent sample library
>(VST instruments are preferred, in reality GM will suffice), that's
>what I need. I looked at the Myriad Harmony Assistant, I'll have to
>check that out.

Any package with custom accidentals or custom lines/spaces or
something like that will work, since they all send MIDI and there
are many excellent microtonal synths available. The key is the
first part.

>You mentioned Finale - can Finale actually play microtonal inflections
>that are entered into it? I thought not, though I might be wrong.

That was the claim. Rick McGowan, you got your ears on?

-Carl

Hi Hans,

>Well, I suspect an all-in-one device suitable for every purpose is
>not realistic... But actually, I have done some thoughts about
>microtonal sequencers, too. Here are them:
>
>First: there is only a problem if we need more than 12 pitches per
>octave. In other cases, standard sequencers/notation programs and the
>standard midi file format work perfectly, together with sound devices
>that support retuning and/or scala which can retune standard midi
>files by inserting pitch-bends.
>
>Second: I would let the sequencer take and produce standard midi,
>keeping tuning details out of the sequencer as much as possible.
>After that, the same as above for devices that support retuning
>and/or scala support.

Agree, agree, and agree.

>What is to be done, in my eyes, would be:
>
>Take a usual standard sequencer, with a piano-roll e.g, and add a
>layer that converts between the paino roll and the actual midi
>events - concrete: notes of one "virtual" microtonal channel would
>have to be distributed to two or more standard midi channels,
>according to a definable pattern. In the case of 72tet, this would
>be, e.g.:
>
>"virtual" note 1 to standard midi note 1 of channel 1
>"virtual" note 2 to standard midi note 1 of channel 2
>"virtual" note 3 to standard midi note 1 of channel 3
>"virtual" note 4 to standard midi note 1 of channel 4
>"virtual" note 5 to standard midi note 1 of channel 5
>"virtual" note 6 to standard midi note 1 of channel 6
>"virtual" note 7 to standard midi note 2 of channel 1
>"virtual" note 8 to standard midi note 2 of channel 2
>
>etc.

Meh. I'd rather have a piano roll that alpha blends to staff
notation (an idea I've been toying with for some years).

>For real-time playing, the converting layer would go into both
>directions (combining notes of several standard midi channels into
>one "virtual" microtonal channel).

Why does the composer care about MIDI channels? This can
be taken care of in the 'tuning setup' phase before playing/
composition begins.

>Usual sequencers often have staff-notation view and piano-roll views
>as alternatives - for our microtonal sequencer, this would be, I
>suppose, a staff-notation view with sagittal notation.

Please don't lock it into one notation concept. Rather, let
the user configure the meaning (in terms of MIDI offsets) of
line and space, and as well a map into any OpenType font file
(for accidentals).

-Carl

The developer of Numerology, a performance oriented sequencer app for
Mac OSX, has been soliciting input from microtonal users for a while
now. If he got enough feedback, he'd implement tuning features,
taking advantage of Core Audio's API which allows frequencies to be
specified, skipping MIDI altogether, if memory serves. Obviously, a
disadvantage of this approach is you have to get a Mac if you don't
use one already, and you can only use softsynths.

http://five12.com/

--- In MakeMicroMusic@yahoogroups.com, "paolovalladolid" <phv40@...>
wrote:
>
> The developer of Numerology, a performance oriented sequencer app for
> Mac OSX, has been soliciting input from microtonal users for a while
> now. If he got enough feedback, he'd implement tuning features,
> taking advantage of Core Audio's API which allows frequencies to be
> specified, skipping MIDI altogether, if memory serves. Obviously, a
> disadvantage of this approach is you have to get a Mac if you don't
> use one already, and you can only use softsynths.
>
> http://five12.com/
>

This sounds cool, but as much as I applaud Mac-land (I will eventually
buy one, b/c I can still run Linux apps on it), a cross-platform
solution would be the ideal. Something done, as mentioned, with
cross-platform GUI toolkits.

But, since Stephen Yi has already done 'Blue', a CSound-frontend-GUI
which has a microtonal piano roll, Carl's wet dream already exists. If
you're ok with CSound as a platform, and why wouldn't you be? Anything
that can be done in any softsynth can be done in CSound, including
creating custom sampling engines.

--- In MakeMicroMusic@yahoogroups.com, "Aaron Krister
Johnson" <aaron@...> wrote:
>
> --- In MakeMicroMusic@yahoogroups.com, "paolovalladolid" <phv40@>
> wrote:
> >
> > The developer of Numerology, a performance oriented sequencer app
for
> > Mac OSX, has been soliciting input from microtonal users for a
while
> > now. If he got enough feedback, he'd implement tuning features,
> > taking advantage of Core Audio's API which allows frequencies to
be
> > specified, skipping MIDI altogether, if memory serves.
Obviously, a
> > disadvantage of this approach is you have to get a Mac if you
don't
> > use one already, and you can only use softsynths.
> >
> > http://five12.com/
> >
>
> This sounds cool, but as much as I applaud Mac-land (I will
eventually
> buy one, b/c I can still run Linux apps on it), a cross-platform
> solution would be the ideal. Something done, as mentioned, with
> cross-platform GUI toolkits.

Understood. I threw Numerology out there mainly for the Mac-using
contingent and wouldn't expect any non-Mac user to convert.

Sigh... I've been working on and learning portaudio as a playform
independent tool which is a much lower level thing. Should I be
learning Csound instead? I like to code in C++ but everything has C++
bindings these days so I assume Csound does also. The last time I
looked at Csound seriously it was pretty low level also but that was
20 or so years ago. I assume it is a lot richer package now.

Is Csound the right path for platform independence?

Oh well. This old dog probably has one more trick in him.

On Tue, Jun 17, 2008 at 4:59 PM, Aaron Krister Johnson
<aaron@...> wrote:
> --- In MakeMicroMusic@yahoogroups.com, "paolovalladolid" <phv40@...>
> wrote:
>
>>
>> The developer of Numerology, a performance oriented sequencer app for
>> Mac OSX, has been soliciting input from microtonal users for a while
>> now. If he got enough feedback, he'd implement tuning features,
>> taking advantage of Core Audio's API which allows frequencies to be
>> specified, skipping MIDI altogether, if memory serves. Obviously, a
>> disadvantage of this approach is you have to get a Mac if you don't
>> use one already, and you can only use softsynths.
>>
>> http://five12.com/
>>
>
> This sounds cool, but as much as I applaud Mac-land (I will eventually
> buy one, b/c I can still run Linux apps on it), a cross-platform
> solution would be the ideal. Something done, as mentioned, with
> cross-platform GUI toolkits.
>
> But, since Stephen Yi has already done 'Blue', a CSound-frontend-GUI
> which has a microtonal piano roll, Carl's wet dream already exists. If
> you're ok with CSound as a platform, and why wouldn't you be? Anything
> that can be done in any softsynth can be done in CSound, including
> creating custom sampling engines.
>
>

--
Steve Morris
barbershopsteve@...
Bass: Unnamed quintet/quartet whatever
Bass: Sounds Of Concord

--- In MakeMicroMusic@yahoogroups.com, "Steve
Morris" <barbershopsteve@...> wrote:
>
> Sigh... I've been working on and learning portaudio as a playform
> independent tool which is a much lower level thing. Should I be
> learning Csound instead? I like to code in C++ but everything has
C++
> bindings these days so I assume Csound does also. The last time I
> looked at Csound seriously it was pretty low level also but that was
> 20 or so years ago. I assume it is a lot richer package now.
>
> Is Csound the right path for platform independence?

Tough call. Check out ChuCK and Supercollider too. On a slightly
higher level (while still keeping the cost free), there's Pure Data.

I've heard of Processing too, but I understand that is specific to
Java.

Graham Breed wrote:

> I now have a waveguide clarinet running in Csound with magic > temperament. (As it happens the current tuning is with a > noble mediant on the approximate 11:8.) I don't think it's > that much like a clarinet but it sounds good. If you're > interested I could easily adapt it for other tunings with > your MIDI control of choice. It'll almost run out of the > box once you install Csound. It took me a long time to get > the tuning correct (this is a problem with waveguides) and > it still isn't JI-perfect.

I ought to find the time one of these days to do more than just simple experiments with Csound. You can do a reasonably readable notation system with macros -- I've mentioned on the tuning list some of the equivalent letters I used in place of the ASCII version of Sagittal (Z substituting for / , T for | , S for ~ , etc.) So you might write [$B(4)$UZ.] in a Csound score for Sagittal B!!/ in octave 4...

I gave up on the waveguide instruments, not realizing there was a way to get more tuning accuracy out of them. But you can still do quite a bit with buzz or oscil, a few filters, envelopes, some delay, and other effects.

>> The developer of Numerology, a performance oriented sequencer app for
>> Mac OSX, has been soliciting input from microtonal users for a while
>> now. If he got enough feedback, he'd implement tuning features,
>> taking advantage of Core Audio's API which allows frequencies to be
>> specified, skipping MIDI altogether, if memory serves. Obviously, a
>> disadvantage of this approach is you have to get a Mac if you don't
>> use one already, and you can only use softsynths.
>>
>> http://five12.com/
>
>This sounds cool, but as much as I applaud Mac-land (I will eventually
>buy one, b/c I can still run Linux apps on it), a cross-platform
>solution would be the ideal. Something done, as mentioned, with
>cross-platform GUI toolkits.
>
>But, since Stephen Yi has already done 'Blue', a CSound-frontend-GUI
>which has a microtonal piano roll, Carl's wet dream already exists. If
>you're ok with CSound as a platform, and why wouldn't you be? Anything
>that can be done in any softsynth can be done in CSound, including
>creating custom sampling engines.

I said my wet dream was a microtonal score editor, not a microtonal
piano roll. Any sequencer with a piano roll can be microtonal with
a microtonal synth. Take my favorite synth, pianoteq, for example.
How long would it take you to create a physically-modeled piano of
this quality in CSound. Forever and a year, and when you were done
you'd need an 8-core machine to run it in realtime.

-Carl

On Wed, Jun 18, 2008 at 12:26 AM, Carl Lumma <carl@...> wrote:
>>> The developer of Numerology, a performance oriented sequencer app for
>>> Mac OSX, has been soliciting input from microtonal users for a while
>>> now. If he got enough feedback, he'd implement tuning features,
>>> taking advantage of Core Audio's API which allows frequencies to be
>>> specified, skipping MIDI altogether, if memory serves. Obviously, a
>>> disadvantage of this approach is you have to get a Mac if you don't
>>> use one already, and you can only use softsynths.
>>>
>>> http://five12.com/
>>
>>This sounds cool, but as much as I applaud Mac-land (I will eventually
>>buy one, b/c I can still run Linux apps on it), a cross-platform
>>solution would be the ideal. Something done, as mentioned, with
>>cross-platform GUI toolkits.
>>
>>But, since Stephen Yi has already done 'Blue', a CSound-frontend-GUI
>>which has a microtonal piano roll, Carl's wet dream already exists. If
>>you're ok with CSound as a platform, and why wouldn't you be? Anything
>>that can be done in any softsynth can be done in CSound, including
>>creating custom sampling engines.
>
> I said my wet dream was a microtonal score editor, not a microtonal
> piano roll. Any sequencer with a piano roll can be microtonal with
> a microtonal synth. Take my favorite synth, pianoteq, for example.
> How long would it take you to create a physically-modeled piano of
> this quality in CSound. Forever and a year, and when you were done
> you'd need an 8-core machine to run it in realtime.
>
> -Carl

Disturbingly we share the same wet dream. A microtonal score
editor/sequencer would probably be the end-all-be-all for me. But I
think the way to go with it is probably this:

1) Have the software operate in CSound or something
2) Have the software save the result in both its own proprietary
format and also "export" to a MIDI file
3) The MIDI file would be as many channels as is needed to represent
the composition in its entirety. Then you load it up into
Pianoteq/VSL/EastWest/Plinky-Dink's-Magic-Synth and roll with it,
record a few times until you get the whole thing down in audio. then
you simply release your composition to the world and take your pick
from the women that swoon over it.

So for example, if we're composing in 144-et and there is a 31-note
chord split between 6 instruments, and 98 MIDI channels are needed to
represent the entire chord, then we simply save those in sets of 16
for different ports. We then run through them 16 at a time and record
them with our favorite synths, then layer them together (this might
cause a few problems but will most likely work in the majority of
cases), and there you go.

To be honest, anyone that has a text-based interface like microabc
already has it half done. All that remains is for a clever GUI to be
written with some musical notation font and layer that on top of the
text interface. Really isn't that hard.

So here is what we need to actually accomplish this:

1) An existing cross-platform microtonal sequencer, preferably coded in C++
2) An existing cross-platform notation package, preferably coded in C++

If I can get pointed into the right direction as to the pros and cons
of the ones out there, I'll definitely start trying to integrate them
on my own. As of now, there are so many posted in this thread I can't
tell them all apart - what are the advantages of microabc over other
software packages, etc? Do people find any of them buggy, etc.? What
are people's preferences with different programs?

I could use a hobby - I've run out of steam with this string
composition that I've been writing, and I need something else to focus
on. Anyone with coding experience on here who is willing to help out
with this project, feel free - especially with the CSound aspects of
it, which I'm a little rough on.

-Mike

Chuckk Hubbard wrote:

> Yup, Python it is. I've thought about perhaps a C++ version, but I
> don't know the language well enough, and this program is my main
> reason for wanting to program at all...

When Rosegarden started they chose C++ because it was fast for complex code. Probably Python's fast enough now.

>>>> It's probably not too useful for 72-tet in this form, but if there's
>>>> enough interest, I could probably program a sequencer for ET tunings.
>>> I think it would be incredibly useful if there were an ET component to
>>> it. In general, the ability to import scala files would prove to be
>>> invaluable. Temperament, whether equal or not, is musically useful in
>>> and of itself. People do like their miracle and magic temperaments,
>>> you know :P
>> If it works with just intonation, it surely can't be >> difficult to re-map to a temperament. The only problem >> would be that you have to look at it from a harmonic >> perspective.
> > Well, one feature is that one may recenter the entire score (or just
> parts of it) by changing, globally or locally, where 1/1 is. Thus no
> "scale" is necessary; anything within the 13-limit is possible by
> changing 1/1 enough times. No reason this couldn't work with tempered
> tunings, but for most people it isn't necessary.

Depending on whether most people want just intonation or not ...

How's it notated?

>>> what they are. You say that CSound can load soundfonts... Is there any
>>> work being in trying to connect CSound to traditional VST or DXi based
>>> sample libraries or synths? I'm talking about Kontakt, Gigasampler,
>>> Pianoteq, and the like. If so, problem solved.
>> Last I heard the SoundFont support wasn't complete. You >> could use samples from SoundFonts but the envelopes were >> ignored. Or you could use FluidSynth. The problem there is >> that it's an existing synth engine like you suggest. But >> the interface to it is all geared to MIDI -- and therefore >> 12 notes to the octave. The only way to retune it is by >> sending pitch bends, like you would for MIDI.
> > The advantage, as I understand it, is that one can have any number of
> FluidSynth engines, so any number of microtunings simultaneously. I
> asked about it a while back on the Csound list, maybe it's time to do
> some more research.

Can't you do that with MIDI? Well, I suppose once you've written a microtonal instrument using a FluidSynth instance it won't be that complex to use.

> I'm a devoted Linux user myself, but I see that as Rosegarden's main
> limitation; if anyone cares about influencing musicians or broadening
> the microtonal community, cross-platform software is a must.

But I don't think there's anything else. If you can write a cross platform audio sequencer with notation support I'm all for it.

Graham

Herman Miller wrote:
> Graham Breed wrote:
> >> I now have a waveguide clarinet running in Csound with magic >> temperament. (As it happens the current tuning is with a >> noble mediant on the approximate 11:8.) I don't think it's >> that much like a clarinet but it sounds good. If you're >> interested I could easily adapt it for other tunings with >> your MIDI control of choice. It'll almost run out of the >> box once you install Csound. It took me a long time to get >> the tuning correct (this is a problem with waveguides) and >> it still isn't JI-perfect.
> > I ought to find the time one of these days to do more than just simple > experiments with Csound. You can do a reasonably readable notation > system with macros -- I've mentioned on the tuning list some of the > equivalent letters I used in place of the ASCII version of Sagittal (Z > substituting for / , T for | , S for ~ , etc.) So you might write > [$B(4)$UZ.] in a Csound score for Sagittal B!!/ in octave 4...

I use numbers for notes so that's simple in Csound. But Csound scores still aren't that readable because it means one note per line.

> I gave up on the waveguide instruments, not realizing there was a way to > get more tuning accuracy out of them. But you can still do quite a bit > with buzz or oscil, a few filters, envelopes, some delay, and other effects.

See if this works:

http://x31eq.com/tripodcs.zip

I run it as

csound -M 2 -m 38 -d -odac3 tripod-midi.orc midi.sco

but your audio and MIDI numbers will be different. It's based on an example from The Csound Book. I adapted it to work in real time, and in a generic way that can be re-tuned. Then I found I had to adjust the tuning. It's about as simple as you say -- an oscillator with an envelope for the reed sound, a filter, a delay line.

Graham

paolovalladolid wrote:
> --- In MakeMicroMusic@yahoogroups.com, "Steve > Morris" <barbershopsteve@...> wrote:
>> Sigh... I've been working on and learning portaudio as a playform
>> independent tool which is a much lower level thing. Should I be
>> learning Csound instead? I like to code in C++ but everything has > C++
>> bindings these days so I assume Csound does also. The last time I
>> looked at Csound seriously it was pretty low level also but that was
>> 20 or so years ago. I assume it is a lot richer package now.
>>
>> Is Csound the right path for platform independence?
> > Tough call. Check out ChuCK and Supercollider too. On a slightly > higher level (while still keeping the cost free), there's Pure Data.
> > I've heard of Processing too, but I understand that is specific to > Java.

Java's a cross-platform paltform. As are Csound and friends in a way. There's some kind of Java synthesizer API.

No need to code low-level synthesizers in C++, at any rate, unless you have a burning desire to do so.

Graham

Mike Battaglia wrote:

> Disturbingly we share the same wet dream. A microtonal score
> editor/sequencer would probably be the end-all-be-all for me. But I
> think the way to go with it is probably this:
> > 1) Have the software operate in CSound or something

It wouldn't operate in Csound. It might use Csound as a synth engine.

> 2) Have the software save the result in both its own proprietary
> format and also "export" to a MIDI file

Could do, but the format needn't be proprietary, and the Rosegarden team managed to do everything in MIDI.

> 3) The MIDI file would be as many channels as is needed to represent
> the composition in its entirety. Then you load it up into
> Pianoteq/VSL/EastWest/Plinky-Dink's-Magic-Synth and roll with it,
> record a few times until you get the whole thing down in audio. then
> you simply release your composition to the world and take your pick
> from the women that swoon over it.
> > So for example, if we're composing in 144-et and there is a 31-note
> chord split between 6 instruments, and 98 MIDI channels are needed to
> represent the entire chord, then we simply save those in sets of 16
> for different ports. We then run through them 16 at a time and record
> them with our favorite synths, then layer them together (this might
> cause a few problems but will most likely work in the majority of
> cases), and there you go.

There you go -- that's the complexity of using MIDI. With Csound you can write your score file and render it to audio outside real time.

For a sequencer you also need to be able to hear the other parts as you record them. Normal audio sequencers do this fine.

> To be honest, anyone that has a text-based interface like microabc
> already has it half done. All that remains is for a clever GUI to be
> written with some musical notation font and layer that on top of the
> text interface. Really isn't that hard.

I'm glad you're so confident. I count a notation GUI as a hard thing to write. And an audio sequencer for that matter.

> So here is what we need to actually accomplish this:
> > 1) An existing cross-platform microtonal sequencer, preferably coded in C++
> 2) An existing cross-platform notation package, preferably coded in C++

Why preferably C++? The rest of us seem to prefer Python.

> If I can get pointed into the right direction as to the pros and cons
> of the ones out there, I'll definitely start trying to integrate them
> on my own. As of now, there are so many posted in this thread I can't
> tell them all apart - what are the advantages of microabc over other
> software packages, etc? Do people find any of them buggy, etc.? What
> are people's preferences with different programs?

There's Blue, but it uses Csound rather than MIDI and doesn't have the notation. There's Rosegarden, but it's Linux only.

> I could use a hobby - I've run out of steam with this string
> composition that I've been writing, and I need something else to focus
> on. Anyone with coding experience on here who is willing to help out
> with this project, feel free - especially with the CSound aspects of
> it, which I'm a little rough on.

What you're suggesting isn't trivial. Why not have a look at Csound instead?

Graham

On Wed, Jun 18, 2008 at 6:36 AM, Graham Breed <gbreed@...> wrote:
>
> Mike Battaglia wrote:
>
> > Disturbingly we share the same wet dream. A microtonal score
> > editor/sequencer would probably be the end-all-be-all for me. But I
> > think the way to go with it is probably this:
> >
> > 1) Have the software operate in CSound or something
>
> It wouldn't operate in Csound. It might use Csound as a
> synth engine.

Fine. Semantics. You know what I mean.

> > 2) Have the software save the result in both its own proprietary
> > format and also "export" to a MIDI file
>
> Could do, but the format needn't be proprietary, and the
> Rosegarden team managed to do everything in MIDI.

Then that works as well.

> > 3) The MIDI file would be as many channels as is needed to represent
> > the composition in its entirety. Then you load it up into
> > Pianoteq/VSL/EastWest/Plinky-Dink's-Magic-Synth and roll with it,
> > record a few times until you get the whole thing down in audio. then
> > you simply release your composition to the world and take your pick
> > from the women that swoon over it.
> >
> > So for example, if we're composing in 144-et and there is a 31-note
> > chord split between 6 instruments, and 98 MIDI channels are needed to
> > represent the entire chord, then we simply save those in sets of 16
> > for different ports. We then run through them 16 at a time and record
> > them with our favorite synths, then layer them together (this might
> > cause a few problems but will most likely work in the majority of
> > cases), and there you go.
>
> There you go -- that's the complexity of using MIDI. With
> Csound you can write your score file and render it to audio
> outside real time.

Well, with Csound, if I wanted to use a sample library such as
EastWest or Vienna to hear what a microtonal string concerto would
sound like, could I? I thought Csound was basically an extremely
programmable synth.

> For a sequencer you also need to be able to hear the other
> parts as you record them. Normal audio sequencers do this fine.

Not sure what you mean here.

> > To be honest, anyone that has a text-based interface like microabc
> > already has it half done. All that remains is for a clever GUI to be
> > written with some musical notation font and layer that on top of the
> > text interface. Really isn't that hard.
>
> I'm glad you're so confident. I count a notation GUI as a
> hard thing to write. And an audio sequencer for that matter.

Well, the GUI would certainly be time-consuming, but it isn't
unfathomable, nor is the process needed to get from point a to point b
unfathomable. Same with the sequencer. As long as the program has a
general plan to get from start to finish, it becomes much easier.

> > So here is what we need to actually accomplish this:
> >
> > 1) An existing cross-platform microtonal sequencer, preferably coded in C++
> > 2) An existing cross-platform notation package, preferably coded in C++
>
> Why preferably C++? The rest of us seem to prefer Python.

I was listing my own preferences actually. I had the first sentence
that I was going to try to integrate them on my own, but then I moved
it to the paragraph below. I'm unfamiliar with Python, so if I was
going to do it, I was going to use C/C++. Plus, I'd be wary of writing
a complicated CPU-intensive, realtime program like an audio sequencer
or notation suite in a scripting language.

> > If I can get pointed into the right direction as to the pros and cons
> > of the ones out there, I'll definitely start trying to integrate them
> > on my own. As of now, there are so many posted in this thread I can't
> > tell them all apart - what are the advantages of microabc over other
> > software packages, etc? Do people find any of them buggy, etc.? What
> > are people's preferences with different programs?
>
> There's Blue, but it uses Csound rather than MIDI and
> doesn't have the notation. There's Rosegarden, but it's
> Linux only.

I'll check Blue out.

> > I could use a hobby - I've run out of steam with this string
> > composition that I've been writing, and I need something else to focus
> > on. Anyone with coding experience on here who is willing to help out
> > with this project, feel free - especially with the CSound aspects of
> > it, which I'm a little rough on.
>
> What you're suggesting isn't trivial. Why not have a look
> at Csound instead?

If the goal is to be able to easily compose microtonal music using a
notation software and be able to play it through a VST sampler such as
pianoteq or kontakt, how can that be accomplished using Csound
instead?

-Mike

>> To be honest, anyone that has a text-based interface like microabc
>> already has it half done. All that remains is for a clever GUI to be
>> written with some musical notation font and layer that on top of the
>> text interface. Really isn't that hard.
>
> I'm glad you're so confident. I count a notation GUI as a
> hard thing to write. And an audio sequencer for that matter.

//

>> I could use a hobby - I've run out of steam with this string
>> composition that I've been writing, and I need something else to focus
>> on. Anyone with coding experience on here who is willing to help out
>> with this project, feel free - especially with the CSound aspects of
>> it, which I'm a little rough on.
>
> What you're suggesting isn't trivial. Why not have a look
> at Csound instead?

And furthermore, in reference to these two comments, the whole point
is that it _doesn't_ have to be written from scratch. A lot of the
individual portions are already done. We have a few open-source
sequencers that people have written and notation software suites
already. So a lot of the work is already done. The main task is
getting these parts to communicate and then creating a cohesive GUI
for the whole package.

From looking at Csound, it seems much more complicated to build an
entire realistic-sounding orchestra than it would be to just use MIDI.
Or, there's the original idea of having the program save into its own
file format (or some commonly used one or even XML if we wanted), and
then be able to "export" to MIDI so you CAN run it through pianoteq if
you want.

-Mike

Mike Battaglia wrote:
>>> To be honest, anyone that has a text-based interface like microabc
>>> already has it half done. All that remains is for a clever GUI to be
>>> written with some musical notation font and layer that on top of the
>>> text interface. Really isn't that hard.
>> I'm glad you're so confident. I count a notation GUI as a
>> hard thing to write. And an audio sequencer for that matter.
> > //
> >>> I could use a hobby - I've run out of steam with this string
>>> composition that I've been writing, and I need something else to focus
>>> on. Anyone with coding experience on here who is willing to help out
>>> with this project, feel free - especially with the CSound aspects of
>>> it, which I'm a little rough on.
>> What you're suggesting isn't trivial. Why not have a look
>> at Csound instead?
> > And furthermore, in reference to these two comments, the whole point
> is that it _doesn't_ have to be written from scratch. A lot of the
> individual portions are already done. We have a few open-source
> sequencers that people have written and notation software suites
> already. So a lot of the work is already done. The main task is
> getting these parts to communicate and then creating a cohesive GUI
> for the whole package.

So what parts are you planning to re-use? You were talking about MicroABC at the top. Are you familiar with MicroABC? Well, maybe I should say more about this. MicroABC is a pre-processor that produces ABC code. ABC is an ASCII format for notating music. You can convert ABC into PostScript using a C program. ABC programs always seem to be written in C although they're essentially text munging because it's ABC in and PostScript out. Obviously C isn't the best language to do text processing so most of the work gets pushed into the PostScript. The result is that, from both ends, I can't make head or tail of it. So then MicroABC has to somehow produce an output that the next layer can take as input to give a microtonal score. And, it gets better, because the microtonal support requires more than standard ABC can give you there's only one specific ABC processor that works with MicroABC. Even better than that, you need a specific version of the processor, and MicroABC has different code for two different versions.

On the good side, MicroABC will produce microtonal scores and some MIDI output as well. It took a lot of work to get it like that.

On the bad side I can't see how MicroABC would help you write a notation program. You can convert from your native format into ABC and then get a PostScript file. You can even convert that into PDF and use a standard PDF viewer to look at it. It isn't perfect until you print it, but it's readable. But how do you edit that output? You have to know which note the user clicked on. You need to update the display with changes. The existing ABC chain doesn't help you at all with this. The code won't help much because it's C and PostScript. So you're going to end up writing your own notation program from stratch. Which is exactly what I said will be difficult.

If there are open source sequencers, bring them on! Do they do notation? I only know of Rosegarden. Rosegarden is very good but what I've heard about the source code is bad. Going in and adapting that code for microtonality won't be easy because there are 12-equal assumptions all over. And then it isn't cross-platform which means it isn't what we, as a group, want.

>>From looking at Csound, it seems much more complicated to build an
> entire realistic-sounding orchestra than it would be to just use MIDI.
> Or, there's the original idea of having the program save into its own
> file format (or some commonly used one or even XML if we wanted), and
> then be able to "export" to MIDI so you CAN run it through pianoteq if
> you want.

Just use MIDI for what? One advantage of Csound is that there's already a cross platform microtonal sequencer for it: Blue. Maybe you could add notation support to it. As for building an orchestra, yes, it'd be complicated, but you said you wanted a hobby. At least this would give you results one instrument at a time.

That sample library again:

http://wiki.laptop.org/go/Sound_samples

Graham

Mike wrote:
>Disturbingly we share the same wet dream. A microtonal score
>editor/sequencer would probably be the end-all-be-all for me. But I
>think the way to go with it is probably this:
>
>1) Have the software operate in CSound or something
>2) Have the software save the result in both its own proprietary
>format and also "export" to a MIDI file
>3) The MIDI file would be as many channels as is needed to represent
>the composition in its entirety. Then you load it up into
>Pianoteq/VSL/EastWest/Plinky-Dink's-Magic-Synth and roll with it,
>record a few times until you get the whole thing down in audio. then
>you simply release your composition to the world and take your pick
>from the women that swoon over it.

Right! Except the 1st and the 3rd steps are kind of mutually
exclusive, no? I'd leave CSound out of it, myself.

>So for example, if we're composing in 144-et and there is a 31-note
>chord split between 6 instruments, and 98 MIDI channels are needed to
>represent the entire chord,

You only need one MIDI channel for each instrument in the ensumble
if the synth supports realtime retuning messages. If the synth
does not, then you need ceiling(n/128)*instruments channels, where
n is the number of notes per octive in your scale. So in this
case, ceiling(144/128) = 2*6 = 12 channels.

>We then run through them 16 at a time and record
>them with our favorite synths, then layer them together

Not necessary with softsynths, if you can run enough instances
of them at once. An 8-core machine with 7 instances of any
softsynth should definitely take care of you. Alternatively,
you just buy a 7 instances of your favorite rack synth.

>To be honest, anyone that has a text-based interface like microabc
>already has it half done. All that remains is for a clever GUI to be
>written with some musical notation font and layer that on top of the
>text interface. Really isn't that hard.

The GUI part is a lot harder than the microabc part.

>So here is what we need to actually accomplish this:
>
>1) An existing cross-platform microtonal sequencer, preferably coded in C++
>2) An existing cross-platform notation package, preferably coded in C++

'Preferrably coded in my language of choice'. I gotta give you
points for style! But I think we should use C# or some other
modern language. Jeez, I thought you were young or something. :)

>If I can get pointed into the right direction as to the pros and cons
>of the ones out there, I'll definitely start trying to integrate them
>on my own. As of now, there are so many posted in this thread I can't
>tell them all apart - what are the advantages of microabc over other
>software packages, etc? Do people find any of them buggy, etc.? What
>are people's preferences with different programs?
>
>I could use a hobby - I've run out of steam with this string
>composition that I've been writing, and I need something else to focus
>on. Anyone with coding experience on here who is willing to help out
>with this project, feel free - especially with the CSound aspects of
>it, which I'm a little rough on.

Well if C++ is all you know, I guess we'll have to take what we
can get! Just kidding. Go for it!

-Carl

>But I don't think there's anything else. If you can write a
>cross platform audio sequencer with notation support I'm all
>for it.

I'd definitely look hard at Silverlight/Moonlight.

-Carl

>> 1) An existing cross-platform microtonal sequencer, preferably coded in C++
>> 2) An existing cross-platform notation package, preferably coded in C++
>
>Why preferably C++? The rest of us seem to prefer Python.

Python's a great language, but it raises the question of what
widget kit to use. Stuff like gtk really blowz, but you could
do a WPF app in IronPython. Hmm...

http://www.mono-project.com/Moonlight#Alpha_support_for_Silverlight_2.0

>> I could use a hobby - I've run out of steam with this string
>> composition that I've been writing, and I need something else to focus
>> on. Anyone with coding experience on here who is willing to help out
>> with this project, feel free - especially with the CSound aspects of
>> it, which I'm a little rough on.
>
>What you're suggesting isn't trivial. Why not have a look
>at Csound instead?

Please god not CSound.

If you want to write a score editor and insist on doing synthesis
at the same time, the project's chance of failure approximately
quadruples. Keep it MIDI for now, or OSC if you must be weird.

-Carl

>If the goal is to be able to easily compose microtonal music using a
>notation software and be able to play it through a VST sampler such as
>pianoteq or kontakt,

Pianoteq's not a sampler!!!!!!!!! :)

>how can that be accomplished using Csound instead?

You'd replace the "VST sampler" with CSound. Though there
probably is a VST wrapper for CSound at this point (didn't
someone mention one earlier?).

-Carl

>And furthermore, in reference to these two comments, the whole point
>is that it _doesn't_ have to be written from scratch. A lot of the
>individual portions are already done.

What makes you think microabc is suitable for realtime interaction?

>Or, there's the original idea of having the program save into its own
>file format (or some commonly used one or even XML if we wanted), and
>then be able to "export" to MIDI so you CAN run it through pianoteq if
>you want.

Why not have it send MIDI and/or OSC in realtime, as well as
being able to save scores in MIDI and a proprietary format?
The prop. format should probably be XML. There is MusicXML,
but last we looked at it, it wasn't quite suitable for
microtonal use IIRC. Our DTD/schema could start with it,
though, and make the minimum number of changes necessary,
and maybe we could even get them adopted into the standard.

-Carl

There's been a VST wrapper for CSound for a couple of years now, it's part of the distribution.

- Dave

Carl Lumma wrote:
>> If the goal is to be able to easily compose microtonal music using a
>> notation software and be able to play it through a VST sampler such as
>> pianoteq or kontakt,
> > Pianoteq's not a sampler!!!!!!!!! :)
> >> how can that be accomplished using Csound instead?
> > You'd replace the "VST sampler" with CSound. Though there
> probably is a VST wrapper for CSound at this point (didn't
> someone mention one earlier?).
> > -Carl

Well, from here:
http://www.csounds.com/manual/html/vst4cs.html

It seems that data is sent to the VST wrapper by means of the
vstmidiout and vstnote commands, both of which output MIDI. So the
core issue of bypassing MIDI isn't really accomplished here.

Yes, but being as Pianoteq and Kontakt are limited to MIDI input, wouldn't that

On Wed, Jun 18, 2008 at 6:24 PM, Dave Seidel <dave@...> wrote:
> There's been a VST wrapper for CSound for a couple of years now, it's
> part of the distribution.
>
> - Dave
>
> Carl Lumma wrote:
>>> If the goal is to be able to easily compose microtonal music using a
>>> notation software and be able to play it through a VST sampler such as
>>> pianoteq or kontakt,
>>
>> Pianoteq's not a sampler!!!!!!!!! :)
>>
>>> how can that be accomplished using Csound instead?
>>
>> You'd replace the "VST sampler" with CSound. Though there
>> probably is a VST wrapper for CSound at this point (didn't
>> someone mention one earlier?).
>>
>> -Carl
>
>

has anyone been able to do microtonal music with garritan personal
orchestra?

On Wed, Jun 18, 2008 at 7:48 PM, Mike Battaglia <battaglia01@...>
wrote:

> Well, from here:
> http://www.csounds.com/manual/html/vst4cs.html
>
> It seems that data is sent to the VST wrapper by means of the
> vstmidiout and vstnote commands, both of which output MIDI. So the
> core issue of bypassing MIDI isn't really accomplished here.
>
> Yes, but being as Pianoteq and Kontakt are limited to MIDI input, wouldn't
> that
>
> On Wed, Jun 18, 2008 at 6:24 PM, Dave Seidel <dave@...<dave%40superluminal.com>>
> wrote:
> > There's been a VST wrapper for CSound for a couple of years now, it's
> > part of the distribution.
> >
> > - Dave
> >
> > Carl Lumma wrote:
> >>> If the goal is to be able to easily compose microtonal music using a
> >>> notation software and be able to play it through a VST sampler such as
> >>> pianoteq or kontakt,
> >>
> >> Pianoteq's not a sampler!!!!!!!!! :)
> >>
> >>> how can that be accomplished using Csound instead?
> >>
> >> You'd replace the "VST sampler" with CSound. Though there
> >> probably is a VST wrapper for CSound at this point (didn't
> >> someone mention one earlier?).
> >>
> >> -Carl
> >
> >
>
>
>

[Non-text portions of this message have been removed]

Chris V asked,

> has anyone been able to do microtonal music with garritan personal
> orchestra?

Yes.

http://rm-and-jo.laughingsquid.org/FTS-HowTo/MicroOrchestra.html

or Google for "microtuning the orchestra".

And I hear it will be much easier when the next version of GPO has Scala
file support.

Rick

No, that's a package for hosting other VST plugins within CSound. To use CSound as a VST plugin from within another program, like Cubase, you use CSoundVST[1]. (That's what I meant by a wrapper, anyway.) I was wrong about it being in the binary distribution, because of licensing issues, but it is included in the source distribution.

- Dave

[1] http://www.csounds.com/manual/html/CommandCsoundVST.html

Mike Battaglia wrote:
> Well, from here:
> http://www.csounds.com/manual/html/vst4cs.html
> > It seems that data is sent to the VST wrapper by means of the
> vstmidiout and vstnote commands, both of which output MIDI. So the
> core issue of bypassing MIDI isn't really accomplished here.

awesome - thank you so much!!!

On Wed, Jun 18, 2008 at 8:13 PM, Rick McGowan <rick@...> wrote:

> Chris V asked,
>
> > has anyone been able to do microtonal music with garritan personal
> > orchestra?
>
> Yes.
>
> http://rm-and-jo.laughingsquid.org/FTS-HowTo/MicroOrchestra.html
>
> or Google for "microtuning the orchestra".
>
> And I hear it will be much easier when the next version of GPO has Scala
> file support.
>
> Rick
>
>
>

[Non-text portions of this message have been removed]

At 05:13 PM 6/18/2008, you wrote:
>No, that's a package for hosting other VST plugins within CSound. To
>use CSound as a VST plugin from within another program, like Cubase, you
>use CSoundVST[1]. (That's what I meant by a wrapper, anyway.) I was
>wrong about it being in the binary distribution, because of licensing
>issues, but it is included in the source distribution.

In fact the legality of using VST of any kind on Linux is questionable,
which is why there's DSSI.

-Carl

Graham Breed wrote:

> See if this works:
> > http://x31eq.com/tripodcs.zip
> > I run it as
> > csound -M 2 -m 38 -d -odac3 tripod-midi.orc midi.sco
> > but your audio and MIDI numbers will be different. It's > based on an example from The Csound Book. I adapted it to > work in real time, and in a generic way that can be > re-tuned. Then I found I had to adjust the tuning. It's > about as simple as you say -- an oscillator with an envelope > for the reed sound, a filter, a delay line.

I don't think my system is fast enough to run this in real time. I got it to run by overriding the srate and krate (-r 22050 -k 2205), but those things can make a difference in the sound.

>>1) Have the software operate in CSound or something
>>2) Have the software save the result in both its own proprietary
>>format and also "export" to a MIDI file
>>3) The MIDI file would be as many channels as is needed to represent
>>the composition in its entirety. Then you load it up into
>>Pianoteq/VSL/EastWest/Plinky-Dink's-Magic-Synth and roll with it,
>>record a few times until you get the whole thing down in audio. then
>>you simply release your composition to the world and take your pick
>>from the women that swoon over it.
>
> Right! Except the 1st and the 3rd steps are kind of mutually
> exclusive, no? I'd leave CSound out of it, myself.

Well, not quite. The reason I was going to use Csound for the audio
engine is that while we're working IN the software, there would be
none of the problems that we're so used to with MIDI. Then the third
step is just that we can reoutput the file as MIDI perhaps just as an
intermittent step to getting the composition into a VST
softsynth/sampler.

What would you recommend other than Csound?

>>So for example, if we're composing in 144-et and there is a 31-note
>>chord split between 6 instruments, and 98 MIDI channels are needed to
>>represent the entire chord,
>
> You only need one MIDI channel for each instrument in the ensumble
> if the synth supports realtime retuning messages. If the synth
> does not, then you need ceiling(n/128)*instruments channels, where
> n is the number of notes per octive in your scale. So in this
> case, ceiling(144/128) = 2*6 = 12 channels.

Yeah, assuming that for 72-et we want to have an octave and a half
range :P I like your idea though, a spin off of that algorithm might
be effective in cutting the number of MIDI channels down.

But even furthermore, you're talking about the MIDI tuning table
standard, right? Isn't support for that as rare as the realtime
retuning messages? Or do I misunderstand here?

>>We then run through them 16 at a time and record
>>them with our favorite synths, then layer them together
>
> Not necessary with softsynths, if you can run enough instances
> of them at once. An 8-core machine with 7 instances of any
> softsynth should definitely take care of you. Alternatively,
> you just buy a 7 instances of your favorite rack synth.

ha. i sometimes have dreams where I go on bittorrent and manage to
pirate a full nord electro rack. I don't know why or how it gets out
of the computer and into my hand, but it does and it makes sense at
the time.

>>To be honest, anyone that has a text-based interface like microabc
>>already has it half done. All that remains is for a clever GUI to be
>>written with some musical notation font and layer that on top of the
>>text interface. Really isn't that hard.
>
> The GUI part is a lot harder than the microabc part.

Indeed, though I think I'm gonna give it a shot anyway.

>
>>So here is what we need to actually accomplish this:
>>
>>1) An existing cross-platform microtonal sequencer, preferably coded in C++
>>2) An existing cross-platform notation package, preferably coded in C++
>
> 'Preferrably coded in my language of choice'. I gotta give you
> points for style! But I think we should use C# or some other
> modern language. Jeez, I thought you were young or something. :)

Hahaha. Everybody hates C++ now, huh? I definitely wanted to stay away
from Python so that it could be relatively fast and expandable in case
we ever wanted to add audio support or something in the future without
considerable slowdown. Honestly I sort of blew C# off right away as I
avoid Microsoft's oh-so-wonderful "programming innovations" like the
plague nowadays, but if you think it's worthwhile, I'll give it
another looksee. What would you recommend for this project?

>>If I can get pointed into the right direction as to the pros and cons
>>of the ones out there, I'll definitely start trying to integrate them
>>on my own. As of now, there are so many posted in this thread I can't
>>tell them all apart - what are the advantages of microabc over other
>>software packages, etc? Do people find any of them buggy, etc.? What
>>are people's preferences with different programs?
>>
>>I could use a hobby - I've run out of steam with this string
>>composition that I've been writing, and I need something else to focus
>>on. Anyone with coding experience on here who is willing to help out
>>with this project, feel free - especially with the CSound aspects of
>>it, which I'm a little rough on.
>
> Well if C++ is all you know, I guess we'll have to take what we
> can get! Just kidding. Go for it!
>
> -Carl

It isn't all I know, but I definitely like to avoid extremely
high-level programming languages, especially interpreted languages, as
IMO they just cause too many problems when the scope of the project
passes a certain point. Sure it's easy to jump right in, but with a
little more work you can write your own methods for common things here
and have them be infinitely more expandable, avoid bugs in the
interpreter, etc.

I'd like this to be a project that people can contribute to if they
want, so if you have a better idea, I'm all for it.

-Mike

Mike Battaglia wrote:

> Well, not quite. The reason I was going to use Csound for the audio
> engine is that while we're working IN the software, there would be
> none of the problems that we're so used to with MIDI. Then the third
> step is just that we can reoutput the file as MIDI perhaps just as an
> intermittent step to getting the composition into a VST
> softsynth/sampler.

I agree with Carl that there's no point using Csound and then breaking out to MIDI. If you have problems with MIDI you could try looking at Csound without committing yourself to such a big project. In the likely event that you decide you still need some other synths you're better off coding for MIDI (or VST) and ignoring Csound at the application level. You can still use Csound as a MIDI synth if it has something you need.

Audio sequencers handle multi-tracking. That's what I was talking about before. It used to be standard practice in the days when you couldn't run umpteen soft synths simultaneously. You record each part to audio one at a time and then play them back together. You can also mix them outside real time. Starting with an existing audio sequencer means all this work's done for you. What we need is:

- Better support of multiple channels per track

- Piano rolls with different numbers of notes per octave

- Converting between pitch bend and tuning table support

- Staff notation

They get more difficult as you go down the list.

Graham

> I agree with Carl that there's no point using Csound and
> then breaking out to MIDI. If you have problems with MIDI
> you could try looking at Csound without committing yourself
> to such a big project.

Csound is incredible. I've been messing around with it for the past
few days. Probably too much, actually. But I do like programming, and
it would be fun to focus my efforts on a big project, as I haven't
tackled one in a while.

> In the likely event that you decide
> you still need some other synths you're better off coding
> for MIDI (or VST) and ignoring Csound at the application
> level. You can still use Csound as a MIDI synth if it has
> something you need.

Hm, do you think so? Right now I'm confused as to how I'd ever do a
microtonal piece for full orchestra with only 16 channels of pitch
bends and MIDI to work with. Let's say I wrote it in 31tet or
something - to play the 31-tet equivalent of 4:5:6:7 on two different
instruments would require 8 channels right there if the synth you're
working with doesn't support the MIDI tuning standard (which many
don't).

> Audio sequencers handle multi-tracking. That's what I was
> talking about before. It used to be standard practice in
> the days when you couldn't run umpteen soft synths
> simultaneously. You record each part to audio one at a time
> and then play them back together. You can also mix them
> outside real time.

That's what I'm getting at. You go open up Logic or Protools or Sonar
or whatever you want, and throw in the MIDI file, you run as many
channels into your softsynth that you can and run as many softsynths
as your computer can allow, you get the audio version of what you
have, then you repeat. Even if you end up with 128 channels of MIDI
data, and you can only run one 8-channel softsynth at a time - which
means your computer is most likely really, really old - you'd STILL be
able to get the job done just by bouncing it 8 times. Most people I
think will be able to run at least 2 instances of Kontakt at once, so
I think it'd work.

> Starting with an existing audio
> sequencer means all this work's done for you.

That's the plan. But see the above notes on MIDI.

Also, when you talk about audio sequencers right here, are you talking
about GUI-based programs i.e. Logic and such?

Mike Battaglia wrote:
>> I agree with Carl that there's no point using Csound and
>> then breaking out to MIDI. If you have problems with MIDI
>> you could try looking at Csound without committing yourself
>> to such a big project.
> > Csound is incredible. I've been messing around with it for the past
> few days. Probably too much, actually. But I do like programming, and
> it would be fun to focus my efforts on a big project, as I haven't
> tackled one in a while.

Oh, well, that's good. Have fun!

>> In the likely event that you decide
>> you still need some other synths you're better off coding
>> for MIDI (or VST) and ignoring Csound at the application
>> level. You can still use Csound as a MIDI synth if it has
>> something you need.
> > Hm, do you think so? Right now I'm confused as to how I'd ever do a
> microtonal piece for full orchestra with only 16 channels of pitch
> bends and MIDI to work with. Let's say I wrote it in 31tet or
> something - to play the 31-tet equivalent of 4:5:6:7 on two different
> instruments would require 8 channels right there if the synth you're
> working with doesn't support the MIDI tuning standard (which many
> don't).

If it doesn't work with tuning tables, yes. But that means you're restricting yourself to those synthesizers with tuning tables -- which will include Csound. If you use pitch bend tuning you can either use multiple MIDI buses (I didn't think that was a problem these days) or multitrack, as we talk about below. And you'll need one note per channel for legato to work properly in MIDI.

The problem with pitch bend tuning is that your keyboard won't emit the pitch bends. With tuning tables you can enter notes from the keyboard. One thing the software can do is add the pitch bends.

>> Audio sequencers handle multi-tracking. That's what I was
>> talking about before. It used to be standard practice in
>> the days when you couldn't run umpteen soft synths
>> simultaneously. You record each part to audio one at a time
>> and then play them back together. You can also mix them
>> outside real time.
> > That's what I'm getting at. You go open up Logic or Protools or Sonar
> or whatever you want, and throw in the MIDI file, you run as many
> channels into your softsynth that you can and run as many softsynths
> as your computer can allow, you get the audio version of what you
> have, then you repeat. Even if you end up with 128 channels of MIDI
> data, and you can only run one 8-channel softsynth at a time - which
> means your computer is most likely really, really old - you'd STILL be
> able to get the job done just by bouncing it 8 times. Most people I
> think will be able to run at least 2 instances of Kontakt at once, so
> I think it'd work.

If you work on one instrument at a time you'll naturally have one audio track per instrument.

>> Starting with an existing audio
>> sequencer means all this work's done for you.
> > That's the plan. But see the above notes on MIDI.
> > Also, when you talk about audio sequencers right here, are you talking
> about GUI-based programs i.e. Logic and such?

Yes. A MIDI sequencer that can record audio.

Graham

Mike wrote:
>>>1) Have the software operate in CSound or something
>>>2) Have the software save the result in both its own proprietary
>>>format and also "export" to a MIDI file
>>>3) The MIDI file would be as many channels as is needed to represent
>>>the composition in its entirety. Then you load it up into
>>>Pianoteq/VSL/EastWest/Plinky-Dink's-Magic-Synth and roll with it,
>>>record a few times until you get the whole thing down in audio. then
>>>you simply release your composition to the world and take your pick
>>>from the women that swoon over it.
>>
>> Right! Except the 1st and the 3rd steps are kind of mutually
>> exclusive, no? I'd leave CSound out of it, myself.
>
>Well, not quite. The reason I was going to use Csound for the audio
>engine is that while we're working IN the software, there would be
>none of the problems that we're so used to with MIDI.

I still think we'd be best to start off just sending MIDI
and/or OSC and letting the user choose the synthesis tools.
Version 2 can add the native synth engine.

>What would you recommend other than Csound?

I'd probably write a custom toy synth for the purpose.

>>>So for example, if we're composing in 144-et and there is a 31-note
>>>chord split between 6 instruments, and 98 MIDI channels are needed
>>>to represent the entire chord,
>>
>> You only need one MIDI channel for each instrument in the ensumble
>> if the synth supports realtime retuning messages. If the synth
>> does not, then you need ceiling(n/128)*instruments channels, where
>> n is the number of notes per octive in your scale. So in this
>> case, ceiling(144/128) = 2*6 = 12 channels.
>
>Yeah, assuming that for 72-et we want to have an octave and a half
>range :P I like your idea though, a spin off of that algorithm might
>be effective in cutting the number of MIDI channels down.

Yeah, forgot to multiply by the number of octaves required.
It can get a bit unwieldy in extreme cases like this but for
most situations, e.g. a quartet playing in 41-ET, it shouldn't
be too bad.

>But even furthermore, you're talking about the MIDI tuning table
>standard, right? Isn't support for that as rare as the realtime
>retuning messages? Or do I misunderstand here?

I generally prefer controlling synth tuning via Scala files,
but we can certainly offer MTS for those synths that support it.
OSC probably lets you specify the frequency of each note.

>>>So here is what we need to actually accomplish this:
>>>
>>>1) An existing cross-platform microtonal sequencer, preferably
>>>coded in C++
>>>2) An existing cross-platform notation package, preferably
>>>coded in C++
>>
>> 'Preferrably coded in my language of choice'. I gotta give you
>> points for style! But I think we should use C# or some other
>> modern language. Jeez, I thought you were young or something. :)
>
>Hahaha. Everybody hates C++ now, huh? I definitely wanted to stay away
>from Python so that it could be relatively fast and expandable in case
>we ever wanted to add audio support or something in the future without
>considerable slowdown. Honestly I sort of blew C# off right away as I
>avoid Microsoft's oh-so-wonderful "programming innovations" like the
>plague nowadays, but if you think it's worthwhile, I'll give it
>another looksee. What would you recommend for this project?

It'd be nice to have managed code, is all. Could be Java or C#
or whatever. If you know C++, C# should be the easiest of these
to move to.

>It isn't all I know, but I definitely like to avoid extremely
>high-level programming languages, especially interpreted languages, as
>IMO they just cause too many problems when the scope of the project
>passes a certain point. Sure it's easy to jump right in, but with a
>little more work you can write your own methods for common things here
>and have them be infinitely more expandable, avoid bugs in the
>interpreter, etc.

Python can be compiled into a .net project like any other .net
language. I'm not saying you should necessarily use .net, but
it's definitely worth a look. Otherwise it's C++ and gtk or
something like that. I dunno if XUL is up to the task. I guess
it's made strides lately.

-Carl

>>> In the likely event that you decide
>>> you still need some other synths you're better off coding
>>> for MIDI (or VST) and ignoring Csound at the application
>>> level. You can still use Csound as a MIDI synth if it has
>>> something you need.
>>
>> Hm, do you think so? Right now I'm confused as to how I'd ever do a
>> microtonal piece for full orchestra with only 16 channels of pitch
>> bends and MIDI to work with. Let's say I wrote it in 31tet or
>> something - to play the 31-tet equivalent of 4:5:6:7 on two different
>> instruments would require 8 channels right there if the synth you're
>> working with doesn't support the MIDI tuning standard (which many
>> don't).
>
> If it doesn't work with tuning tables, yes. But that means
> you're restricting yourself to those synthesizers with
> tuning tables -- which will include Csound. If you use
> pitch bend tuning you can either use multiple MIDI buses (I
> didn't think that was a problem these days) or multitrack,
> as we talk about below. And you'll need one note per
> channel for legato to work properly in MIDI.

Well, on Windows, for the Microsoft/Roland GM synth I wasn't aware you
could have more than one MIDI bus running on it at at time. If you
can, then I have no reservations about going straight to MIDI.
Otherwise, to use MIDI, I'd have to package some MIDI synth with the
program, similar to how Scala does with Megamid, I suppose. That would
most likely be way, way easier than using Csound.

I think before I even start thinking about this I need to get a better
idea of how you guys have been approaching this via MIDI, as I've run
into a few interesting ideas in this thread on the last 24 hours that
I didn't think were even possible.

> The problem with pitch bend tuning is that your keyboard
> won't emit the pitch bends. With tuning tables you can
> enter notes from the keyboard. One thing the software can
> do is add the pitch bends.

Not sure what you mean here... You're referring to recording MIDI in
live from a keyboard? It seems like if I could get a MIDI echo
function set up, the problem would be solved.

>> That's what I'm getting at. You go open up Logic or Protools or Sonar
>> or whatever you want, and throw in the MIDI file, you run as many
>> channels into your softsynth that you can and run as many softsynths
>> as your computer can allow, you get the audio version of what you
>> have, then you repeat. Even if you end up with 128 channels of MIDI
>> data, and you can only run one 8-channel softsynth at a time - which
>> means your computer is most likely really, really old - you'd STILL be
>> able to get the job done just by bouncing it 8 times. Most people I
>> think will be able to run at least 2 instances of Kontakt at once, so
>> I think it'd work.
>
> If you work on one instrument at a time you'll naturally
> have one audio track per instrument.

What do you mean? Are you talking about the best way for someone to
bounce from a VST synth to an audio track? Or how you'd like the GUI
to look in which multiple channels can be represented in the same
track?

>>> Starting with an existing audio
>>> sequencer means all this work's done for you.
>>
>> That's the plan. But see the above notes on MIDI.
>>
>> Also, when you talk about audio sequencers right here, are you talking
>> about GUI-based programs i.e. Logic and such?
>
> Yes. A MIDI sequencer that can record audio.
>
> Graham

Well, if an open source sequencer like that exists, that would be a
great starting point. Now that I'm becoming aware of some of the ways
to get MIDI to do various microtonal things, it seems to have become a
lot easier.

-Mike

Mike wrote:
>Well, on Windows, for the Microsoft/Roland GM synth
//
>Otherwise, to use MIDI, I'd have to package some MIDI synth with the
>program, similar to how Scala does with Megamid, I suppose. That would
>most likely be way, way easier than using Csound.

What happened to cross-platformness (Microsoft GM is only standard
on Windows)? Also, why must you bundle it with a synth? Can't
users provide their own?

-Carl

> I still think we'd be best to start off just sending MIDI
> and/or OSC and letting the user choose the synthesis tools.
> Version 2 can add the native synth engine.

Well, if we use OSC, what synths are out there that operate on it?
Hell, I'll package one with it if it's that much easier to work with
than MIDI. MIDI has some limitations in my mind that seem
insurmountable, but maybe you have some insight on how to get past
them. Namely, it seems like if I'm going to want to play a four-note
chord in 31tet, for example, I'm going to use four MIDI channels,
unless the synth supports MTS, which I can't find any that do except
NI's FM8. So I think relying on the MTS support to be there would be
futile, as even if we left it up to the user to choose the synth
tools, what tools are there that support it? Even Scala has the option
of doing it with channel swapping and pitch-bending, which is fine. So
we'd need to make sure that we can run multiple instances of whatever
softsynth the user chooses, or else the program simply won't work.

>>What would you recommend other than Csound?
>
> I'd probably write a custom toy synth for the purpose.

Yeah. I think that in the grand scheme of things for this project
having a custom synth shouldn't be that important anyway. We're really
looking for people to be able to write microtonal music and hear what
it would sound like in real time. People are going to want to use
their own synths anyway, but if they can't get Kontakt running enough
times to handle all of the MIDI channels they might need, then it's
useless. So the option should be available to do that if the user has
the CPU power/RAM for it, but if not, then the option should also
exist to use at least some other GM synth we can have 8 instances of
running if need be.

So I envision a "high-level" and a "low-level" of operation. Perhaps
in realtime, if their computer isn't powerful enough, they'll still be
able to work with some standard cheesy GM sounds, and then when
they're done they can bounce to audio by rerecording a few times like
we were discussing above.

> I generally prefer controlling synth tuning via Scala files,
> but we can certainly offer MTS for those synths that support it.
> OSC probably lets you specify the frequency of each note.

What do you mean via scala files? How would that let us get more than
12 notes per octave when dealing with MIDI? We'd still need pitch
bends, right?

> It'd be nice to have managed code, is all. Could be Java or C#
> or whatever. If you know C++, C# should be the easiest of these
> to move to.

Well, managed code certainly makes portability easier, but if we're
talking about eventually turning this into something that could run
several VST softsynths at once and have a whole audio suite and such
while still maintaining a rigid 960 ppq, won't that be much harder
using something with a VM? Even if for right now it isn't that
complicated, it seems like using a VM is going to impose a ceiling on
the whole thing. I haven't messed with .NET too much, so maybe its VM
is capable of handling that sort of thing.

> Python can be compiled into a .net project like any other .net
> language. I'm not saying you should necessarily use .net, but
> it's definitely worth a look. Otherwise it's C++ and gtk or
> something like that. I dunno if XUL is up to the task. I guess
> it's made strides lately.

If the performance hit by using .NET wouldn't be too severe, I'd do it
in that. The jury seems to be still out on that one though.

-Mike

Mike Battaglia wrote:

> Well, on Windows, for the Microsoft/Roland GM synth I wasn't aware you
> could have more than one MIDI bus running on it at at time. If you
> can, then I have no reservations about going straight to MIDI.
> Otherwise, to use MIDI, I'd have to package some MIDI synth with the
> program, similar to how Scala does with Megamid, I suppose. That would
> most likely be way, way easier than using Csound.

Cross that bridge when you come to it.

> I think before I even start thinking about this I need to get a better
> idea of how you guys have been approaching this via MIDI, as I've run
> into a few interesting ideas in this thread on the last 24 hours that
> I didn't think were even possible.

That's good! What I actually did with MIDI is have one instrument at a time running on the Capybara and multitrack them in the sequencer. Partly that was the easiest way of using Kyma.

>> The problem with pitch bend tuning is that your keyboard
>> won't emit the pitch bends. With tuning tables you can
>> enter notes from the keyboard. One thing the software can
>> do is add the pitch bends.
> > Not sure what you mean here... You're referring to recording MIDI in
> live from a keyboard? It seems like if I could get a MIDI echo
> function set up, the problem would be solved.

Either live from the keyboard or with a step input. With Cubasis I could click on a note and play a note to change the pitch. With more than 12 notes to the octave I could never remember where each note was on the piano roll but I knew how to play them. So when I was drawing in notes I'd just start with an arbitrary pitch and play the keyboard to set it right.

I don't know if you have ambitions for keyboards with more than 128 notes to the octave. I could record multiple channels from my Ztar not problem but the piano roll gave no indication of which channel was which. That was enough of a pain that I generally avoided using such mappings with a sequencer even when I was recording one instrument at a time so there was no problem with the channels clashing. These details can make a difference and the people who wrote the sequencer never thought of it being used the way I was using it.

>> If you work on one instrument at a time you'll naturally
>> have one audio track per instrument.
> > What do you mean? Are you talking about the best way for someone to
> bounce from a VST synth to an audio track? Or how you'd like the GUI
> to look in which multiple channels can be represented in the same
> track?

The best way to record MIDI synthesizers. Maybe VST does it more naturally. I've never used VST.

>>>> Starting with an existing audio
>>>> sequencer means all this work's done for you.
>>> That's the plan. But see the above notes on MIDI.
>>>
>>> Also, when you talk about audio sequencers right here, are you talking
>>> about GUI-based programs i.e. Logic and such?
>> Yes. A MIDI sequencer that can record audio.
> > Well, if an open source sequencer like that exists, that would be a
> great starting point. Now that I'm becoming aware of some of the ways
> to get MIDI to do various microtonal things, it seems to have become a
> lot easier.

That's exactly the problem. There's Rosegarden but it's Linux only. Maybe you could see what else is out there. Probably what you'll lose is the notation because that's what Rosegarden specializes in.

Graham

>>Well, on Windows, for the Microsoft/Roland GM synth
> //
>>Otherwise, to use MIDI, I'd have to package some MIDI synth with the
>>program, similar to how Scala does with Megamid, I suppose. That would
>>most likely be way, way easier than using Csound.
>
> What happened to cross-platformness (Microsoft GM is only standard
> on Windows)? Also, why must you bundle it with a synth? Can't
> users provide their own?

Well, it's definitely gonna be cross-platform. I was just pointing out
one example. I expect there will be a large portion of users that
don't have any softsynths to use and so I don't want to cut them out.
If the solution is simple as packaging a synth with the program like
Scala does, I'd rather do that just so it's operable out of the box
than have to deal with

a) the possibility that we won't be able to run enough instances of
whatever synth the user has to be able to play music
b) the possibility that the user won't be able to find a synth that
WILL work and leave them stranded

So I really think it important to have at least one CPU-friendly
GM-capable synth at least just so the program will run. Right now I'm
not sure to do it any other way - I don't know any softsynths I can
run so many times that I'll be able to handle a full orchestra, for
example, and the program has to be able to deal with projects on that
level of complexity. Megamid might actually be the solution - I'm
looking at it now, and it seems pretty lightweight and portable.

-Mike

Mike Battaglia wrote:

> Well, managed code certainly makes portability easier, but if we're
> talking about eventually turning this into something that could run
> several VST softsynths at once and have a whole audio suite and such
> while still maintaining a rigid 960 ppq, won't that be much harder
> using something with a VM? Even if for right now it isn't that
> complicated, it seems like using a VM is going to impose a ceiling on
> the whole thing. I haven't messed with .NET too much, so maybe its VM
> is capable of handling that sort of thing.

Audio sequencers worked 10 years ago so they should work now with managed code. Besides most of the audio-level work should be in a native code library anyway. Java has a built-in sequencer class for example. If Silverlight does the same that'd be a reason for using it.

Graham

--- In MakeMicroMusic@yahoogroups.com, Graham Breed <gbreed@...> wrote:
>
> Chuckk Hubbard wrote:

> > Well, one feature is that one may recenter the entire score (or just
> > parts of it) by changing, globally or locally, where 1/1 is. Thus no
> > "scale" is necessary; anything within the 13-limit is possible by
> > changing 1/1 enough times. No reason this couldn't work with tempered
> > tunings, but for most people it isn't necessary.
>
> Depending on whether most people want just intonation or not ...

I mean, if people are working in tempered tunings they don't
necessarily need to recenter the score each time they change "keys".
With JI, I find it far better to change 16/15 to 1/1 if I switch to
that tonality rather than using intervals like 32/25.
Also, I'm far more passionate about extended JI, personally, and to be
honest, I'm writing it for selfish reasons.

> How's it notated?

If you're using numbers right now, I think you'll like this:
Fractions. I stand fully behind Partch in this regard; I have lost
interest in infinite systems of accidentals. I say if you want to use
JI to any great extent you have to use the numbers. Once my program
actually WORKS, then others will be free to recast it with whatever
notation they like, or to make whatever requests/suggestions. It'll
be open source.
One big score- since a 1/2-inch staff is too small for all the
possibilities- with all instruments in the same area; each note has a
triangle whose size shows amplitude, and a thin horizontal line to
show duration. To the left of this shape is a fraction showing
frequency ratio, and on top and bottom another fraction showing the
local tonal center.
If one chooses to use "held note" and individual voice support in
Csound, i.e., "i1.1 0 -1 ...." in score language, all notes with the
same instrument/voice combination will have a light line connecting
them. This way one can do portamentos and such, and slurs are visible.

> > The advantage, as I understand it, is that one can have any number of
> > FluidSynth engines, so any number of microtunings simultaneously. I
> > asked about it a while back on the Csound list, maybe it's time to do
> > some more research.
>
> Can't you do that with MIDI? Well, I suppose once you've
> written a microtonal instrument using a FluidSynth instance
> it won't be that complex to use.

Not sure, I haven't had much luck in the past. Csound does it all
internally and just puts out the sound, so good for it. I'll have a
default soundfont instrument for those who want to use soundfonts and
not bother with Csound code, and then the option to manually edit
one's own Csound orchestra. Anyway, the soundfont specification would
take months to sort out and code into my program, so, since I'm using
Csound for audio anyway, I'll be using it to load soundfonts too.

>
> > I'm a devoted Linux user myself, but I see that as Rosegarden's main
> > limitation; if anyone cares about influencing musicians or broadening
> > the microtonal community, cross-platform software is a must.
>
> But I don't think there's anything else. If you can write a
> cross platform audio sequencer with notation support I'm all
> for it.

I best get busy.
Unfortunately I've been invited (commanded) to head out to the
countryside for the weekend, so I may be facing some delays.

-Chuckk

--- In MakeMicroMusic@yahoogroups.com, Carl Lumma <carl@...> wrote:
>
> At 05:13 PM 6/18/2008, you wrote:
> >No, that's a package for hosting other VST plugins within CSound. To
> >use CSound as a VST plugin from within another program, like
Cubase, you
> >use CSoundVST[1]. (That's what I meant by a wrapper, anyway.) I was
> >wrong about it being in the binary distribution, because of licensing
> >issues, but it is included in the source distribution.
>
> In fact the legality of using VST of any kind on Linux is questionable,
> which is why there's DSSI.

As I understand it there's no legality problems *using* VST on Linux,
only with *distributing* it. Debian won't carry it, but I believe
Steinberg releases the actual code, they just have a license
restricting its use that Debian and GNU don't appreciate. But it
ought to work fine on Linux, and a Linux user is in no way legally
restricted to running only free software on his or her system. Me,
I'm happy to use DSSI and LADSPA, so I don't recall ever having tried
to use VST on Linux.
Csound also now supports writing one's own LADSPA plugins in the
Csound language, but last I read there wasn't much interest in DSSI
among the developers since there was lots of talk about a new
standard, what was it called?? LV2?

-Chuckk

I was rereading through the posts on this list, and somehow I missed this one.

Graham wrote:
>> And furthermore, in reference to these two comments, the whole point
>> is that it _doesn't_ have to be written from scratch. A lot of the
>> individual portions are already done. We have a few open-source
>> sequencers that people have written and notation software suites
>> already. So a lot of the work is already done. The main task is
>> getting these parts to communicate and then creating a cohesive GUI
>> for the whole package.
>
> So what parts are you planning to re-use? You were talking
> about MicroABC at the top. Are you familiar with MicroABC?
> Well, maybe I should say more about this. MicroABC is a
> pre-processor that produces ABC code. ABC is an ASCII
> format for notating music. You can convert ABC into
> PostScript using a C program. ABC programs always seem to
> be written in C although they're essentially text munging
> because it's ABC in and PostScript out. Obviously C isn't
> the best language to do text processing so most of the work
> gets pushed into the PostScript. The result is that, from
> both ends, I can't make head or tail of it. So then
> MicroABC has to somehow produce an output that the next
> layer can take as input to give a microtonal score. And, it
> gets better, because the microtonal support requires more
> than standard ABC can give you there's only one specific ABC
> processor that works with MicroABC. Even better than that,
> you need a specific version of the processor, and MicroABC
> has different code for two different versions.
>
> On the good side, MicroABC will produce microtonal scores
> and some MIDI output as well. It took a lot of work to get
> it like that.
>
> On the bad side I can't see how MicroABC would help you
> write a notation program. You can convert from your native
> format into ABC and then get a PostScript file. You can
> even convert that into PDF and use a standard PDF viewer to
> look at it. It isn't perfect until you print it, but it's
> readable. But how do you edit that output? You have to
> know which note the user clicked on. You need to update the
> display with changes. The existing ABC chain doesn't help
> you at all with this. The code won't help much because it's
> C and PostScript. So you're going to end up writing your
> own notation program from stratch. Which is exactly what I
> said will be difficult.

I wish I had seen this earlier. I just did a whole bunch of research
into MicroABC to hit this dead end. So I'm going to have to look
elsewhere.

> If there are open source sequencers, bring them on! Do they
> do notation? I only know of Rosegarden. Rosegarden is very
> good but what I've heard about the source code is bad.
> Going in and adapting that code for microtonality won't be
> easy because there are 12-equal assumptions all over. And
> then it isn't cross-platform which means it isn't what we,
> as a group, want.

So I've been doing some research into possibly porting Rosegarden, and
from what I've found, people keep saying that the challenge doesn't
lie in porting the GUI or really anything except the audio engine,
which is currently tied into Linux's ALSA/JACK stuff. Since we've been
discussing really only MIDI so far, this might not be a problem. And
when it becomes a problem, we might be able to port it using portaudio
or something.

But in reference to this:

> Rosegarden is very good but what I've heard about the source code is bad.

Are you saying that the source code is bad for -microtonality- because
it's tied into 12-equal? Or are you saying that the source code is bad
because it's poorly written or unreadable or something?

> Just use MIDI for what? One advantage of Csound is that
> there's already a cross platform microtonal sequencer for
> it: Blue. Maybe you could add notation support to it. As
> for building an orchestra, yes, it'd be complicated, but you
> said you wanted a hobby. At least this would give you
> results one instrument at a time.
>
> That sample library again:
>
> http://wiki.laptop.org/go/Sound_samples
>
> Graham

I must have missed the post about this earlier, but damn. That's
insane. There are so many free resources out there now for almost
anything you might want to do with audio, really amazing.

-Mike

--- In MakeMicroMusic@yahoogroups.com, Graham Breed <gbreed@...>
wrote:
>
> Mike Battaglia wrote:
> >
> > Well, if an open source sequencer like that exists, that would be
> > a great starting point. Now that I'm becoming aware of some of
> > the ways to get MIDI to do various microtonal things, it seems to
> > have become a lot easier.
>
> That's exactly the problem. There's Rosegarden but it's
> Linux only. Maybe you could see what else is out there.
> Probably what you'll lose is the notation because that's
> what Rosegarden specializes in.
>

I am thinking about Jazz++ (http://jazzplusplus.sourceforge.net). No
notation and no audio recording, but open source and running on both
Windows and Linux, currently being ported to the wxWidgets framework
so it will run on Mac, too. Dunno how far the portation has
proceeded, though.
--
Hans Straub

--- In MakeMicroMusic@yahoogroups.com, Carl Lumma <carl@...> wrote:
>
>
> >What is to be done, in my eyes, would be:
> >
> >Take a usual standard sequencer, with a piano-roll e.g, and add a
> >layer that converts between the paino roll and the actual midi
> >events - concrete: notes of one "virtual" microtonal channel would
> >have to be distributed to two or more standard midi channels,
> >according to a definable pattern. In the case of 72tet, this would
> >be, e.g.:
> >
> >"virtual" note 1 to standard midi note 1 of channel 1
> >"virtual" note 2 to standard midi note 1 of channel 2
> >"virtual" note 3 to standard midi note 1 of channel 3
> >"virtual" note 4 to standard midi note 1 of channel 4
> >"virtual" note 5 to standard midi note 1 of channel 5
> >"virtual" note 6 to standard midi note 1 of channel 6
> >"virtual" note 7 to standard midi note 2 of channel 1
> >"virtual" note 8 to standard midi note 2 of channel 2
> >
> >etc.
>
> Meh. I'd rather have a piano roll that alpha blends to staff
> notation (an idea I've been toying with for some years).
>

What do you mean exactly here? A piano roll with a staff above it,
and edit operations at any place propagated to the other in realtime?
--
Hans Straub

> I am thinking about Jazz++ (http://jazzplusplus.sourceforge.net). No
> notation and no audio recording, but open source and running on both
> Windows and Linux, currently being ported to the wxWidgets framework
> so it will run on Mac, too. Dunno how far the portation has
> proceeded, though.
> --
> Hans Straub

Looks interesting, also I've been looking at this one:

http://tse3.sourceforge.net/ - An open source "sequencer engine" - all
it lacks is a GUI (yes, any GUI at all). Might be useful though.

http://noteedit.berlios.de/index.html might be interesting to start
with/port as well.

-Mike

On Thu, Jun 19, 2008 at 6:47 AM, Mike Battaglia <battaglia01@...> wrote:
>> I am thinking about Jazz++ (http://jazzplusplus.sourceforge.net). No
>> notation and no audio recording, but open source and running on both
>> Windows and Linux, currently being ported to the wxWidgets framework
>> so it will run on Mac, too. Dunno how far the portation has
>> proceeded, though.
>> --
>> Hans Straub
>
> Looks interesting, also I've been looking at this one:
>
> http://tse3.sourceforge.net/ - An open source "sequencer engine" - all
> it lacks is a GUI (yes, any GUI at all). Might be useful though.
>

I use Csound and a front end I wrote 15 years ago in Turbo Pascal.
Pascal is a little dated, but I still have customers running 40 year
old FORTRAN and 25 year old COBOL applications, so I feel like I'm in
good company.

Csound has a very stiff learning curve, as others have discovered. I
climbed it ten years ago, and have enjoyed it in the limited way I use
it ever since. If anyone tries it for microtonal work, I'd be glad to
help with that aspect of it. It does allow for any pitch and any
instrument with no practical limitations except CPU speed and storage.
I use it as a sampler, and then control as many aspects as possible to
humanize the sound. Fake but accurate is my motto. I generally have
about 10 different instruments with 53 or fewer pitches per octave,
and it fits well in that range.

A recent experiment is a realization of John Cage's In a Landscape in
53-TET. You can hear it here: http://bumpermusic.blogspot.com .

Prent Rodgers

Mike Battaglia wrote:

> So I've been doing some research into possibly porting Rosegarden, and
> from what I've found, people keep saying that the challenge doesn't
> lie in porting the GUI or really anything except the audio engine,
> which is currently tied into Linux's ALSA/JACK stuff. Since we've been
> discussing really only MIDI so far, this might not be a problem. And
> when it becomes a problem, we might be able to port it using portaudio
> or something.

There's talk of KDE working on Windows. I don't know how much that'll help.

> But in reference to this:
> >> Rosegarden is very good but what I've heard about the source code is bad.
> > Are you saying that the source code is bad for -microtonality- because
> it's tied into 12-equal? Or are you saying that the source code is bad
> because it's poorly written or unreadable or something?

From what I've heard it's not easy to go in and make changes. The code is very complex and tied to the original requirements -- 12-equal, Linux, and so on.

Graham

Chuckk Hubbard wrote:

> I mean, if people are working in tempered tunings they don't
> necessarily need to recenter the score each time they change "keys". > With JI, I find it far better to change 16/15 to 1/1 if I switch to
> that tonality rather than using intervals like 32/25.
> Also, I'm far more passionate about extended JI, personally, and to be
> honest, I'm writing it for selfish reasons.

If a temperament's being notated with ratios maybe you'll want to recenter it to show there's been a comma shift. Or maybe the temperament would deal with that itself.

>> How's it notated?
> > If you're using numbers right now, I think you'll like this:
> Fractions. I stand fully behind Partch in this regard; I have lost
> interest in infinite systems of accidentals. I say if you want to use
> JI to any great extent you have to use the numbers. Once my program
> actually WORKS, then others will be free to recast it with whatever
> notation they like, or to make whatever requests/suggestions. It'll
> be open source.
> One big score- since a 1/2-inch staff is too small for all the
> possibilities- with all instruments in the same area; each note has a
> triangle whose size shows amplitude, and a thin horizontal line to
> show duration. To the left of this shape is a fraction showing
> frequency ratio, and on top and bottom another fraction showing the
> local tonal center.
> If one chooses to use "held note" and individual voice support in
> Csound, i.e., "i1.1 0 -1 ...." in score language, all notes with the
> same instrument/voice combination will have a light line connecting
> them. This way one can do portamentos and such, and slurs are visible.

It'd be usable for temperaments, then, but not ideal.

> Not sure, I haven't had much luck in the past. Csound does it all
> internally and just puts out the sound, so good for it. I'll have a
> default soundfont instrument for those who want to use soundfonts and
> not bother with Csound code, and then the option to manually edit
> one's own Csound orchestra. Anyway, the soundfont specification would
> take months to sort out and code into my program, so, since I'm using
> Csound for audio anyway, I'll be using it to load soundfonts too.

Er, if the Csound folks can embed FluidSynth can't you embed it directly as well?

Graham

--- In MakeMicroMusic@yahoogroups.com, "Mike
Battaglia" <battaglia01@...> wrote:
> Well, if we use OSC, what synths are out there that operate on it?

Reaktor and Bidule were listed as OSC compatible on the old OSC site
at Berkeley, but neither are listed at the http://
opensoundcontrol.org/implementations.

Perhaps there is some confusion as to which products directly support
OSC and which require some sort of wrapper app. There are wrapper
apps for interfacing OSC with Ableton Live, Max/MSP, etc. I think
its safe to say all OSC compatible synths (with or without wrappers)
are softsynths.

> What do you mean via scala files? How would that let us get more
than
> 12 notes per octave when dealing with MIDI? We'd still need pitch
> bends, right?

Unless the synth on the receiving side can load Scala files or
respond to MTS, in which case no pitch bend would be required.

One possible workaround is to offer a patch conversion tool that will
convert someone's favorite softsynth patch into a patch format that
works with synth that is compatible with Scala and/or MTS. The Nord
Modular G2 community for example produced Yamaha DX-series patch->G2
patch conversion tools. A Scala->G2 loader project was also started,
but it's incomplete due to no .kbm file support (the loader assumes
you want to start on the MIDI Note Number 0 or something like that,
and makes other assumptions about your keyboard mapping, due to lack
of, again Scala .kbm file support).

Just ideas here...

--- In MakeMicroMusic@yahoogroups.com, "Prent Rodgers"
<prentrodgers@...> wrote:
>
> I use Csound and a front end I wrote 15 years ago in Turbo Pascal.
> Pascal is a little dated, but I still have customers running 40 year
> old FORTRAN and 25 year old COBOL applications, so I feel like I'm
in
> good company.
>
> Csound has a very stiff learning curve, as others have discovered. I
> climbed it ten years ago, and have enjoyed it in the limited way I
use
> it ever since. If anyone tries it for microtonal work, I'd be glad
to
> help with that aspect of it. It does allow for any pitch and any
> instrument with no practical limitations except CPU speed and
storage.
> I use it as a sampler, and then control as many aspects as possible
to
> humanize the sound. Fake but accurate is my motto. I generally have
> about 10 different instruments with 53 or fewer pitches per octave,
> and it fits well in that range.
>
> A recent experiment is a realization of John Cage's In a Landscape
in
> 53-TET. You can hear it here: http://bumpermusic.blogspot.com .
>
> Prent Rodgers
>
From Robert. There are trillions of dollars invested in the
computer codes of FORTRAN and COBOL and governments around the world
are not about to throw it away. So you are certainly in illustrious
company. Your music is interesting and quite exotic sounding at times
and deserves closer attention.

At 01:17 AM 6/19/2008, you wrote:
>> I still think we'd be best to start off just sending MIDI
>> and/or OSC and letting the user choose the synthesis tools.
>> Version 2 can add the native synth engine.
>
>Well, if we use OSC, what synths are out there that operate on it?

A few. Reaktor that I can remember. There are others.

>Hell, I'll package one with it if it's that much easier to work with
>than MIDI. MIDI has some limitations in my mind that seem
>insurmountable, but maybe you have some insight on how to get past
>them. Namely, it seems like if I'm going to want to play a four-note
>chord in 31tet, for example, I'm going to use four MIDI channels,

Why??

>unless the synth supports MTS, which I can't find any that do except
>NI's FM8.

Hare are some MTS-compatible synths:

Alphakanal Automat
AnaMark softsynth
Audio Compositor
Big Tick: Angelina, Rainbow and Rhino
Bitheadz: Unity
Cakewalk: Dimension Pro, Rapture, Z3TA+
Camel Audio: Cameleon5000
ChucK
E-mu: Morpheus, Proteus series
Ensoniq: EPS/EPS16/ASR10, VFX
Fluidsynth
Korg: M1, M1R, X5DR, OASYS PCI soundcard
LinPlug: Albino 2, Alpha 2, CronoX, Octopus, Organ 3 and Sophistry
Native Instruments: FM7, Pro-52/Pro-53
Pure Data
Roland: GS & JV/XP families, Fantom-X6/X7/X8, V-Synth 2.0, VSC, SC-8850
Synapse Audio: Orion Pro softsynth
Timidity
Tobybear Helios
Turtle Beach: Multisound, Monterey, Maui, Tropez, Rio
VAZ Plus
VirSyn Cube, Cantor and TERA 2
Yamaha DX7II/TX802SY77/TG77/SY99/VL-1/VL-7/TX81Z/DX11/DX27/DX100/V50
Yamaha XG family
Yamaha VL70m
Wusik Wusikstation v2
Zebra 2.0

Here are some .scl-compatible synths:

AlsaModularSynth by Matthias Nagorni
Astralis v2 by Homegrown Sounds
Blue by Steven Yi and Tracker Sound Object
Cantor by VirSyn Software Synthesizer
Dimension Pro 1.2, Rapture, and Z3TA+ by Cakewalk
Lounge Lizard by Applied Acoustics
Max Magic Microtuner by Victor Cerullo
Nord Modular G2 via G2ools converter by Ian Sayer
Oblivion VSTi for Windows by David R. Frost
Pianoteq by MODARTT
Scala Microtuners for Kontakt and SynthEdit by Robert Strauss
TERA 2.0 by VirSyn
Virtual Pipe Organs by Garritan
ZynAddSubFX software synthesizer by Paul Nasca

>Even Scala has the option
>of doing it with channel swapping and pitch-bending, which is fine.

It's great for quick-and-dirty, but it's a hack and not suitable
for serious production. I'd suggest we not implement it.

>> I generally prefer controlling synth tuning via Scala files,
>> but we can certainly offer MTS for those synths that support it.
>> OSC probably lets you specify the frequency of each note.
>
>What do you mean via scala files? How would that let us get more than
>12 notes per octave when dealing with MIDI? We'd still need pitch
>bends, right?

If a synth supports them, you don't need any pitch bends, no.
You just send plain MIDI and the synth knows how to tune it. It
can be a pain when doing channel splitting, but other than that
it's a great way to work.

>> It'd be nice to have managed code, is all. Could be Java or C#
>> or whatever. If you know C++, C# should be the easiest of these
>> to move to.
>
>Well, managed code certainly makes portability easier, but if we're
>talking about eventually turning this into something that could run
>several VST softsynths at once and have a whole audio suite and such
>while still maintaining a rigid 960 ppq, won't that be much harder
>using something with a VM?

It's not very often you can beat a modern garbage collector by
hand coding. More likely you'll f-up and have memory leaks.

>Even if for right now it isn't that
>complicated, it seems like using a VM is going to impose a ceiling on
>the whole thing. I haven't messed with .NET too much, so maybe
>its VM is capable of handling that sort of thing.

.net isn't a VM, it's native binaries at the end of the day.
And I don't see why you're equating our performance with
softsynth load. Whatever VST or MIDI synths we'll use will
be binaries. The OS takes care of the multitasking, not us.
And unless we inspect the system and make a guess, we won't
be able to know how many instances of which synth the user
will be able to run. But your idea of picking a lightweight
one and bundling it is good. Except instead of a bundle you
could just put it in the system requirements and that saves
you any licensing trouble.

>> Python can be compiled into a .net project like any other .net
>> language. I'm not saying you should necessarily use .net, but
>> it's definitely worth a look. Otherwise it's C++ and gtk or
>> something like that. I dunno if XUL is up to the task. I guess
>> it's made strides lately.
>
>If the performance hit by using .NET wouldn't be too severe,
>I'd do it in that. The jury seems to be still out on that
>one though.

It does? All windows development has been .net for years.
Even if you write C++ I believe it's compiled to .net CLR
first.

-Carl

>> In fact the legality of using VST of any kind on Linux is questionable,
>> which is why there's DSSI.
>
>As I understand it there's no legality problems *using* VST on Linux,
>only with *distributing* it. Debian won't carry it, but I believe
>Steinberg releases the actual code, they just have a license
>restricting its use that Debian and GNU don't appreciate.

I forget what Paul Davies told me, but I think it was worse
than that. I don't recall that the code was released, and
whatever the situation was, it was a snafu since Yamaha's
acquisition of Steinberg -- they simply weren't replying to
correspondence. That was a year ago already so maybe things
have improved.

-Carl

Carl says:
> It does? All windows development has been .net for years.
> Even if you write C++ I believe it's compiled to .net CLR
> first.

All of them. I think you overstate this a little. This is only true of
projects using Microsoft development tools. That is a lot of them but a long
way from all of them. Especially open source cross platform projects. We are
talking cross platform here I hope.

[Non-text portions of this message have been removed]

Hi Hans!

>> Meh. I'd rather have a piano roll that alpha blends to staff
>> notation (an idea I've been toying with for some years).
>
>What do you mean exactly here? A piano roll with a staff above it,
>and edit operations at any place propagated to the other in realtime?

The key ingredients in my scheme are:

* Not WSYWIG, in the sense that the onscreen display is
optimized for onscreen editing. There's a separate step
for page layout if you want to print. This is a core
part of the philosophy.

* So the editing display shows a linear timeline, not a
"page". The key is that 'now' is always at the same
spot. We don't move a caret across the screen and then
update, we *smoothly scroll* content under a fixed caret.
Very important for *fast* mouse editing, and works well
with a 'object properties inspector' paradigm.

* The timeline can show a regular staff with notes, or
a piano roll, with a slider for alpha blending between
them. Music notation is equivalent to a piano roll,
except: 1. duration is expressed with flags and notehead
coloring instead of bar length 2. vertical position
increments aren't all the same.
If the slider is moved all the way to the piano roll
position (so the traditional notes are invisible) the
only difference between what you'll see and a traditional
piano roll will be #2 (the unequal vertical increment).

* The point of the piano roll view is: 1. to assist people
who aren't good with regular music notation and 2. to
assist people who aren't good with the advanced
polyrhythmic notations (like Henry Cowell's) I intend to
implement, which would be everybody.

* As already mentioned, when creating a new score we are
prompted to specify the vertical increments (ie < L L s >
etc.), accidental font file, MIDI offsets into the
accidentals, and notehead shape and flag meanings.
We'll ship with a few presets, including one that gives
common practice notation, and users can save their own.

-Carl

At 03:47 AM 6/19/2008, you wrote:
>> I am thinking about Jazz++ (http://jazzplusplus.sourceforge.net). No
>> notation and no audio recording, but open source and running on both
>> Windows and Linux, currently being ported to the wxWidgets framework
>> so it will run on Mac, too. Dunno how far the portation has
>> proceeded, though.
>> --
>> Hans Straub
>
>Looks interesting, also I've been looking at this one:
>
>http://tse3.sourceforge.net/ - An open source "sequencer engine" - all
>it lacks is a GUI (yes, any GUI at all). Might be useful though.

Guys: if you want to write a notation package (VERY hard problem),
my advice is to forget about sequencer functionality. One
thing at a time.

-Carl

At 09:36 AM 6/19/2008, you wrote:
> Carl says:
>> It does? All windows development has been .net for years.
>> Even if you write C++ I believe it's compiled to .net CLR
>> first.
>
>All of them. I think you overstate this a little. This is only true of
>projects using Microsoft development tools. That is a lot of them but
>a long way from all of them.

You're right, "Windows development" was too inclusive a term.

>Especially open source cross platform
>projects. We are talking cross platform here I hope.

Yes, cross-platform is obviously essential for this community.
I'm not a Microsoft developer or even shareholder any more, but
their technology is superior in my opinion. Silverlight isn't
mature yet and it may even be too early to do an application
like this with it, but it's the only thing out there that could
reasonably be called a next-generation API.

The application I'm envisioning has a 'flat' interface like
Ableton Live.

Hey- fruity loops is cross-platform. What widgets do they use?
Something custom probably. I wonder if it's open...

-Carl

On Thu, Jun 19, 2008 at 12:55 PM, Carl Lumma <carl@...> wrote:
> I'm not a Microsoft developer or even shareholder any more, but
> their technology is superior in my opinion.

Speaking of subjective experience I find that most people mean "It
does what I expect" when they say something is superior even though
their expectations are heavily colored by experience. That is a large
part of the reason that it can be dangerous to trust your own
experience. I think it is the primary explanation for all those
flame/holy wars between Mac vs Microsoft vs Linux advocates or vi vs
emacs text editor fanatics (or if you are old enough Ford vs Chevy
fans.)

I first noticed this phenomenon many years ago (1982) when I had seen
the Lisa and the Xerox Star but the Macintosh wasn't out yet and we
were trying to design a graphical GUI type interface for a scientific
instrument. We had a viewing window on the display. The contents were
controlled by a joystick. Left and right scrolled the image and front
to back zoomed. The big question was "when the joystick is pushed
right which direction should the pixels move?" I polled almost
everyone in the company (I added a switch that allowed people to try
either one.) To my surprise it was a 50/50 split. When I asked why for
most it was an instinctive feel that they didn't understand but they
were passionately convinced was right anyway. I've seen this kind of
thing a lot, an apparently arbitrary split in the opinions of
intelligent people. I'm sure most people on this list have had similar
experiences. Two groups of intelligent competent people will have
totally and sometimes violently opposing views on something relatively
trivial seeming to an outsider or newcomer.

I eventually tracked this particular one down. It was a difference in
a paradigm most people didn't even know they were using. Half the
people thought the joystick was connected to the image and they wanted
the image to move with the joystick. The other half thought the
joystick was connected to the window and moving the joysick should
move the window over the image. This group expected the image to move
counter to the direction of joystick movement.

The people most passionate about their feelings about the joystick
motion (i.e. the most biased) were the ones that least understood the
paradigms (derived from something in their personal experience) that
were forcing their perceptions of correct and incorrect joystick
behavior. In reality there was no right answer. Both paradigms were
useful once the user understood them. In fact once I recast the
question to include the underlying paradigms most people got bored and
no longer had strong opinions. In other words when I asked if the
joystick should be connected to the window and move it over the image,
or connected to the image and move it under the window the answer was
generally some form of "whatever."

We don't need to start any silly flame wars here on the superiority of
one computer or development system over another (other than to note
that the single most likely explanation is that Bill Gates is in fact
the devil incarnate) but we do need to be sensitive to how our history
of audible experience colors our current perception if we expect to be
able to trust our own experience and use it to teach us something
basic about the general questions of human perception of music.

And we do need to be sure that any cross platform software we develop
should work especially well on the clearly superior Macintosh OS-X
platform used by all right thinking people. ;-)

Pax vobiscum

Barbershop Steve

At 12:16 PM 6/19/2008, you wrote:
>On Thu, Jun 19, 2008 at 12:55 PM, Carl Lumma <carl@...> wrote:
>> I'm not a Microsoft developer or even shareholder any more, but
>> their technology is superior in my opinion.
>
>Speaking of subjective experience I find that most people mean "It
>does what I expect" when they say something is superior even though
>their expectations are heavily colored by experience.

In this their technology allows one to:
1. Code in any of 30+ languages against the same object library.
I can't say the object library is any good, but I can't name one
that is, either.
2. Use fully vectorized, 3D-accelerated widgets, and control layout
using human-readable XML.
3. Compile to native code on three platforms.
4. Run it in a browser or on the desktop.

Name another API that offers this and I'll bake you a cake.
Nobody else does #1. Flash is closest on #2 as it's vector-based,
but it wasn't meant to be a GUI kit, isn't 3D that I know of,
and is a proprietary binary format. AIR is the desperate
response from Adobe.

>We don't need to start any silly flame wars here on the superiority of
>one computer or development system over another (other than to note
>that the single most likely explanation is that Bill Gates is in fact
>the devil incarnate)

Steve Jobs is far worse, yet Gates is deionized while Jobs is
lionized. I work for him and I'm saying that.

>And we do need to be sure that any cross platform software we develop
>should work especially well on the clearly superior Macintosh OS-X
>platform used by all right thinking people. ;-)

If I worked for Microsoft I could use a Mac at work. I can't
use Windows where I work now. The simple truth of the matter
is that the Macintosh Intel/OSX platform is better if you want
to run unix software, and better for average users. For power
users Windows XP is a superior OS. Most of its interactions
are far superior, from mouse acceleration on up, IF you know
how to configure it. If you're an average user the configuration
burden isn't worth it, and I recommend a Mac. I've worked on
both platforms my whole life and I can demonstrate that the
optimal usage patterns on Windows are superior to the optimal
ones on OS X.

-Carl

1. Code in any of 30+ languages against the same object library.

I can't say the object library is any good, but I can't name one

that is, either.

2. Use fully vectorized, 3D-accelerated widgets, and control layout

using human-readable XML.

3. Compile to native code on three platforms.

4. Run it in a browser or on the desktop.

   In fact, JAVA does virtually all of these except using XML for the control
layout (as WPF does).  Extensions to Java can allow vector graphics ala Flash,
plus the fact that Java can produce HTML in a browser and pass parameters to
Flash through that HTML for integration.  And, of course, JAVA Applets can run in
a browser...and even do virtually anything regular JAVA can with the help of certificates.
   Also, JAVA runs under virtually all operating systems, while systems to run .NET under
alternative operating systems are often far from flawless.  And while .NET may be faster / native code...most of what JAVA and .NET are used for, IE business programming, is mostly limited by the speed of the user and network, not the processor.  One exception would be server software programming, where Visual C++.Net gains a valuable speed advantage.
----Steve Jobs is far worse, yet Gates is deionized while Jobs is

lionized. I work for him and I'm saying that.
   I will agree, at least in part.  Jobs is the equivalent of an avant-garde artist who has have things his way...hence his arrogance with keeping the price of the original Macintosh far too high as Microsoft was taking over, still assuming all the real artists/multimedia/music developers would 100% HAVE to stick to the Mac due to its superior interface alone...and also being picky about supporting 3rd party vendor support.
   Microsoft may have a monopoly, and they are much less efficient than they should be, but at least they are making some efforts to satisfy customer business demands, instead of trying to guess their demands before the customer says what they are.  Such guessing may be responsible for Apple's fantastic user interface, but also for their lack of ability to integrate with what productivity software and systems other software vendors create, thus making it a very inaccessible platform who those who want to do power-user type work across systems or need compatibility with most current software.

--- On Thu, 6/19/08, Carl Lumma <carl@...> wrote:
From: Carl Lumma <carl@...>
Subject: Re: [MMM] Re: Composition Software...?
To: MakeMicroMusic@yahoogroups.com
Date: Thursday, June 19, 2008, 12:49 PM

At 12:16 PM 6/19/2008, you wrote:

>On Thu, Jun 19, 2008 at 12:55 PM, Carl Lumma <carl@...> wrote:

>> I'm not a Microsoft developer or even shareholder any more, but

>> their technology is superior in my opinion.

>

>Speaking of subjective experience I find that most people mean "It

>does what I expect" when they say something is superior even though

>their expectations are heavily colored by experience.

In this their technology allows one to:

1. Code in any of 30+ languages against the same object library.

I can't say the object library is any good, but I can't name one

that is, either.

2. Use fully vectorized, 3D-accelerated widgets, and control layout

using human-readable XML.

3. Compile to native code on three platforms.

4. Run it in a browser or on the desktop.

Name another API that offers this and I'll bake you a cake.

Nobody else does #1. Flash is closest on #2 as it's vector-based,

but it wasn't meant to be a GUI kit, isn't 3D that I know of,

and is a proprietary binary format. AIR is the desperate

response from Adobe.

>We don't need to start any silly flame wars here on the superiority of

>one computer or development system over another (other than to note

>that the single most likely explanation is that Bill Gates is in fact

>the devil incarnate)

Steve Jobs is far worse, yet Gates is deionized while Jobs is

lionized. I work for him and I'm saying that.

>And we do need to be sure that any cross platform software we develop

>should work especially well on the clearly superior Macintosh OS-X

>platform used by all right thinking people. ;-)

If I worked for Microsoft I could use a Mac at work. I can't

use Windows where I work now. The simple truth of the matter

is that the Macintosh Intel/OSX platform is better if you want

to run unix software, and better for average users. For power

users Windows XP is a superior OS. Most of its interactions

are far superior, from mouse acceleration on up, IF you know

how to configure it. If you're an average user the configuration

burden isn't worth it, and I recommend a Mac. I've worked on

both platforms my whole life and I can demonstrate that the

optimal usage patterns on Windows are superior to the optimal

ones on OS X.

-Carl

[Non-text portions of this message have been removed]

Hm, a bunch of this discussion is starting to get pretty far away from the
topic of our list and looks more like a general discussion of programming
and platforms. So perhaps it could be taken somewhere else?

Thanks,
Rick

At 01:14 PM 6/19/2008, you wrote:
>>1. Code in any of 30+ languages against the same object library.
>>I can't say the object library is any good, but I can't name one
>>that is, either.
>>2. Use fully vectorized, 3D-accelerated widgets, and control layout
>>using human-readable XML.
>>3. Compile to native code on three platforms.
>>4. Run it in a browser or on the desktop.
>
> In fact, JAVA does virtually all of these except using XML for
>the control layout (as WPF does).

Java allows you to choose from 30 languages? It has 3D widgets?

>Extensions to Java can allow vector graphics ala Flash,

How easy would it be to build a completely vector UI, from the
fonts on up?

>plus the fact that Java can produce HTML in a browser and pass
>parameters to Flash through that HTML for integration.

Oh yeah, that sounds like a winning control flow.

>And, of course, JAVA Applets can run in a browser...

Silverlight runs in a browser. Java applets are practically
(actually?) deprecated, thanks in part to JVM standardization
failure (only part of which is MS's fault), and in part to
the fact that they run like crap.

> Also, JAVA runs under virtually all operating systems, while
>systems to run .NET under alternative operating systems are often
>far from flawless.

It's not mature yet, I admit.

>And while .NET may be faster / native code...

Yup. Of course if you really like Java, you can use it with
.net (in the form of J#).

>most of what JAVA and .NET are used
>for, IE business programming, is mostly limited by the speed of the
>user and network, not the processor.

Java is used mostly for server-side programming. .Net is used
also for consumer desktop applications.

> Microsoft may have a monopoly,

In case you hadn't noticed, Apple also have a monopoly.
One, it has been argued, more restrictive than anything
Microsoft ever had.

>and they are much less efficient than they should be,

You can say that again. They're basically doomed. They have
an unfixable cultural problem outside the company, and a
nearly unfixable operations problem inside (as long as Balmer
is at the helm, for sure), and the classic "aging empire"
problem with their codebase.

>Such guessing may be
>responsible for Apple's fantastic user interface,

The secret is making products for yourself, not for the "customer".
That's why Apple succeeds while Microsoft is floundering.

-Carl

At 01:26 PM 6/19/2008, you wrote:
>Hm, a bunch of this discussion is starting to get pretty far away from the
>topic of our list and looks more like a general discussion of programming
>and platforms. So perhaps it could be taken somewhere else?
>
>Thanks,
> Rick

Sorry, will do. Please: anyone involved feel free to reply to
me offlist, or on metatuning. -Carl

On Thu, Jun 19, 2008 at 12:45 PM, Carl Lumma <carl@...> wrote:
> Hi Hans!
>
>>> Meh. I'd rather have a piano roll that alpha blends to staff
>>> notation (an idea I've been toying with for some years).
>>
>>What do you mean exactly here? A piano roll with a staff above it,
>>and edit operations at any place propagated to the other in realtime?
>
> The key ingredients in my scheme are:
>
> * Not WSYWIG, in the sense that the onscreen display is
> optimized for onscreen editing. There's a separate step
> for page layout if you want to print. This is a core
> part of the philosophy.

I like this idea. So you're basically saying we'd have two views - one
that's just a linear timeline in staff notation a la Cakewalk's scheme
and one that's closer to the Finale/Sibelius end of things. I do like
that.

As much as I'm starting to hate Sonar (mainly for the lame skinning
they added recently), there are some things from their software that I
think are a really good idea. Namely that they have it modularized
like this:

- In memory there is a master list of events that occur at a specific
time. These events can be MIDI, they can be audio samples, they can be
automation data, they can be "text."
- There are different "views" that simply process and display this
list. There is a piano roll view, there is a staff view, there is an
"event list" view, there is the track view.
- All the views really are are just different GUI's that deal with the
same data. The staff view ignores audio data and automation data and
focuses only on MIDI note messages.
- Then, from this master list of events, when playing, we can output
as MIDI, we can output as OSC, we can output however we want. They
store the event data for a "note," for example, as having a velocity
and a duration (i.e. they don't have two separate events stored in
memory for a note off and a note on). This will make it incredibly
easy to expand to other formats in the future.

> * So the editing display shows a linear timeline, not a
> "page". The key is that 'now' is always at the same
> spot. We don't move a caret across the screen and then
> update, we *smoothly scroll* content under a fixed caret.
> Very important for *fast* mouse editing, and works well
> with a 'object properties inspector' paradigm.
>
> * The timeline can show a regular staff with notes, or
> a piano roll, with a slider for alpha blending between
> them. Music notation is equivalent to a piano roll,
> except: 1. duration is expressed with flags and notehead
> coloring instead of bar length 2. vertical position
> increments aren't all the same.
> If the slider is moved all the way to the piano roll
> position (so the traditional notes are invisible) the
> only difference between what you'll see and a traditional
> piano roll will be #2 (the unequal vertical increment).

How would we line them up horizontally? Would we never stretch the
staff notation? That might look weird. Or maybe we could have it so
that for a bar if there are 32nd notes crammed in after a bar of just
quarter notes, we could just stretch that bar of the piano roll.

> * The point of the piano roll view is: 1. to assist people
> who aren't good with regular music notation and 2. to
> assist people who aren't good with the advanced
> polyrhythmic notations (like Henry Cowell's) I intend to
> implement, which would be everybody.

> * As already mentioned, when creating a new score we are
> prompted to specify the vertical increments (ie < L L s >
> etc.), accidental font file, MIDI offsets into the
> accidentals, and notehead shape and flag meanings.
> We'll ship with a few presets, including one that gives
> common practice notation, and users can save their own.

Number of lines as well. This all sounds good. I can tell you've been
thinking about this for a while. :P

-Mike

Mike wrote:
>>
>>>> Meh. I'd rather have a piano roll that alpha blends to staff
>>>> notation (an idea I've been toying with for some years).
>>>
>>>What do you mean exactly here? A piano roll with a staff above it,

No, over top it.

>> The key ingredients in my scheme are:
>>
>> * Not WSYWIG, in the sense that the onscreen display is
>> optimized for onscreen editing. There's a separate step
>> for page layout if you want to print. This is a core
>> part of the philosophy.
>
>I like this idea. So you're basically saying we'd have two views - one
>that's just a linear timeline in staff notation a la Cakewalk's scheme
>and one that's closer to the Finale/Sibelius end of things. I do like
>that.

Lots of score editors have these two views, but none scroll
the content in linear mode. I wouldn't even waste 1.0 bandwidth
doing an editable "paper" view. I'd just put a some page break
and basic layout controls in a print preview window and call
it done.

>As much as I'm starting to hate Sonar (mainly for the lame skinning
>they added recently), there are some things from their software that I
>think are a really good idea. Namely that they have it modularized
>like this:
>
>- In memory there is a master list of events that occur at a specific
>time. These events can be MIDI, they can be audio samples, they can be
>automation data, they can be "text."

Yep, that's how DAWs work. There isn't a single DAW on Earth
with a good notation feature, and there's a reason for that.
If you put audio or anything fancy in your thing, I'm offering
odds it'll suck.

>> * The timeline can show a regular staff with notes, or
>> a piano roll, with a slider for alpha blending between
>> them. Music notation is equivalent to a piano roll,
>> except: 1. duration is expressed with flags and notehead
>> coloring instead of bar length 2. vertical position
>> increments aren't all the same.
>> If the slider is moved all the way to the piano roll
>> position (so the traditional notes are invisible) the
>> only difference between what you'll see and a traditional
>> piano roll will be #2 (the unequal vertical increment).
>
>How would we line them up horizontally? Would we never stretch the
>staff notation?

Right!

>That might look weird.

http://www.musanim.com

Just imagine the score on top. It's a no-brainer.

>> * The point of the piano roll view is: 1. to assist people
>> who aren't good with regular music notation and 2. to
>> assist people who aren't good with the advanced
>> polyrhythmic notations (like Henry Cowell's) I intend to
>> implement, which would be everybody.
>
>> * As already mentioned, when creating a new score we are
>> prompted to specify the vertical increments (ie < L L s >
>> etc.), accidental font file, MIDI offsets into the
>> accidentals, and notehead shape and flag meanings.
>> We'll ship with a few presets, including one that gives
>> common practice notation, and users can save their own.
>
>Number of lines as well.

Yes.

>This all sounds good. I can tell you've been
>thinking about this for a while. :P

Yup!

-Carl

>>Hell, I'll package one with it if it's that much easier to work with
>>than MIDI. MIDI has some limitations in my mind that seem
>>insurmountable, but maybe you have some insight on how to get past
>>them. Namely, it seems like if I'm going to want to play a four-note
>>chord in 31tet, for example, I'm going to use four MIDI channels,
>
> Why??

Because if the synth DOESN'T support MTS, that's what we're left to
work with. 4 channels and pitch bending.

>>unless the synth supports MTS, which I can't find any that do except
>>NI's FM8.
>
> Hare are some MTS-compatible synths:
>
> Alphakanal Automat
> AnaMark softsynth
> Audio Compositor
> Big Tick: Angelina, Rainbow and Rhino
> Bitheadz: Unity
> Cakewalk: Dimension Pro, Rapture, Z3TA+
> Camel Audio: Cameleon5000
> ChucK
> E-mu: Morpheus, Proteus series
> Ensoniq: EPS/EPS16/ASR10, VFX
> Fluidsynth
> Korg: M1, M1R, X5DR, OASYS PCI soundcard
> LinPlug: Albino 2, Alpha 2, CronoX, Octopus, Organ 3 and Sophistry
> Native Instruments: FM7, Pro-52/Pro-53
> Pure Data
> Roland: GS & JV/XP families, Fantom-X6/X7/X8, V-Synth 2.0, VSC, SC-8850
> Synapse Audio: Orion Pro softsynth
> Timidity
> Tobybear Helios
> Turtle Beach: Multisound, Monterey, Maui, Tropez, Rio
> VAZ Plus
> VirSyn Cube, Cantor and TERA 2
> Yamaha DX7II/TX802SY77/TG77/SY99/VL-1/VL-7/TX81Z/DX11/DX27/DX100/V50
> Yamaha XG family
> Yamaha VL70m
> Wusik Wusikstation v2
> Zebra 2.0
>
> Here are some .scl-compatible synths:
>
> AlsaModularSynth by Matthias Nagorni
> Astralis v2 by Homegrown Sounds
> Blue by Steven Yi and Tracker Sound Object
> Cantor by VirSyn Software Synthesizer
> Dimension Pro 1.2, Rapture, and Z3TA+ by Cakewalk
> Lounge Lizard by Applied Acoustics
> Max Magic Microtuner by Victor Cerullo
> Nord Modular G2 via G2ools converter by Ian Sayer
> Oblivion VSTi for Windows by David R. Frost
> Pianoteq by MODARTT
> Scala Microtuners for Kontakt and SynthEdit by Robert Strauss
> TERA 2.0 by VirSyn
> Virtual Pipe Organs by Garritan
> ZynAddSubFX software synthesizer by Paul Nasca

Cool. So if we package Timidity with it, then nobody would have any
kind of a problem ever.

>>Even Scala has the option
>>of doing it with channel swapping and pitch-bending, which is fine.
>
> It's great for quick-and-dirty, but it's a hack and not suitable
> for serious production. I'd suggest we not implement it.

One of my main objectives with this is that people shouldn't be
limited to the software you see up there. It's quite an impressive
range, but certainly not complete by any means. And users are likely
going to have their own synths that they're going to want to use, of
which a few very significant ones are left out of that list above. Not
to mention the people that use an external MIDI rack or keyboard for
their sounds, of which support is even more rare. So having a little
switch in the preferences that lets you use the pitch bending and
channel swapping to work with those resources greatly expands the
usefulness of the program.

I don't suggest we implement that as fundamental on the application
layer - rather that it is one possible way to output the note data
from the events list. It could also be output via MTS, or by working
something out with a Scala file, or by using OSC.

And the very least, it might be useful within the context of an
"export to MIDI," so that people can at least go to their favorite
synths when they're done. But if we're going to implement that it
would likely not be too much of a stretch to just have it work in real
time as well.

>>What do you mean via scala files? How would that let us get more than
>>12 notes per octave when dealing with MIDI? We'd still need pitch
>>bends, right?
>
> If a synth supports them, you don't need any pitch bends, no.
> You just send plain MIDI and the synth knows how to tune it. It
> can be a pain when doing channel splitting, but other than that
> it's a great way to work.

So you're not limited to 12-note subsets of various scales? Does it
retune the note spectrum from 0-127 as MTS does?

Also, what do you mean by channel splitting?

> .net isn't a VM, it's native binaries at the end of the day.
> And I don't see why you're equating our performance with
> softsynth load. Whatever VST or MIDI synths we'll use will
> be binaries. The OS takes care of the multitasking, not us.
> And unless we inspect the system and make a guess, we won't
> be able to know how many instances of which synth the user
> will be able to run. But your idea of picking a lightweight
> one and bundling it is good. Except instead of a bundle you
> could just put it in the system requirements and that saves
> you any licensing trouble.

Yeah. Or at least a link on the sourceforge page.

Part of my objective with this project is that I want it to tie into
the stuff we were discussing about the workshop. I want it to be
something I can point at the average musician on a mac or a PC or even
linux and say "check this program out" and they can roll with it, so
that more people can easily compose microtonal music. And I want them
to be able to roll with it without even having to know anything about
VST or MTS or Scala at all if they don't already know, so that they
can just use their system's native GM synth or one that we would
bundle. Of course, for power users that know more about it, there
should be a way to configure all of the behind-the-scenes stuff as
well.

I guess rather than equate the performance with softsynth load, it
would be better to equate the performance with audio load. If we're
going to do it in C#, and if the performance hit means that we will
NEVER be able to add in an audio section to the sequencer if we want
to, then I think we'd best work from a lower level paradigm. On the
other hand, it seems you're saying that working in .NET won't be too
much slower, and if that's the case, then I'd rather work in C# than
C++ anyway, as it's just easier.

>>> Python can be compiled into a .net project like any other .net
>>> language. I'm not saying you should necessarily use .net, but
>>> it's definitely worth a look. Otherwise it's C++ and gtk or
>>> something like that. I dunno if XUL is up to the task. I guess
>>> it's made strides lately.
>>
>>If the performance hit by using .NET wouldn't be too severe,
>>I'd do it in that. The jury seems to be still out on that
>>one though.
>
> It does? All windows development has been .net for years.
> Even if you write C++ I believe it's compiled to .net CLR
> first.

Hm. I'll have to do more research. Now visions of C# run through my head.

-Mike

>>>>What do you mean exactly here? A piano roll with a staff above it,
>
> No, over top it.

Ah yeah, I meant "above it" in like a 3-dimensional sense. Seems like
the vertical disparity could get pretty extreme between the two
though, especially if you're using something like 31-tet where sharps
and flats and half sharps and half flats can all be on the same staff
notation line. If you don't mind that though, it seems like it'd be a
neat idea.

Another way to counter that would be to have unequal vertical piano
roll spacing, so that in 12tet, for example, the "D" space on the
staff is really two spaces half the size on the piano roll for D and
D#, and then E would be twice as large as it has no sharp. Then we
could alpha blend them perfectly. Maybe we'd have to increase the zoom
on the staff.

For higher-order temperaments such as 31 or 72 equal, this method
might not work, although it might be very possible to find a setting
in which the piano roll is zoomed in enough to see and the staff is
zoomed in just enough that it's still somewhat readable. I guess if
you're used to working within a piano roll kind of environment, you'll
be used to zoom like that anyway, so it might work.

>>> The key ingredients in my scheme are:
>>>
>>> * Not WSYWIG, in the sense that the onscreen display is
>>> optimized for onscreen editing. There's a separate step
>>> for page layout if you want to print. This is a core
>>> part of the philosophy.
>>
>>I like this idea. So you're basically saying we'd have two views - one
>>that's just a linear timeline in staff notation a la Cakewalk's scheme
>>and one that's closer to the Finale/Sibelius end of things. I do like
>>that.
>
> Lots of score editors have these two views, but none scroll
> the content in linear mode. I wouldn't even waste 1.0 bandwidth
> doing an editable "paper" view. I'd just put a some page break
> and basic layout controls in a print preview window and call
> it done.
>
>>As much as I'm starting to hate Sonar (mainly for the lame skinning
>>they added recently), there are some things from their software that I
>>think are a really good idea. Namely that they have it modularized
>>like this:
>>
>>- In memory there is a master list of events that occur at a specific
>>time. These events can be MIDI, they can be audio samples, they can be
>>automation data, they can be "text."
>
> Yep, that's how DAWs work. There isn't a single DAW on Earth
> with a good notation feature, and there's a reason for that.
> If you put audio or anything fancy in your thing, I'm offering
> odds it'll suck.

Well I'm certainly not stuck on audio just yet. I was referring mainly
to planning ahead so that the functionality wouldn't be a nightmare to
add in the future.

>>> * The timeline can show a regular staff with notes, or
>>> a piano roll, with a slider for alpha blending between
>>> them. Music notation is equivalent to a piano roll,
>>> except: 1. duration is expressed with flags and notehead
>>> coloring instead of bar length 2. vertical position
>>> increments aren't all the same.
>>> If the slider is moved all the way to the piano roll
>>> position (so the traditional notes are invisible) the
>>> only difference between what you'll see and a traditional
>>> piano roll will be #2 (the unequal vertical increment).
>>
>>How would we line them up horizontally? Would we never stretch the
>>staff notation?
>
> Right!
>
>>That might look weird.
>
> http://www.musanim.com
>
> Just imagine the score on top. It's a no-brainer.

Well, I get what you're saying, but it seems like it would be more
useful for a piano roll than for a staff if that's the case. Being
able to stretch the width of a measure is pretty vital for scoring. A
good portion of the time that I spend while scoring anything is making
sure it's readable and preparing it as such by moving measure lengths
around and the like. If your goal for the staff is to simply be an
alpha overlay on top of the piano roll and so that the piano roll is
the main intent of focus with the staff as kind of an added bit of
information, then that would work. But if you want the staff in this
view to be readable in its own right then I would suggest at least
having the option of stretching the measures in the staff if the user
wants. Then again, it's your idea, so whatever you'd like.

-Mike

--- In MakeMicroMusic@yahoogroups.com, "Mike Battaglia"
<battaglia01@...> wrote:
>
> >>Hell, I'll package one with it if it's that much easier to
> >>work with than MIDI. MIDI has some limitations in my mind
> >>that seem insurmountable, but maybe you have some insight
> >>on how to get past them. Namely, it seems like if I'm going
> >>to want to play a four-note chord in 31tet, for example,
> >>I'm going to use four MIDI channels,
> >
> > Why??
>
> Because if the synth DOESN'T support MTS, that's what we're left to
> work with. 4 channels and pitch bending.

Not if it support scala files, like (wait for it) Pianoteq!

> Cool. So if we package Timidity with it, then nobody would have any
> kind of a problem ever.

Timidity is a midi renderer. It doesn't work in realtime.
sfz comes to mind as a good candidate, but I forget the
details and I don't know what happened to it after the
Cakewalk acquisition.

> > It's great for quick-and-dirty, but it's a hack and not suitable
> > for serious production. I'd suggest we not implement it.
>
> One of my main objectives with this is that people shouldn't be
> limited to the software you see up there.

Thems the breaks.

> It's quite an impressive
> range,

It's not complete. It's just what Scala happens to support
for its MTS dumps.

> Not
> to mention the people that use an external MIDI rack or keyboard for
> their sounds, of which support is even more rare.

Most Emu, Yamaha, Roland, and all Kurweil racks support MTS.
Probably the vast majority of racks currently in service,
though I believe Roland is limited to 12 notes/octave.

> So having a little
> switch in the preferences that lets you use the pitch bending and
> channel swapping to work with those resources greatly expands the
> usefulness of the program.

Maybe Manuel would give you his Ada code, which you could
translate over.

> >>What do you mean via scala files? How would that let us get
> >>more than 12 notes per octave when dealing with MIDI? We'd
> >>still need pitch bends, right?
> >
> > If a synth supports them, you don't need any pitch bends, no.
> > You just send plain MIDI and the synth knows how to tune it. It
> > can be a pain when doing channel splitting, but other than that
> > it's a great way to work.
>
> So you're not limited to 12-note subsets of various scales?

No !

> Does it retune the note spectrum from 0-127 as MTS does?

It can, and it does one better: if you specify fewer notes,
it periodically tiles the MIDI note spectrum for you.

> Also, what do you mean by channel splitting?

Splitting a large scale across multiple MIDI channels. Then
you have to have a different Scala file for each channel,
and load each in a different instance of the synth, and
then make sure you've sent the right channels to the right
instances.

> Part of my objective with this project is that I want it to tie into
> the stuff we were discussing about the workshop.

It'll probably take at least a year to get an alpha. How
many large-scale projects have you undertaken before?

> I guess rather than equate the performance with softsynth load, it
> would be better to equate the performance with audio load.

???

> If we're
> going to do it in C#, and if the performance hit means that we will
> NEVER be able to add in an audio section to the sequencer if we want
> to,

Audio playback will always be easier than synthesis, so I'm not
sure what you mean.

> >>> Python can be compiled into a .net project like any other .net
> >>> language. I'm not saying you should necessarily use .net, but
> >>> it's definitely worth a look. Otherwise it's C++ and gtk or
> >>> something like that. I dunno if XUL is up to the task. I guess
> >>> it's made strides lately.
> >>
> >>If the performance hit by using .NET wouldn't be too severe,

There's no performance hit with .Net vs. C++.

-Carl

> Another way to counter that would be to have unequal vertical piano
> roll spacing, so that in 12tet, for example, the "D" space on the
> staff is really two spaces half the size on the piano roll for D
> and D#, and then E would be twice as large as it has no sharp. Then
> we could alpha blend them perfectly. Maybe we'd have to increase
> the zoom on the staff.

The roll spacing could be a subdivision of the staff spacing, yes.
Or you could just not show that information on the roll (it would
be available in the note-bar properties inspector).

> >>How would we line them up horizontally? Would we never stretch
> >>the staff notation?
> >
> > Right!
> >
> >>That might look weird.
> >
> > http://www.musanim.com
> >
> > Just imagine the score on top. It's a no-brainer.
>
> Well, I get what you're saying, but it seems like it would be more
> useful for a piano roll than for a staff if that's the case. Being
> able to stretch the width of a measure is pretty vital for scoring.

I thought you meant stretch the note heads or something.
No, that's a good point. But I usually compose with fixed
measure widths, myself. In fact, I hate the way Sibelius
does measure widths automatically.

> A good portion of the time that I spend while scoring anything
> is making sure it's readable and preparing it as such by moving
> measure lengths around and the like.

I'm viewing this mainly as a computer-based composing tool,
with the idea of humans playing the output as a secondary
goal. Since the number of musicians who are able let alone
willing to read the custom notations I'd create is exactly
zero... However that might change if there was compelling
music written in them. And for that to happen, we need a
compelling composition interface.

> But if you want the staff in this
> view to be readable in its own right then I would suggest at
> least having the option of stretching the measures in the staff
> if the user wants. Then again, it's your idea, so whatever
> you'd like.

It's based on a prototype score editor called Rational
Composer, which was written in Pascal by my first
microtonal music teacher, Denny Genovese.

-Carl

Carl, Carl, Carl ...

I guess my humor is too subtle. Picking on Bill Gates and advocating
Mac OSX was intended to be a joke that exemplified my case. It looks
like you missed my entire point and only responded by taking seriously
the apparently too weak attempt at humor at the end. I'm sorry I did
it so poorly because I was hoping you would enjoy the joke.

I don't know if any platform is superior. I can't honestly advocate
one over another. I'm a professional. I use what tools I need to and
try (not very hard as this thread proves) to stay out of silly
religious wars with narrow minded zealots. I have preferences but
unlike you I have no idea if my experience is general enough to push
my preferences on anyone else. I can't say I have worked on any of
these platforms all my life because except for UNIX itself they didn't
exist for the first third of my career. On the other hand I have used
them all (including UNIX) basically as long as they have existed. The
more experience I gain the more I realize that there is no single best
anything. Choice of tools ALWAYS depends on the application, size of
project and the skill sets of the developers.

All I know is that I can find competent people to articulately and
compellingly demonstrate why their own favorite software platform,
tuning strategy, sequencer, soft synth etc. is superior. Your
arguments are an example. They sure sound good but anyone can make a
list of what their favorite platform supports. In my experience all
platforms have their strengths. You list strengths that you think are
compelling. Frankly it looks like the resume of a web developer to me.
I'm an embedded software kind of guy who is as likely as not to write
my own lightweight OS to go into my analytical instrument or robot or
whatever. Nothing on your list appeals to me. We sit on opposite ends
of the space of code developers. Most of what I expect out a
development platform you probably consider unimportant.

The wise person is the one who realizes that almost anything they say
and believe about a complex subject is tinged by layer after layer of
bias, most of which is unwitting. No one is intentionally biased. Most
people try to be honest and objective. Most people think they ARE
honest and objective. Most people are neither.

--- In MakeMicroMusic@yahoogroups.com, Carl Lumma <carl@...> wrote:
> something like that. I dunno if XUL is up to the task. I guess
> it's made strides lately.

XUL from what I was able to gather is almost purely an interface
builder for Mozilla/Firefox. If you're just looking for a
Mozilla-hosted front end for something else that does the dirty work
of synthesis and sequencing, then it might be somewhat useful.
However it appears to be primarily intended for forms (ie. enter your
name here, your address there, etc.).

>> something like that. I dunno if XUL is up to the task. I guess
>> it's made strides lately.
>
>XUL from what I was able to gather is almost purely an interface
>builder for Mozilla/Firefox. If you're just looking for a
>Mozilla-hosted front end for something else that does the dirty work
>of synthesis and sequencing, then it might be somewhat useful.
>However it appears to be primarily intended for forms (ie. enter your
>name here, your address there, etc.).

XUL is a toolkit designed by the Mozilla project, much
as gtk was designed for the gimp project. Either can be
used for any application. Judging by Firefox 3, XUL has
made great strides recently in what it can do. I don't
know if it's up to a score editor but it's worth looking at.

-Carl

very nice pieces on that page!

/^_,',',',_ //^ /Kraig Grady_ ^_,',',',_
Mesotonal Music from:
_'''''''_ ^North/Western Hemisphere: North American Embassy of Anaphoria Island <http://anaphoria.com/>

_'''''''_ ^South/Eastern Hemisphere:
Austronesian Outpost of Anaphoria <http://anaphoriasouth.blogspot.com/>

',',',',',',',',',',',',',',',',',',',',',',',',',',',',',

Prent Rodgers wrote:
>
>
>
> A recent experiment is a realization of John Cage's In a Landscape in
> 53-TET. You can hear it here: http://bumpermusic.blogspot.com > <http://bumpermusic.blogspot.com> .
>
> Prent Rodgers
>
>

Prent Rodgers wrote:
> I use Csound and a front end I wrote 15 years ago in Turbo Pascal.
> Pascal is a little dated, but I still have customers running 40 year
> old FORTRAN and 25 year old COBOL applications, so I feel like I'm in
> good company. > > Csound has a very stiff learning curve, as others have discovered. I
> climbed it ten years ago, and have enjoyed it in the limited way I use
> it ever since. If anyone tries it for microtonal work, I'd be glad to
> help with that aspect of it. It does allow for any pitch and any
> instrument with no practical limitations except CPU speed and storage.
> I use it as a sampler, and then control as many aspects as possible to
> humanize the sound. Fake but accurate is my motto. I generally have
> about 10 different instruments with 53 or fewer pitches per octave,
> and it fits well in that range.
> > A recent experiment is a realization of John Cage's In a Landscape in
> 53-TET. You can hear it here: http://bumpermusic.blogspot.com . > > Prent Rodgers

One thing that's great about Csound is how easy it is to change the tuning (with tables, macros, etc.), and how many different options there are for specifying the pitch. I think it's a good option for anyone who does a lot of work with JI (which can be a mess to work with in MIDI), and your music is certainly a good illustration of that. (What other system can handle a 15-limit tonality diamond with such ease?)

--- In MakeMicroMusic@yahoogroups.com, Kraig Grady <kraiggrady@...>
wrote:
>
> very nice pieces on that page!
>
>
> /^_,',',',_ //^ /Kraig Grady_ ^_,',',',_
> Mesotonal Music from:
> _'''''''_ ^North/Western Hemisphere:
> North American Embassy of Anaphoria Island <http://anaphoria.com/>
>
> _'''''''_ ^South/Eastern Hemisphere:
> Austronesian Outpost of Anaphoria
<http://anaphoriasouth.blogspot.com/>
>
> ',',',',',',',',',',',',',',',',',',',',',',',',',',',',',
>
>
>
>
> Prent Rodgers wrote:
> >
> >
> >
> > A recent experiment is a realization of John Cage's In a
Landscape in
> > 53-TET. You can hear it here: http://bumpermusic.blogspot.com
> > <http://bumpermusic.blogspot.com> .
> >
> > Prent Rodgers
> >
> >
>
From Robert. The genre of music called Exotica and made famous by
Martin Denny et al is still alive and kicking with the sounds of
Prent Rodgers and his unique style and superb craftmanship.

>> Because if the synth DOESN'T support MTS, that's what we're left to
>> work with. 4 channels and pitch bending.
>
> Not if it support scala files, like (wait for it) Pianoteq!

Indeed. You're in love with pianoteq, aren't you?

> Timidity is a midi renderer. It doesn't work in realtime.
> sfz comes to mind as a good candidate, but I forget the
> details and I don't know what happened to it after the
> Cakewalk acquisition.

Damn. I'm sure something's out there though.

>> One of my main objectives with this is that people shouldn't be
>> limited to the software you see up there.
>
> Thems the breaks.

The who?

>> It's quite an impressive
>> range,
>
> It's not complete. It's just what Scala happens to support
> for its MTS dumps.
>
>> Not
>> to mention the people that use an external MIDI rack or keyboard for
>> their sounds, of which support is even more rare.
>
> Most Emu, Yamaha, Roland, and all Kurweil racks support MTS.
> Probably the vast majority of racks currently in service,
> though I believe Roland is limited to 12 notes/octave.

Off the top of my head, I know people who use a Yamaha Motif classic,
a Korg Triton, and a Nord Stage for a main MIDI hookup (except for the
Nord Stage, which is used for E.P.'s only). None of those keyboards
support MTS.

>> So having a little
>> switch in the preferences that lets you use the pitch bending and
>> channel swapping to work with those resources greatly expands the
>> usefulness of the program.
>
> Maybe Manuel would give you his Ada code, which you could
> translate over.

>> >>What do you mean via scala files? How would that let us get
>> >>more than 12 notes per octave when dealing with MIDI? We'd
>> >>still need pitch bends, right?
>> >
>> > If a synth supports them, you don't need any pitch bends, no.
>> > You just send plain MIDI and the synth knows how to tune it. It
>> > can be a pain when doing channel splitting, but other than that
>> > it's a great way to work.
>>
>> So you're not limited to 12-note subsets of various scales?
>
> No !
>
>> Does it retune the note spectrum from 0-127 as MTS does?
>
> It can, and it does one better: if you specify fewer notes,
> it periodically tiles the MIDI note spectrum for you.
>
>> Also, what do you mean by channel splitting?
>
> Splitting a large scale across multiple MIDI channels. Then
> you have to have a different Scala file for each channel,
> and load each in a different instance of the synth, and
> then make sure you've sent the right channels to the right
> instances.
>
>> Part of my objective with this project is that I want it to tie into
>> the stuff we were discussing about the workshop.
>
> It'll probably take at least a year to get an alpha. How
> many large-scale projects have you undertaken before?

I mean that I want it to be under the same principles of expanding the
reach of microtonal music and making current musicians out there more
capable of writing microtonal stuff. It would make it a lot easier to
do that if it worked without people needing to provide their own
synth. Hell, if the project gets advanced enough, maybe we'd be able
to put in

Admittedly the largest scale project I've worked on before would be
the equivalent of just the notation GUI or just the sequencer or just
the piano roll. To put them all together is a step up, so it's going
to be a learning process.

>> I guess rather than equate the performance with softsynth load, it
>> would be better to equate the performance with audio load.
>
> ???

I would like to leave the project open to be a full DAW if that's the
route it takes at a later date. If we do it in C# do you think that
the project would be even capable of handling the kind of load that a
full DAW would need, with multiple audio channels and VST effects and
such? Because if not, then

>> If we're
>> going to do it in C#, and if the performance hit means that we will
>> NEVER be able to add in an audio section to the sequencer if we want
>> to,
>
> Audio playback will always be easier than synthesis, so I'm not
> sure what you mean.

Hm. Well I'm not looking to program a synth engine, just have one that
people can use. In logic, 60 audio channels will slow you down, 60
midi channels will likely not, unless they're all using softsynths.

>> >>> Python can be compiled into a .net project like any other .net
>> >>> language. I'm not saying you should necessarily use .net, but
>> >>> it's definitely worth a look. Otherwise it's C++ and gtk or
>> >>> something like that. I dunno if XUL is up to the task. I guess
>> >>> it's made strides lately.
>> >>
>> >>If the performance hit by using .NET wouldn't be too severe,
>
> There's no performance hit with .Net vs. C++.

I'll have to take a look at C# then. Not having to worry about memory
management would be a plus, although I have never looked at it before,
so I don't know if coding in it for the first time for a project like
this would be inadvisable. Might be easy to jump over, though.

> The roll spacing could be a subdivision of the staff spacing, yes.
> Or you could just not show that information on the roll (it would
> be available in the note-bar properties inspector).

Or, we could have sharps and flats be in different colors, or just
draw a "b" or a "#" sign inside the roll or something.

I'm going to set up a sourceforge page for this, I guess, and keep
looking at what software is out there. I think that given what we've
got we should start with a sequencer engine like TSE3 - that way the
internal DAW stuff is done and we can focus entirely on the piano roll
and staff notation.
________

Here's some stuff I've found on sourceforge:

http://musickit.sourceforge.net/index.html - C/Objective-C Music/MIDI/audio API
http://guidolib.sourceforge.net/ - Music notation API/library

I think we should start with one of the sequencer API's like the TSE3
one posted above, as then the hard work of actually making the
sequencer is done, and the focus can mainly go to the notation
GUI/piano roll. If we can get just a cross-platform notation GUI done
and tie it into an already-built sequencer, that would be in and of
itself a huge step forward, and it doesn't seem to be too unattainable
of a goal. It's exciting to be around people with so many ideas - if
we can get a solid core, we'd be able to extend this as far as people
can come up with.

Also, Carl, if we do this in C# and .NET, how hard would it be to tie
THAT in with Python? Usually is a bad idea to have programmers
contributing in more than one main language, I know, but maybe at
least for plugin development or something, it would be useful
somewhere down the line, since everyone here seems to already know it.

-Mike

>>> I guess rather than equate the performance with softsynth load, it
>>> would be better to equate the performance with audio load.
>>
>> ???
>
>I would like to leave the project open to be a full DAW if that's the
>route it takes at a later date. If we do it in C# do you think that
>the project would be even capable of handling the kind of load that a
>full DAW would need, with multiple audio channels and VST effects and
>such? Because if not, then

Once again, VSTs are external binaries and their performance has
nothing to do with our code. Yes C# is plenty fast.

>>> If we're
>>> going to do it in C#, and if the performance hit means that we will
>>> NEVER be able to add in an audio section to the sequencer if we want
>>> to,
>>
>> Audio playback will always be easier than synthesis, so I'm not
>> sure what you mean.
>
>Hm. Well I'm not looking to program a synth engine, just have one that
>people can use. In logic, 60 audio channels will slow you down, 60
>midi channels will likely not, unless they're all using softsynths.

MIDI is easier than audio only if you're synthesizing outside of the
computer.

>>> >>> Python can be compiled into a .net project like any other .net
>>> >>> language. I'm not saying you should necessarily use .net, but
>>> >>> it's definitely worth a look. Otherwise it's C++ and gtk or
>>> >>> something like that. I dunno if XUL is up to the task. I guess
>>> >>> it's made strides lately.
>>> >>
>>> >>If the performance hit by using .NET wouldn't be too severe,
>>
>> There's no performance hit with .Net vs. C++.
>
>I'll have to take a look at C# then. Not having to worry about memory
>management would be a plus, although I have never looked at it before,
>so I don't know if coding in it for the first time for a project like
>this would be inadvisable. Might be easy to jump over, though.

The one catch with .net is you'll be bludgeoned into using
Visual Studio. It's not a bad IDE on paper but some people don't
like the particulars. If you've already done C++ development on
Windows, chances are you're familiar with Visual Studio so it
won't be a problem. Better look into the state of the Moonlight
union before going ahead though.

>> The roll spacing could be a subdivision of the staff spacing, yes.
>> Or you could just not show that information on the roll (it would
>> be available in the note-bar properties inspector).
>
>Or, we could have sharps and flats be in different colors,

Colors are reserved for voices on the roll, since you don't
have flag direction to show that.

>or just draw a "b" or a "#" sign inside the roll or something.

Could do.

>I'm going to set up a sourceforge page for this, I guess, and keep
>looking at what software is out there. I think that given what we've
>got we should start with a sequencer engine like TSE3 - that way the
>internal DAW stuff is done and we can focus entirely on the piano roll
>and staff notation.

Nobody will fault you for trying!

-Carl

Mike Battaglia wrote:

> Also, Carl, if we do this in C# and .NET, how hard would it be to tie
> THAT in with Python? Usually is a bad idea to have programmers
> contributing in more than one main language, I know, but maybe at
> least for plugin development or something, it would be useful
> somewhere down the line, since everyone here seems to already know it.

People here like using it, which is the point.

It's easy to use Python in .NET. You can access .NET classes directly from Python. I'm not sure how Python classes get wrapped up for .NET. It's slower because there's still a Python interpreter running in the .NET environment, but it also takes advantage of the .NET optimizations so it's about the same as CPython (Python written in C). If there's bottleneck you can re-code it in C or C# the same way you re-code in a native language for CPython.

The main thing you can't do is use Python libraries that live outside .NET. So a CPython and IronPython project wouldn't be able to share much at the low level.

Similar applies to Java but even slower.

Graham

Mike Battaglia wrote:

> Well, I get what you're saying, but it seems like it would be more
> useful for a piano roll than for a staff if that's the case. Being
> able to stretch the width of a measure is pretty vital for scoring. A
> good portion of the time that I spend while scoring anything is making
> sure it's readable and preparing it as such by moving measure lengths
> around and the like. If your goal for the staff is to simply be an
> alpha overlay on top of the piano roll and so that the piano roll is
> the main intent of focus with the staff as kind of an added bit of
> information, then that would work. But if you want the staff in this
> view to be readable in its own right then I would suggest at least
> having the option of stretching the measures in the staff if the user
> wants. Then again, it's your idea, so whatever you'd like.

There's a difference between a sequencer that has a notation view and a notation editor that has MIDI support. Carl mentioned it but I won't go and look for the reference.

A sequencer with MIDI support needn't give good printable output (sometimes there's no printing at all) but it means there's something for musicians who like that sort or thing. There's a big gap in the market here because you need explicit microtonal support and no sequencers have it. Then again if you get used to the piano roll view there isn't much that doesn't generalize to different scales. It's only a staff notation problem.

Notation editors are more about getting the notes in easily and producing good printed output. They're likely not that useful for good sounding output because you can't adjust the rhythms or set controllers. I believe there are notation editors that can handle microtonal accidentals and even custom staffs. So the only advantage you're getting by writing the whole thing from scratch with microtonality in mind is that the MIDI supports a bit better.

I can see the point of separating these two kinds of programs. The tweaks you make to look good and sound good are entirely different. So you should decide which it is you're aiming for.

Personally, I think text-based systems like ABC will do what I need for printed notation. Entering the ASCII needn't be harder than the keystrokes you would have used in a GUI. And thing like LilyPond are optimized for very good output. Rosegarden gives LilyPond output you know, which is an interesting way of bridging the gap.

So what *I'd* be interested in is a sequencer with a better way of using microtonal scales. As Csound's good enough for me I'll have a look at Blue one day and follow Chuck's project. I don't know if you're heading anywhere interesting.

Anyway, to the question, the unscientific selection of music books I have next to the computer shows equal-width measures on any given line regardless of whether it's numerical notation, staff notation or guitar tablature. I've got a sheet of bawu music in numerical notation that has unequal measures but it's a cheaply printed thing they gave me with the instrument.

Graham

Hi Mike and everyone else,

--- In MakeMicroMusic@yahoogroups.com, "Mike Battaglia"
<battaglia01@...> wrote:
>
> Just curious, what do people use nowadays to compose?
> I've been doing things using Sonar 7 in 72-tet by using
> 6 MIDI channels detuned by the right amount. As you
> might imagine, this is a huge pain to have to deal with,
> and I might imagine there's an easier way that people are
> working with.
>
> I did some searching of the archives and I found some
> notation software, but nothing stuck out at me for
> composition. Can anyone point me in a better direction?

I haven't been reading any of the lists much lately,
and i see that over the last week this post has spawned
a huge discussion. Of course i'm popping in here to
toot my own horn ...

http://tonalsoft.com/tonescape.aspx

It's the only tool i use to compose microtonal music
these days ;-)

I'm disappointed to see that no-one here has even
mentioned it yet in this big thread ... i'll warn you
in advance that even tho we have made it available
for free, it only runs under Windows XP or Vista,
and even then only if you have exactly the right
hardware and DirectX. But it has always worked on
my machines and i love it! :)

-monz
http://tonalsoft.com/tonescape.aspx
Tonescape microtonal music software

--- In MakeMicroMusic@yahoogroups.com, Carl Lumma <carl@...> wrote:
> XUL is a toolkit designed by the Mozilla project, much
> as gtk was designed for the gimp project. Either can be
> used for any application. Judging by Firefox 3, XUL has
> made great strides recently in what it can do. I don't
> know if it's up to a score editor but it's worth looking at.

So you've coded in XUL for Firefox 3? I admit I've only done XUL for
a few weeks for Firefox 2. When I worked with it, all XUL really
was, was just a set of markup tags. Any code that had to do real
work, such as accessing the file system or making HTTP requests, had
to be done in code that was referenced by the XUL tags - code written
in Javascript, Java, or something else.

That's what I meant by my "interface builder only" comment. You
can't use XUL to edit a score by itself. You have to interface it
with other technologies to do real work.

Graham Breed wrote:
>Personally, I think text-based systems like ABC will do what
>I need for printed notation. Entering the ASCII needn't be
>harder than the keystrokes you would have used in a GUI.

How do you know which notes you want?

-Carl

monz wrote:

>http://tonalsoft.com/tonescape.aspx
>
>It's the only tool i use to compose microtonal music
>these days ;-)
>
>I'm disappointed to see that no-one here has even
>mentioned it yet in this big thread ... i'll warn you
>in advance that even tho we have made it available
>for free, it only runs under Windows XP or Vista,
>and even then only if you have exactly the right
>hardware and DirectX. But it has always worked on
>my machines and i love it! :)

It's not a score editor, which is why I didn't mention
it. But I will say it's not only a .net project but
also by far the best microtonal GUI application ever
written in terms of the quality and richness of the
interactions. Unfortunately, the codebase is closed
and not being developed. It was also written before
Silverlight existed.

-Carl

--- In MakeMicroMusic@yahoogroups.com, "Prent Rodgers"
<prentrodgers@...> wrote:
>
> I use Csound and a front end I wrote 15 years ago in Turbo Pascal.
> Pascal is a little dated, but I still have customers running 40 year
> old FORTRAN and 25 year old COBOL applications, so I feel like I'm in
> good company.
>
> Csound has a very stiff learning curve, as others have discovered. I
> climbed it ten years ago, and have enjoyed it in the limited way I use
> it ever since. If anyone tries it for microtonal work, I'd be glad to
> help with that aspect of it. It does allow for any pitch and any
> instrument with no practical limitations except CPU speed and storage.
> I use it as a sampler, and then control as many aspects as possible to
> humanize the sound. Fake but accurate is my motto.

Could you give some details how you humanize it?

For that is one big disadvantage of Csound - that the eprformance
is "fake".

If not for that, the duo Csound-Blue would probably be close to perfect
to what we want here.
--
Hans Straub

At 07:54 AM 6/20/2008, you wrote:
>--- In MakeMicroMusic@yahoogroups.com, Carl Lumma <carl@...> wrote:
>> XUL is a toolkit designed by the Mozilla project, much
>> as gtk was designed for the gimp project. Either can be
>> used for any application. Judging by Firefox 3, XUL has
>> made great strides recently in what it can do. I don't
>> know if it's up to a score editor but it's worth looking at.
>
>So you've coded in XUL for Firefox 3?

No, but I played with in a bit in Firefox 1, and FF3 looks to
be doing things that couldn't be done then (in the awesomebar
and the downloads window, to name two places).

>When I worked with it, all XUL really was, was just a set
>of markup tags. Any code that had to do real work, such as
>accessing the file system or making HTTP requests

Yes, it's a widget kit, not a programming language.
See for example
http://en.wikipedia.org/wiki/Comparison_of_user_interface_markup_languages

>That's what I meant by my "interface builder only" comment. You
>can't use XUL to edit a score by itself. You have to interface it
>with other technologies to do real work.

I was talking about cross-platform widgets.

-Carl

>If not for that, the duo Csound-Blue would probably be close to perfect
>to what we want here.

It's nothing close to what I want. :(

-Carl

--- In MakeMicroMusic@yahoogroups.com, "monz" <joemonz@...> wrote:
> I'm disappointed to see that no-one here has even
> mentioned it yet in this big thread ...

That would be because...

1. It isn't a readily available 'product', with ongoing support. Then
there is what you wrote:

> i'll warn you
> in advance that even tho we have made it available
> for free, it only runs under Windows XP or Vista,
> and even then only if you have exactly the right
> hardware and DirectX.

2. No MIDI support
3. No keyboard input
4. No use of external/soft instruments (synth/sampler/whatever)

Monz, it still has the beginnings of a decent environment, mostly
suited to tuning research. It is a *long* way from being a microtonal
music creation platform, certainly when stacked up against the current
state of the art in these things in the 12tet world.

I hope things change in the future and you can find a way to get back
in and get Tonescape/Tonalsoft into a fully implemented product.

Cheers,
Jon

--- In MakeMicroMusic@yahoogroups.com, Carl Lumma <carl@...> wrote:
> Yes, it's a widget kit, not a programming language.
> See for example
> http://en.wikipedia.org/wiki/
Comparison_of_user_interface_markup_languages
>
> >That's what I meant by my "interface builder only" comment. You
> >can't use XUL to edit a score by itself. You have to interface it
> >with other technologies to do real work.
>
> I was talking about cross-platform widgets.

Ah, this wasn't clear to me before. Cool.

Carl Lumma wrote:
> Graham Breed wrote:
>> Personally, I think text-based systems like ABC will do what >> I need for printed notation. Entering the ASCII needn't be >> harder than the keystrokes you would have used in a GUI.
> > How do you know which notes you want?

Well, that's composition, isn't it?

Graham

--- In MakeMicroMusic@yahoogroups.com, Graham Breed <gbreed@...> wrote:
> >> Personally, I think text-based systems like ABC will do what
> >> I need for printed notation. Entering the ASCII needn't be
> >> harder than the keystrokes you would have used in a GUI.
> >
> > How do you know which notes you want?
>
> Well, that's composition, isn't it?
>
>
> Graham

Being able to composing music with notation is a kind
of language skill. Let's say you invent a notation system
nobody's ever used for a tuning nobody's ever heard.
It's hard enough getting skill in a system that's been
drilled into your brain since before you were born,
with tutorials and teachers and schools available for
free and charge in every corner of the globe. How are
you going to get those skills for your new invention?

I submit that the only way is to have an interactive
notation editor that'll play notes when you click on 'em,
chords when you select them, sections of the score when
you hit play, etc.

-Carl

At 02:26 PM 6/20/2008, you wrote:
>--- In MakeMicroMusic@yahoogroups.com, Carl Lumma <carl@...> wrote:
>> Yes, it's a widget kit, not a programming language.
>> See for example
>> http://en.wikipedia.org/wiki/
>> Comparison_of_user_interface_markup_languages
>>
>> >That's what I meant by my "interface builder only" comment. You
>> >can't use XUL to edit a score by itself. You have to interface it
>> >with other technologies to do real work.
>>
>> I was talking about cross-platform widgets.
>
>Ah, this wasn't clear to me before. Cool.

Er, somehow this got back on-list.

-Carl

Hi Jon,

> I hope things change in the future and you can find a way
> to get back in and get Tonescape/Tonalsoft into a fully
> implemented product.

Well, of course so do i!

What we need is an infusion of money. Tonescape has reached
a state of complexity that makes it impossible for Chris
to maintain and continue to develop by himself. Our plan
was (is) to hire a team of programmers to work under him.
But right at the moment we reached that point (January 2006),
we lost our major investor.

As i've said, we have some plans to open up the source
code at least a little ... but right now both of us are
struggling to just keep earning money to support our
families and stay afloat.

-monz
http://tonalsoft.com/tonescape.aspx
Tonescape microtonal music software

--- In MakeMicroMusic@yahoogroups.com, Carl Lumma <carl@...> wrote:
>
> >If not for that, the duo Csound-Blue would probably be close to
perfect
> >to what we want here.
>
> It's nothing close to what I want. :(
>

Alright - no staff notation editor, that's true.
How about providing Blue with that?
--
Hans Straub

Carl Lumma wrote:

> Being able to composing music with notation is a kind
> of language skill. Let's say you invent a notation system
> nobody's ever used for a tuning nobody's ever heard.
> It's hard enough getting skill in a system that's been
> drilled into your brain since before you were born,
> with tutorials and teachers and schools available for
> free and charge in every corner of the globe. How are
> you going to get those skills for your new invention?

That's something of a leading question. Notation is a form of writing. Yes, you can call it a language skill, but it isn't language. Reading with an alphabet wasn't drilled into me before I was born and I picked it up easily enough with only my mother teaching me. The notations we're talking about are exactly like standard notation but with different ways of writing pitches.

Tuning systems with new mappings may be more like a new language. Personally I learn them by playing keyboards and working on theory. Not much through notation. When I use notation it's to record ideas so that I can work out what I was thinking or to work out theories. The point is that there should be a clear mapping between the keyboard, the theory, and the notation. The more you work with the notation the more intuitive it becomes but it's still useful if it's a code that you have to think about.

The answer, anyway, is practice.

> I submit that the only way is to have an interactive
> notation editor that'll play notes when you click on 'em,
> chords when you select them, sections of the score when
> you hit play, etc.

No, the only way to learn how to compose music is to compose music. If you start with a score and you use a computer then you have to be able to get it into the computer easily and hear the results. Hence MicroABC does the job fine. What you're talking about is fine and dandy but a sequencer with staff notation support is good enough for it. That also helps the people who build their ideas by playing keyboards and drawing in piano roll views and want to see how the staff notation ties in with it.

The importance of typesetting music you've already composed is that you can share your thinking with others. We're good at sharing tunings and the ideas behind them on these lists but we aren't developing shared musical idioms. In language terms, we're all developing idiolects. A score to back up a recording helps a lot to follow the notation. Maybe sharing scores will help us to move on from being a community of musical idiots ;-)

Graham

>> It's nothing close to what I want. :(
>
>Alright - no staff notation editor, that's true.
>How about providing Blue with that?

If somebody does I'll definitely be interested. But *all*
I want is a notation editor, so there's no need to start
with Blue for my sake. -Carl

At 07:55 AM 6/21/2008, you wrote:
>Carl Lumma wrote:
>
>> Being able to composing music with notation is a kind
>> of language skill. Let's say you invent a notation system
>> nobody's ever used for a tuning nobody's ever heard.
>> It's hard enough getting skill in a system that's been
>> drilled into your brain since before you were born,
>> with tutorials and teachers and schools available for
>> free and charge in every corner of the globe. How are
>> you going to get those skills for your new invention?
>
>That's something of a leading question. Notation is a form
>of writing. Yes, you can call it a language skill, but it
>isn't language.

Sure it's a language. You could define it in Backus–Naur Form.

>Reading with an alphabet wasn't drilled
>into me before I was born and I picked it up easily enough
>with only my mother teaching me.

You didn't have reading/writing in school?. You didn't
read existing text in the world all around you?

>The notations we're
>talking about are exactly like standard notation but with
>different ways of writing pitches.

And rhythms (I've mentioned Cowell notations several times now).

>Tuning systems with new mappings may be more like a new
>language. Personally I learn them by playing keyboards and
>working on theory. Not much through notation.

The bottom line is, the time it takes you to transcribe a
a normal piece of music in normal notation after hearing it
for the first time is inversely related to how successful
you'll be writing new music in standard notation.
The same relation applies to microtonal music/notation.

>> I submit that the only way is to have an interactive
>> notation editor that'll play notes when you click on 'em,
>> chords when you select them, sections of the score when
>> you hit play, etc.
>
>No, the only way to learn how to compose music is to compose
>music. If you start with a score

Where do you get the score from? You're going to work
it out with pen and paper on the keyboard? That's one
way. In effect, you're doing the same thing manually
with the keyboard that I'm suggestion the notation
editor do.

-Carl

Hi All,

This is Steven Yi, the author of blue. I'm a reader of this list but
don't often post, but I thought I'd reply to just mention some
thoughts about this thread. (Admittedly, I have only been
half-following the thread, so some things I say may already be
mentioned; if so, apologies.)

That said, I think Csound, for my needs, is probably the most
expressive tool I've come across. I use python within blue using an
orchestral composition library I wrote that helps to abstract out the
different musical ideas (I write a score and have a Performer object
"perform" the score, applying the performer's location, with the
"accuracy" of the performer affecting things like pitch and
amplitude). I've found write text processing functions in python
fairly easy, and because Csound's note values are doubles or floats
and not integers bound to 0-127 like MIDI, it's quite flexible. I've
also used the Random NoteProcessor built into blue to just add a tiny
random value to note start times or durations or amplitudes to
"humanize" a score. It's easy to apply per-score-object, to a track,
or to the entire score. It's also very easy for me to explore limited
aleatory and glissandi sound masses, both of which are very
interesting to me. I know one person has used blue to work with
granular clouds spatialized using Ambisonics, and others have used a
number of other tools liks Common Music within blue for their work.
This is just to get an idea of some of the ways blue/csound have been
used.

In regards to microtonal work, I've used the microtonal PianoRoll in
blue to experiment with ideas with different scales. Also to note,
the Tracker object in blue also supports Scala scales, so one can
compose using a tracker interface if one is familiar with those, but
also can work with Scala scales. I have some python code that reads in
Scala scales and works with my orchestral composition library and have
used that in the past too. These days I haven't been exploring scales
as much as I have been hand-tuning notes to get the right sound,
usually justly-tuning the notes from a base note when appropriate.

I am sure blue is not an ideal software for everyone, but it allows me
to explore the musical sounds and ideas I'm interested in very well
and it seems to work well for others I know too. However, I'd just
say that if you're going to start a new piece of software, something
that is of great concern to me is the long-term for software. It
pains me to see so much software written and abandoned and users of
that software now having no way to really access their work now.
People invest time in a tool to create musical work that hopefully has
a great deal of meaning to them, and I think it is sad to see when
either they can not access the tool any longer to continue their work
or even just to see what they had done.

This happens for a number of reasons, but the most common being the
developer of the software retires from working on it, or the operating
system or the language is used to develop the software no longer is
around. Lots of musical work done with software for OS9 can not be
accessed anymore. Lots of small musical software tools written in
Visual Basic are difficult to find source for. In these situations,
if the operating system decides to end support for backwards
compatibility, as it seems is often the case with Mac, there is a risk
of losing one's work.

This has been a big thought on my mind in terms of the long-term
development of music and hope that other developers consider these
things when building a piece of software and making it available to
other people to use.

Also, I'd implore that if you don't find blue for you and do start to
look at building your own software, that you might want to a least
take a look at blue's design to look at how some of the issues
regarding support for microtonal work have been addressed, and to feel
free to take a look at code and take anything that may be of use (it
is open source after all). Also, let me know if you do and hopefully
some healthy exchange of ideas can occur. =)

Thanks!
steven

--- In MakeMicroMusic@yahoogroups.com, "hstraub64" <straub@...> wrote:
>
> --- In MakeMicroMusic@yahoogroups.com, "Prent Rodgers"
> <prentrodgers@> wrote:
> >
> > I use Csound and a front end I wrote 15 years ago in Turbo Pascal.
> > Pascal is a little dated, but I still have customers running 40 year
> > old FORTRAN and 25 year old COBOL applications, so I feel like I'm in
> > good company.
> >
> > Csound has a very stiff learning curve, as others have discovered. I
> > climbed it ten years ago, and have enjoyed it in the limited way I use
> > it ever since. If anyone tries it for microtonal work, I'd be glad to
> > help with that aspect of it. It does allow for any pitch and any
> > instrument with no practical limitations except CPU speed and storage.
> > I use it as a sampler, and then control as many aspects as possible to
> > humanize the sound. Fake but accurate is my motto.
>
> Could you give some details how you humanize it?
>
> For that is one big disadvantage of Csound - that the eprformance
> is "fake".
>
> If not for that, the duo Csound-Blue would probably be close to perfect
> to what we want here.
> --
> Hans Straub
>

Steven,

Thanks for the insighteful discussion. It really hits home I have lost
generations of work (and with it lost the ability to build on it and
move forward) because I wrote dedicated music software for specific
platforms. This time around I want to use supported tools as much as
possible so I can focus on my tiny additions, which are big to me but
small in terms of the software necessary to support it. At first I
thought I would start with portaudio, I had been convinced by posts
here that cSound was the way to go. You make me think that blue itself
on top of cSound is even closer to my needs. I'll start looking at it
next. If you have any pointers to the best way to get started. I would
appreciate hearing about it. Most attractive is the idea of developing
a library in python. I want to create my own musical objects,
operations on those objects and relationships between objects.

Barbershop Steve

Thanks for chipping in. There is so much wonderful software out there
these days that it can be paralyzing to attempt to make an intelligent
choice.

On Sat, Jun 21, 2008 at 2:34 PM, Steven Yi <stevenyi@...> wrote:
> Hi All,
>
> This is Steven Yi, the author of blue. I'm a reader of this list but
> don't often post, but I thought I'd reply to just mention some
> thoughts about this thread. (Admittedly, I have only been
> half-following the thread, so some things I say may already be
> mentioned; if so, apologies.)
>
> That said, I think Csound, for my needs, is probably the most
> expressive tool I've come across. I use python within blue using an
> orchestral composition library I wrote that helps to abstract out the
> different musical ideas (I write a score and have a Performer object
> "perform" the score, applying the performer's location, with the
> "accuracy" of the performer affecting things like pitch and
> amplitude). I've found write text processing functions in python
> fairly easy, and because Csound's note values are doubles or floats
> and not integers bound to 0-127 like MIDI, it's quite flexible. I've
> also used the Random NoteProcessor built into blue to just add a tiny
> random value to note start times or durations or amplitudes to
> "humanize" a score. It's easy to apply per-score-object, to a track,
> or to the entire score. It's also very easy for me to explore limited
> aleatory and glissandi sound masses, both of which are very
> interesting to me. I know one person has used blue to work with
> granular clouds spatialized using Ambisonics, and others have used a
> number of other tools liks Common Music within blue for their work.
> This is just to get an idea of some of the ways blue/csound have been
> used.
>
> In regards to microtonal work, I've used the microtonal PianoRoll in
> blue to experiment with ideas with different scales. Also to note,
> the Tracker object in blue also supports Scala scales, so one can
> compose using a tracker interface if one is familiar with those, but
> also can work with Scala scales. I have some python code that reads in
> Scala scales and works with my orchestral composition library and have
> used that in the past too. These days I haven't been exploring scales
> as much as I have been hand-tuning notes to get the right sound,
> usually justly-tuning the notes from a base note when appropriate.
>
> I am sure blue is not an ideal software for everyone, but it allows me
> to explore the musical sounds and ideas I'm interested in very well
> and it seems to work well for others I know too. However, I'd just
> say that if you're going to start a new piece of software, something
> that is of great concern to me is the long-term for software. It
> pains me to see so much software written and abandoned and users of
> that software now having no way to really access their work now.
> People invest time in a tool to create musical work that hopefully has
> a great deal of meaning to them, and I think it is sad to see when
> either they can not access the tool any longer to continue their work
> or even just to see what they had done.
>
> This happens for a number of reasons, but the most common being the
> developer of the software retires from working on it, or the operating
> system or the language is used to develop the software no longer is
> around. Lots of musical work done with software for OS9 can not be
> accessed anymore. Lots of small musical software tools written in
> Visual Basic are difficult to find source for. In these situations,
> if the operating system decides to end support for backwards
> compatibility, as it seems is often the case with Mac, there is a risk
> of losing one's work.
>
> This has been a big thought on my mind in terms of the long-term
> development of music and hope that other developers consider these
> things when building a piece of software and making it available to
> other people to use.
>
> Also, I'd implore that if you don't find blue for you and do start to
> look at building your own software, that you might want to a least
> take a look at blue's design to look at how some of the issues
> regarding support for microtonal work have been addressed, and to feel
> free to take a look at code and take anything that may be of use (it
> is open source after all). Also, let me know if you do and hopefully
> some healthy exchange of ideas can occur. =)
>
> Thanks!
> steven
>
> --- In MakeMicroMusic@yahoogroups.com, "hstraub64" <straub@...> wrote:
>>
>> --- In MakeMicroMusic@yahoogroups.com, "Prent Rodgers"
>> <prentrodgers@> wrote:
>> >
>> > I use Csound and a front end I wrote 15 years ago in Turbo Pascal.
>> > Pascal is a little dated, but I still have customers running 40 year
>> > old FORTRAN and 25 year old COBOL applications, so I feel like I'm in
>> > good company.
>> >
>> > Csound has a very stiff learning curve, as others have discovered. I
>> > climbed it ten years ago, and have enjoyed it in the limited way I use
>> > it ever since. If anyone tries it for microtonal work, I'd be glad to
>> > help with that aspect of it. It does allow for any pitch and any
>> > instrument with no practical limitations except CPU speed and storage.
>> > I use it as a sampler, and then control as many aspects as possible to
>> > humanize the sound. Fake but accurate is my motto.
>>
>> Could you give some details how you humanize it?
>>
>> For that is one big disadvantage of Csound - that the eprformance
>> is "fake".
>>
>> If not for that, the duo Csound-Blue would probably be close to perfect
>> to what we want here.
>> --
>> Hans Straub
>>
>
>

--
Steve Morris
barbershopsteve@...
Bass: Boston Wailers
Bass: Sounds Of Concord
Motto: Old age and treachery will always prevail over youth and skill

I agree with Carl on this one. I have been frustrated all my (longish)
life with the limitations of existing musical notation systems. I have
sympathy for people who want to create alternatives. However I find
that I have wasted substantial amounts of time and energy trying to
support various systems I invented mostly with software I have
written. The result has been that I haven't got much done musically
and worse what little I have done could not be easily shared. When I
invented my own worlds I not only removed myself from the broader
music community, I also made it difficult for people to join mine. In
retrospect I would have gotten more done if I had just bit the bullet
and embraced traditional musical notation systems, especially now that
I realize that useful enhancements and alternatives already exist.

Carl is particularly on point when he says that music notation is a
language. It is not just LIKE a language, it IS a language. Like math
it is a specialized language.

On Sat, Jun 21, 2008 at 12:49 PM, Carl Lumma <carl@lumma.org> wrote:
> At 07:55 AM 6/21/2008, you wrote:
>>Carl Lumma wrote:
>>
>>> Being able to composing music with notation is a kind
>>> of language skill. Let's say you invent a notation system
>>> nobody's ever used for a tuning nobody's ever heard.
>>> It's hard enough getting skill in a system that's been
>>> drilled into your brain since before you were born,
>>> with tutorials and teachers and schools available for
>>> free and charge in every corner of the globe. How are
>>> you going to get those skills for your new invention?
>>
>>That's something of a leading question. Notation is a form
>>of writing. Yes, you can call it a language skill, but it
>>isn't language.
>
> Sure it's a language. You could define it in Backus–Naur Form.
>
>>Reading with an alphabet wasn't drilled
>>into me before I was born and I picked it up easily enough
>>with only my mother teaching me.
>
> You didn't have reading/writing in school?. You didn't
> read existing text in the world all around you?
>
>>The notations we're
>>talking about are exactly like standard notation but with
>>different ways of writing pitches.
>
> And rhythms (I've mentioned Cowell notations several times now).
>
>>Tuning systems with new mappings may be more like a new
>>language. Personally I learn them by playing keyboards and
>>working on theory. Not much through notation.
>
> The bottom line is, the time it takes you to transcribe a
> a normal piece of music in normal notation after hearing it
> for the first time is inversely related to how successful
> you'll be writing new music in standard notation.
> The same relation applies to microtonal music/notation.
>
>>> I submit that the only way is to have an interactive
>>> notation editor that'll play notes when you click on 'em,
>>> chords when you select them, sections of the score when
>>> you hit play, etc.
>>
>>No, the only way to learn how to compose music is to compose
>>music. If you start with a score
>
> Where do you get the score from? You're going to work
> it out with pen and paper on the keyboard? That's one
> way. In effect, you're doing the same thing manually
> with the keyboard that I'm suggestion the notation
> editor do.
>
> -Carl
>
>

--
Steve Morris
barbershopsteve@...
Bass: Boston Wailers
Bass: Sounds Of Concord
Motto: Old age and treachery will always prevail over youth and skill

Hi Steve,

Regarding pointers to getting started to evaluate if blue/csound may
work for you, I have to say that there is a manual for blue but that
it has some outdated information unfortunately. I had another user
recently I worked with over email to help get started and he had given
some useful suggestions on areas of the manual which were a bit
deficient. I haven't yet had a chance to get started to revise the
manual, but am starting to plan going through page by page and doing a
reworking of it.

Until then, I'd suggest getting a bit familiar with Csound by
downloading and installing it and reading Dr. Boulanger's TOOTS at:

http://www.csounds.com/toots/

These are a pretty good introduction to Csound. There's also the
manual at:

http://www.csounds.com/manual

There's been some discusson on the Csound list about deficiencies of
both for new users and it's all been very valid critique, but for now,
those are the usual ways in.

After that I'd recommend downloading blue from:

http://www.csounds.com/stevenyi/blue

For microtonal music, I'd highly recommend opening Dave Seidel's
pieces in the examples/pieces/daveSeidel folder. You can see some of
the different ways he's approached microtonal work with the different
soundObjects in blue.

My guess is that the above all won't all make sense at first though,
so I'd say please feel free to email me anytime and I'll try to walk
through anything that you get stuck on. I've also generally extended
invitations that I'm available to do phone calls if you'd like to
schedule a chat about all this. It might save some time when getting
started to see if this is really a set of tools that will help you
achieve what you'd like musically or not. (However, I'm gone from
Sunday through Friday, so will not be really available this next week,
but any time after that will be replying to emails and available for
calls).

Thanks and good luck!
steven

--- In MakeMicroMusic@yahoogroups.com, "Steve Morris"
<barbershopsteve@...> wrote:
>
> Steven,
>
> Thanks for the insighteful discussion. It really hits home I have lost
> generations of work (and with it lost the ability to build on it and
> move forward) because I wrote dedicated music software for specific
> platforms. This time around I want to use supported tools as much as
> possible so I can focus on my tiny additions, which are big to me but
> small in terms of the software necessary to support it. At first I
> thought I would start with portaudio, I had been convinced by posts
> here that cSound was the way to go. You make me think that blue itself
> on top of cSound is even closer to my needs. I'll start looking at it
> next. If you have any pointers to the best way to get started. I would
> appreciate hearing about it. Most attractive is the idea of developing
> a library in python. I want to create my own musical objects,
> operations on those objects and relationships between objects.
>
> Barbershop Steve
>
> Thanks for chipping in. There is so much wonderful software out there
> these days that it can be paralyzing to attempt to make an intelligent
> choice.
>
> On Sat, Jun 21, 2008 at 2:34 PM, Steven Yi <stevenyi@...> wrote:
> > Hi All,
> >
> > This is Steven Yi, the author of blue. I'm a reader of this list but
> > don't often post, but I thought I'd reply to just mention some
> > thoughts about this thread. (Admittedly, I have only been
> > half-following the thread, so some things I say may already be
> > mentioned; if so, apologies.)
> >
> > That said, I think Csound, for my needs, is probably the most
> > expressive tool I've come across. I use python within blue using an
> > orchestral composition library I wrote that helps to abstract out the
> > different musical ideas (I write a score and have a Performer object
> > "perform" the score, applying the performer's location, with the
> > "accuracy" of the performer affecting things like pitch and
> > amplitude). I've found write text processing functions in python
> > fairly easy, and because Csound's note values are doubles or floats
> > and not integers bound to 0-127 like MIDI, it's quite flexible. I've
> > also used the Random NoteProcessor built into blue to just add a tiny
> > random value to note start times or durations or amplitudes to
> > "humanize" a score. It's easy to apply per-score-object, to a track,
> > or to the entire score. It's also very easy for me to explore limited
> > aleatory and glissandi sound masses, both of which are very
> > interesting to me. I know one person has used blue to work with
> > granular clouds spatialized using Ambisonics, and others have used a
> > number of other tools liks Common Music within blue for their work.
> > This is just to get an idea of some of the ways blue/csound have been
> > used.
> >
> > In regards to microtonal work, I've used the microtonal PianoRoll in
> > blue to experiment with ideas with different scales. Also to note,
> > the Tracker object in blue also supports Scala scales, so one can
> > compose using a tracker interface if one is familiar with those, but
> > also can work with Scala scales. I have some python code that reads in
> > Scala scales and works with my orchestral composition library and have
> > used that in the past too. These days I haven't been exploring scales
> > as much as I have been hand-tuning notes to get the right sound,
> > usually justly-tuning the notes from a base note when appropriate.
> >
> > I am sure blue is not an ideal software for everyone, but it allows me
> > to explore the musical sounds and ideas I'm interested in very well
> > and it seems to work well for others I know too. However, I'd just
> > say that if you're going to start a new piece of software, something
> > that is of great concern to me is the long-term for software. It
> > pains me to see so much software written and abandoned and users of
> > that software now having no way to really access their work now.
> > People invest time in a tool to create musical work that hopefully has
> > a great deal of meaning to them, and I think it is sad to see when
> > either they can not access the tool any longer to continue their work
> > or even just to see what they had done.
> >
> > This happens for a number of reasons, but the most common being the
> > developer of the software retires from working on it, or the operating
> > system or the language is used to develop the software no longer is
> > around. Lots of musical work done with software for OS9 can not be
> > accessed anymore. Lots of small musical software tools written in
> > Visual Basic are difficult to find source for. In these situations,
> > if the operating system decides to end support for backwards
> > compatibility, as it seems is often the case with Mac, there is a risk
> > of losing one's work.
> >
> > This has been a big thought on my mind in terms of the long-term
> > development of music and hope that other developers consider these
> > things when building a piece of software and making it available to
> > other people to use.
> >
> > Also, I'd implore that if you don't find blue for you and do start to
> > look at building your own software, that you might want to a least
> > take a look at blue's design to look at how some of the issues
> > regarding support for microtonal work have been addressed, and to feel
> > free to take a look at code and take anything that may be of use (it
> > is open source after all). Also, let me know if you do and hopefully
> > some healthy exchange of ideas can occur. =)
> >
> > Thanks!
> > steven
> >
> > --- In MakeMicroMusic@yahoogroups.com, "hstraub64" <straub@> wrote:
> >>
> >> --- In MakeMicroMusic@yahoogroups.com, "Prent Rodgers"
> >> <prentrodgers@> wrote:
> >> >
> >> > I use Csound and a front end I wrote 15 years ago in Turbo Pascal.
> >> > Pascal is a little dated, but I still have customers running 40
year
> >> > old FORTRAN and 25 year old COBOL applications, so I feel like
I'm in
> >> > good company.
> >> >
> >> > Csound has a very stiff learning curve, as others have
discovered. I
> >> > climbed it ten years ago, and have enjoyed it in the limited
way I use
> >> > it ever since. If anyone tries it for microtonal work, I'd be
glad to
> >> > help with that aspect of it. It does allow for any pitch and any
> >> > instrument with no practical limitations except CPU speed and
storage.
> >> > I use it as a sampler, and then control as many aspects as
possible to
> >> > humanize the sound. Fake but accurate is my motto.
> >>
> >> Could you give some details how you humanize it?
> >>
> >> For that is one big disadvantage of Csound - that the eprformance
> >> is "fake".
> >>
> >> If not for that, the duo Csound-Blue would probably be close to
perfect
> >> to what we want here.
> >> --
> >> Hans Straub
> >>
> >
> >
>
>
>
> --
> Steve Morris
> barbershopsteve@...
> Bass: Boston Wailers
> Bass: Sounds Of Concord
> Motto: Old age and treachery will always prevail over youth and skill
>

Steve Morris wrote:

> Carl is particularly on point when he says that music notation is a
> language. It is not just LIKE a language, it IS a language. Like math
> it is a specialized language.

Sure, or maybe it's a language like dance steps are a language. It isn't a language like English is a language and it isn't anywhere near as difficult to learn.

Graham

Graham wrote:
>> Carl is particularly on point when he says that music notation is a
>> language. It is not just LIKE a language, it IS a language. Like math
>> it is a specialized language.
>
>Sure, or maybe it's a language like dance steps are a
>language. It isn't a language like English is a language
>and it isn't anywhere near as difficult to learn.

I don't know if it's as difficult to learn, but it's very
difficult to learn. Lots of people take lessons as kids,
but very very few attain any kind of proficiency at
sightreading, let alone composing.

But really the question is how to make it easiest. And
the way to do that is by reducing the feedback time between
the learner and the environment. A notation editor where
notes are objects that can be played with a click is
probably going to provide faster feedback than sitting at
the keyboard with a clipboard and a pencil and going back
and forth. Speaking from personal experience, I certainly
found this to be the case when I started learning to compose
in standard notation (having already practiced sightreading
in choirs, and with the trumpet and piano, for most of
my life).

-Carl

I wrote:
>But really the question is how to make it easiest. And
>the way to do that is by reducing the feedback time between
>the learner and the environment. A notation editor where
>notes are objects that can be played with a click is
>probably going to provide faster feedback than sitting at
>the keyboard with a clipboard and a pencil and going back
>and forth.

Though clearly both methods will be helpful, owing to the
motor associations involved with playing the keyboard.

-Carl

Carl Lumma wrote:

> I don't know if it's as difficult to learn, but it's very
> difficult to learn. Lots of people take lessons as kids,
> but very very few attain any kind of proficiency at
> sightreading, let alone composing.

Sure, because we took lessons but we didn't practice. Playing an instrument's difficult. Sight reading's difficult because you need to have a concept of each note you need to play and, depending on the instrument, the finger positions. You need to map symbols to those concepts very quickly. Figuring out how to a play a tune on a piano's a lot easier.

I don't think the music lessons people take as kids help at all with composition. They learn to copy but not to create. It's also possible to compose without being able to read -- Django Reinhardt tried by with dictation.

> But really the question is how to make it easiest. And
> the way to do that is by reducing the feedback time between
> the learner and the environment. A notation editor where
> notes are objects that can be played with a click is
> probably going to provide faster feedback than sitting at
> the keyboard with a clipboard and a pencil and going back
> and forth. Speaking from personal experience, I certainly
> found this to be the case when I started learning to compose
> in standard notation (having already practiced sightreading
> in choirs, and with the trumpet and piano, for most of
> my life).

Speaking from my experience, I found writing music on paper and transcribing it into Csound to be very boring. I got a lot more music finished when I was using a keyboard and a sequencer. Now I've got a keyboard again maybe I'll get back into the swing. I seem to be the only one in these parts who cares about logical keyboard mappings.

The feedback time is certainly not optimized by spending months of concentrated effort writing a score editor that you and few other people are going to use.

If you work with ABC you'll start to learn how to read raw ABC. Then you can make the changes directly and use the output when you want to see a score. GUIs are worth having but I don't think they'll reduce the feedback time because you have to tell the computer what you mean somehow. ABC's a pretty dense format.

Graham

Carl Lumma wrote:
> At 07:55 AM 6/21/2008, you wrote:

>> That's something of a leading question. Notation is a form >> of writing. Yes, you can call it a language skill, but it >> isn't language.
> > Sure it's a language. You could define it in Backus�Naur Form.

So you could call it a language in the sense that programming languages are languages. But not in the sense that English is a language.

>> Reading with an alphabet wasn't drilled >> into me before I was born and I picked it up easily enough >> with only my mother teaching me.
> > You didn't have reading/writing in school?. You didn't
> read existing text in the world all around you?

I could read when I started school. I couldn't read anything before I was born.

>> The notations we're >> talking about are exactly like standard notation but with >> different ways of writing pitches.
> > And rhythms (I've mentioned Cowell notations several times now).

You mention that. It's orthogonal to microtonal support.

>> Tuning systems with new mappings may be more like a new >> language. Personally I learn them by playing keyboards and >> working on theory. Not much through notation.
> > The bottom line is, the time it takes you to transcribe a
> a normal piece of music in normal notation after hearing it
> for the first time is inversely related to how successful
> you'll be writing new music in standard notation.
> The same relation applies to microtonal music/notation.

Maybe, because to transcribe easily you need to understand the music. Converting from one notation to another is a different exercise. It's not that difficult to learn a new notation when you're already familiar with the musical rules.

>>> I submit that the only way is to have an interactive
>>> notation editor that'll play notes when you click on 'em,
>>> chords when you select them, sections of the score when
>>> you hit play, etc.
>> No, the only way to learn how to compose music is to compose >> music. If you start with a score
> > Where do you get the score from? You're going to work
> it out with pen and paper on the keyboard? That's one
> way. In effect, you're doing the same thing manually
> with the keyboard that I'm suggestion the notation
> editor do.

Or pencil and paper. Either from your head or with a keyboard. So in effect what you said isn't the only way after all.

Graham

Graham wrote:
>I don't think the music lessons people take as kids help at
>all with composition.

Depends what kind of lessons they are, I guess, and whether
the kid is interested in them or being forced to take them.

> They learn to copy but not to create.

The foundation is incredibly useful when going to create
later. One could not write Ulysses without a muscle-memory
command of basic English.

> It's also possible to compose without being able to read
>-- Django Reinhardt tried by with dictation.

Sure, if you have a composer or music editor willing to
take your dictation, then feh! on a GUI score editor!

>> But really the question is how to make it easiest. And
>> the way to do that is by reducing the feedback time between
>> the learner and the environment. A notation editor where
>> notes are objects that can be played with a click is
>> probably going to provide faster feedback than sitting at
>> the keyboard with a clipboard and a pencil and going back
>> and forth. Speaking from personal experience, I certainly
>> found this to be the case when I started learning to compose
>> in standard notation (having already practiced sightreading
>> in choirs, and with the trumpet and piano, for most of
>> my life).
>
>Speaking from my experience, I found writing music on paper
>and transcribing it into Csound to be very boring. I got a
>lot more music finished when I was using a keyboard and a
>sequencer.

And I think you'd get even more done with an interactive
notation editor.

>I seem to be the only one in these
>parts who cares about logical keyboard mappings.

On these lists or where you're staying now?

>The feedback time is certainly not optimized by spending
>months of concentrated effort writing a score editor that
>you and few other people are going to use.

I don't think that makes any sense.

>If you work with ABC you'll start to learn how to read raw
>ABC. Then you can make the changes directly and use the
>output when you want to see a score. GUIs are worth having
>but I don't think they'll reduce the feedback time because
>you have to tell the computer what you mean somehow. ABC's
>a pretty dense format.

Have you ever used a notation editor like Finale or Sibelius?

-Carl

>> You didn't have reading/writing in school?. You didn't
>> read existing text in the world all around you?
>
>I could read when I started school.

Well you were exceptional, or you started school late. I
learned to read childrens' books when I was in Kindergarten
and Consumer Reports articles when I was in 3rd grade and
scientific papers when I was in my early 20's. Last I
attempted Ulysses I still couldn't hack it. Probably should
try now.

>>> Tuning systems with new mappings may be more like a new
>>> language. Personally I learn them by playing keyboards and
>>> working on theory. Not much through notation.
>>
>> The bottom line is, the time it takes you to transcribe a
>> a normal piece of music in normal notation after hearing it
>> for the first time is inversely related to how successful
>> you'll be writing new music in standard notation.
>> The same relation applies to microtonal music/notation.
>
>Maybe, because to transcribe easily you need to understand
>the music. Converting from one notation to another is a
>different exercise. It's not that difficult to learn a new
>notation when you're already familiar with the musical rules.

Some of the skill crosses over, but... I'm good enough at
transcription to test out of first-year music theory classes
at a conservatory, but I can't put pajara music down on
paper to save my life, let alone learning all of the umpteen
systems I might want to compose in.

-Carl

Carl Lumma wrote:
>>> You didn't have reading/writing in school?. You didn't
>>> read existing text in the world all around you?
>> I could read when I started school.
> > Well you were exceptional, or you started school late. I
> learned to read childrens' books when I was in Kindergarten
> and Consumer Reports articles when I was in 3rd grade and
> scientific papers when I was in my early 20's. Last I
> attempted Ulysses I still couldn't hack it. Probably should
> try now.

I started school full time at 5 and could read anything I wanted to by then. Most schools systems these days will teach reading before then but it wasn't the fashion at the time, so I learnt at home. Let's assume I wasn't reading at all at 3, so it took me at most 2 years. Probably about the same for you. I've been using decimal notation for 7.

Do you think you would understand Ulysses if somebody read it to you? Actually I did hear some of Joyce's version read on the radio a long time ago, probably when I was 10. I don't remember it being at all difficult :-S Maybe it was only the easy bits, or I didn't notice what I wasn't understanding. Homer's still Greek to me in print or sound.

> Some of the skill crosses over, but... I'm good enough at
> transcription to test out of first-year music theory classes
> at a conservatory, but I can't put pajara music down on
> paper to save my life, let alone learning all of the umpteen
> systems I might want to compose in.

If you can play it on a keyboard, you can copy the notes onto paper, can't you? If you can't play it on a keyboard then you haven't internalized the system. It isn't a problem with notation.

I didn't know numerical notation at all a few years ago and never studied it systematically, but I reckon I could convert between it and staff notation for simple, diatonic music (which is all I ever see in numerical notation). I can also read chords from a staff and convert them to chord symbols, then play them on a guitar. I can't transcribe to save my life so sue me.

Frequently I don't know what my music -- especially the harmony -- will sound like when I write it. But I've learnt rules that make it sound good and if they don't work I can go back and revise it. If I start with a tune in my head I need a keyboard to work it out.

I don't think about umpteen different systems. I think about different mappings from JI, usually 11-limit. A lot of the same rules carry over. Don't you think in the 7-limit with your barbershop experience?

Graham

Carl Lumma wrote:
> Graham wrote:
>> I don't think the music lessons people take as kids help at >> all with composition.
> > Depends what kind of lessons they are, I guess, and whether
> the kid is interested in them or being forced to take them.

IME children who learn music formally learn to play other people's music and not write their own. Maybe there are different kinds of lessons somewhere out there.

>> They learn to copy but not to create.
> > The foundation is incredibly useful when going to create
> later. One could not write Ulysses without a muscle-memory
> command of basic English.

The Odyssey was probably composed by an illiterate poet. So you mean Joyce? I checked Wikipedia. It says he wrote a poem at the age of 9, with 3 years of formal schooling, and was greatly encouraged by his father. That supports my argument -- that he couldn't have written Ulysses without a lot of experience of creative writing.

>> It's also possible to compose without being able to read >> -- Django Reinhardt tried by with dictation.
> > Sure, if you have a composer or music editor willing to
> take your dictation, then feh! on a GUI score editor!

If your music's good enough you'll attract the attention. You know as well as I do the contribution made to Jazz and other musics by musicians who couldn't read notation. If you prefer to write that's fine and you need the tools but composition's still a long way from manipulation of notation.

>> Speaking from my experience, I found writing music on paper >> and transcribing it into Csound to be very boring. I got a >> lot more music finished when I was using a keyboard and a >> sequencer.
> > And I think you'd get even more done with an interactive
> notation editor.

You can prove anything from a counter factual.

>> I seem to be the only one in these >> parts who cares about logical keyboard mappings.
> > On these lists or where you're staying now?

Either but the latter's hardly surprising.

>> The feedback time is certainly not optimized by spending >> months of concentrated effort writing a score editor that >> you and few other people are going to use.
> > I don't think that makes any sense.

We're not talking about a score editor magically bursting into life and taking or leaving it. Somebody has to put a lot of effort into writing it. Of course, it's up to you whether you want to do that instead of composing with existing tools and improving them incrementally. My original point is that you need to decide what you want -- score editor or sequencer with score view.

>> If you work with ABC you'll start to learn how to read raw >> ABC. Then you can make the changes directly and use the >> output when you want to see a score. GUIs are worth having >> but I don't think they'll reduce the feedback time because >> you have to tell the computer what you mean somehow. ABC's >> a pretty dense format.
> > Have you ever used a notation editor like Finale or Sibelius?

No.

Graham

Graham wrote:
>Frequently I don't know what my music -- especially the
>harmony -- will sound like when I write it.

Well, that's a problem in my book. If it works for you,
then don't worry about a notation editor.

>I don't think about umpteen different systems. I think
>about different mappings from JI, usually 11-limit. A lot
>of the same rules carry over. Don't you think in the
>7-limit with your barbershop experience?

The harmony isn't the thing the system, the scale is.
Barbershop's still diatonic.

-Carl

Graham wrote:
>You know as well as I do the contribution made to Jazz and
>other musics by musicians who couldn't read notation. If
>you prefer to write that's fine and you need the tools but
>composition's still a long way from manipulation of notation.

It requires an instrument. A bass guitar is an instrument,
a Tonal Plexus is an instrument, and a piece of paper and a
pencil is an instrument in the hands of those who can hear
out notation. It's just pulp otherwise.

The point is to create the music, not record it. You may
be thinking of notation as a way to record music. That's
not how I think of it and if that's all notation was it
would be completely obsolete today.

>>> I seem to be the only one in these
>>> parts who cares about logical keyboard mappings.
>>
>> On these lists or where you're staying now?
>
>Either but the latter's hardly surprising.

I take it you don't consider sagittal logical, or the
generalized MOS notation I (and to some extent Herman Miller)
advocate.

-Carl

quickly off topic which i hope will be tolerated in us just passing 'bloomsday'.
most people i know found that if they started with the question and answer segment,
second to last chapter, found it much easier.
there is also a chart with the themes , colors, style that also helps, but i no longer have this

/^_,',',',_ //^ /Kraig Grady_ ^_,',',',_
Mesotonal Music from:
_'''''''_ ^North/Western Hemisphere: North American Embassy of Anaphoria Island <http://anaphoria.com/>

_'''''''_ ^South/Eastern Hemisphere:
Austronesian Outpost of Anaphoria <http://anaphoriasouth.blogspot.com/>

',',',',',',',',',',',',',',',',',',',',',',',',',',',',',

Graham Breed wrote:
>
>
>
> Do you think you would understand Ulysses if somebody read
> it to you? Actually I did hear some of Joyce's version read
> on the radio a long time ago, probably when I was 10. I
> don't remember it being at all difficult :-S Maybe it was
> only the easy bits, or I didn't notice what I wasn't
> understanding. Homer's still Greek to me in print or sound.
>
>
> Recent Activity
>
> *
> 1
> New Members
> </makemicromusic/members;_ylc=X3oDMTJmdHQzZjFwBF9TAzk3MzU5NzE0BGdycElkAzM2NjA0NjUEZ3Jwc3BJZAMxNzA1MDIzODY1BHNlYwN2dGwEc2xrA3ZtYnJzBHN0aW1lAzEyMTQxNDEzMzg->
> *
> 1
> New Files
> </makemicromusic/files;_ylc=X3oDMTJnc204Y3U5BF9TAzk3MzU5NzE0BGdycElkAzM2NjA0NjUEZ3Jwc3BJZAMxNzA1MDIzODY1BHNlYwN2dGwEc2xrA3ZmaWxlcwRzdGltZQMxMjE0MTQxMzM4>
>
> Visit Your Group > </makemicromusic;_ylc=X3oDMTJlOTl1MnA2BF9TAzk3MzU5NzE0BGdycElkAzM2NjA0NjUEZ3Jwc3BJZAMxNzA1MDIzODY1BHNlYwN2dGwEc2xrA3ZnaHAEc3RpbWUDMTIxNDE0MTMzOA--> >
> Need traffic?
>
> Drive customers > <http://us.ard.yahoo.com/SIG=13ovh2pvq/M=493064.12016308.12445700.8674578/D=grplch/S=1705023865:NC/Y=YAHOO/EXP=1214148539/L=/B=htUXAELaX.w-/J=1214141339005490/A=3848644/R=0/SIG=131l83flq/*http://searchmarketing.yahoo.com/arp/srchv2.php?o=US2006&cmp=Yahoo&ctv=Groups5&s=Y&s2=&s3=&b=50>
>
> With search ads
>
> on Yahoo!
>
> Do-It-Yourselfers
>
> Find Y! Groups > <http://us.ard.yahoo.com/SIG=13o61o8m8/M=493064.12117566.12537396.8674578/D=grplch/S=1705023865:NC/Y=YAHOO/EXP=1214148539/L=/B=h9UXAELaX.w-/J=1214141339005490/A=5170409/R=0/SIG=11gfoiqic/*http://advision.webevents.yahoo.com/craftsman/>
>
> on Lawn & garden,
>
> homes and autos.
>
> Cat Fanatics
>
> on Yahoo! Groups > <http://us.ard.yahoo.com/SIG=13oq6o4ob/M=493064.12016263.12445670.8674578/D=grplch/S=1705023865:NC/Y=YAHOO/EXP=1214148539/L=/B=iNUXAELaX.w-/J=1214141339005490/A=4836040/R=0/SIG=11olbte0b/*http://advision.webevents.yahoo.com/catzone/index.html>
>
> Find people who are
>
> crazy about cats.
>
> .
>
>

Carl Lumma wrote:
> Graham wrote:
>> I don't think about umpteen different systems. I think >> about different mappings from JI, usually 11-limit. A lot >> of the same rules carry over. Don't you think in the >> 7-limit with your barbershop experience?
> > The harmony isn't the thing the system, the scale is.
> Barbershop's still diatonic.

Pajara isn't a scale, it's a way of mapping 7-limit JI to a simpler system. What happens if you try and convert 7-limit barbershop to pajara? This is an example of where notated examples would be useful because I'm not sure how the discussion can progress without them.

Graham

Carl Lumma wrote:
> Graham wrote:
>> You know as well as I do the contribution made to Jazz and >> other musics by musicians who couldn't read notation. If >> you prefer to write that's fine and you need the tools but >> composition's still a long way from manipulation of notation.
> > It requires an instrument. A bass guitar is an instrument,
> a Tonal Plexus is an instrument, and a piece of paper and a
> pencil is an instrument in the hands of those who can hear
> out notation. It's just pulp otherwise.

Rubbish. You can use an instrument to play the music.

> The point is to create the music, not record it. You may
> be thinking of notation as a way to record music. That's
> not how I think of it and if that's all notation was it
> would be completely obsolete today.

That depends on what you mean by "notation" and "record".

>>>> I seem to be the only one in these >>>> parts who cares about logical keyboard mappings.
>>> On these lists or where you're staying now?
>> Either but the latter's hardly surprising.
> > I take it you don't consider sagittal logical, or the
> generalized MOS notation I (and to some extent Herman Miller)
> advocate.

I don't consider either sagittal or whatever MOS notation you refer to to be keyboard mappings since you mention it. But, if you're going to put that remark under the microscope, I can see that it was too hasty. There are people who talk about how nice it would be to have generalized keyboards but by and large don't put their money where their mouths are. I'm the only one I've noticed who thinks about how notations relate to alternative 7+5 mappings.

Graham

Actually Erv mapped out quite a few of his tubes in 5-7 patterns. in fact at his house, he doesn't play on the bosanquet as much.
http://anaphoria.com/patent2.PDF for example.
The Japanese have been very fond of the 5-7 relation for very long. In 12 tone systems, it is still my favorite preoccupation.
I use the 11 different forms of 7 tones scales used in India which gives me 19 different pentatonics. how these interact is quite a bit of material to work with which at some point i will put up.

/^_,',',',_ //^ /Kraig Grady_ ^_,',',',_
Mesotonal Music from:
_'''''''_ ^North/Western Hemisphere: North American Embassy of Anaphoria Island <http://anaphoria.com/>

_'''''''_ ^South/Eastern Hemisphere:
Austronesian Outpost of Anaphoria <http://anaphoriasouth.blogspot.com/>

',',',',',',',',',',',',',',',',',',',',',',',',',',',',',

Graham Breed wrote:
>
> Carl Lumma wrote:
> > Graham wrote:
> >> You know as well as I do the contribution made to Jazz and
> >> other musics by musicians who couldn't read notation. If
> >> you prefer to write that's fine and you need the tools but
> >> composition's still a long way from manipulation of notation.
> >
> > It requires an instrument. A bass guitar is an instrument,
> > a Tonal Plexus is an instrument, and a piece of paper and a
> > pencil is an instrument in the hands of those who can hear
> > out notation. It's just pulp otherwise.
>
> Rubbish. You can use an instrument to play the music.
>
> > The point is to create the music, not record it. You may
> > be thinking of notation as a way to record music. That's
> > not how I think of it and if that's all notation was it
> > would be completely obsolete today.
>
> That depends on what you mean by "notation" and "record".
>
> >>>> I seem to be the only one in these
> >>>> parts who cares about logical keyboard mappings.
> >>> On these lists or where you're staying now?
> >> Either but the latter's hardly surprising.
> >
> > I take it you don't consider sagittal logical, or the
> > generalized MOS notation I (and to some extent Herman Miller)
> > advocate.
>
> I don't consider either sagittal or whatever MOS notation
> you refer to to be keyboard mappings since you mention it.
> But, if you're going to put that remark under the
> microscope, I can see that it was too hasty. There are
> people who talk about how nice it would be to have
> generalized keyboards but by and large don't put their money
> where their mouths are. I'm the only one I've noticed who
> thinks about how notations relate to alternative 7+5 mappings.
>
> Graham
>
>

At 06:30 AM 6/23/2008, you wrote:
>Carl Lumma wrote:
>> Graham wrote:
>>> I don't think about umpteen different systems. I think
>>> about different mappings from JI, usually 11-limit. A lot
>>> of the same rules carry over. Don't you think in the
>>> 7-limit with your barbershop experience?
>>
>> The harmony isn't the thing the system, the scale is.
>> Barbershop's still diatonic.
>
>Pajara isn't a scale,

Now now, let's not say things like this. We both know
we both know what pajara is and is not. I said
'generalized MOS notation I advocate' or some such.

>What happens if you try and convert 7-limit barbershop
>to pajara?

It would probably sound surprisingly normal, since 12 is a
pajara system. But of course I meant the decatonic
scales in pajara.

-Carl

At 06:40 AM 6/23/2008, you wrote:
>Carl Lumma wrote:
>> Graham wrote:
>>> You know as well as I do the contribution made to Jazz and
>>> other musics by musicians who couldn't read notation. If
>>> you prefer to write that's fine and you need the tools but
>>> composition's still a long way from manipulation of notation.
>>
>> It requires an instrument. A bass guitar is an instrument,
>> a Tonal Plexus is an instrument, and a piece of paper and a
>> pencil is an instrument in the hands of those who can hear
>> out notation. It's just pulp otherwise.
>
>Rubbish. You can use an instrument to play the music.

What music?

>> The point is to create the music, not record it. You may
>> be thinking of notation as a way to record music. That's
>> not how I think of it and if that's all notation was it
>> would be completely obsolete today.
>
>That depends on what you mean by "notation" and "record".

Record in the most generic sense, and notation in the
must generic sense.

-Carl

On Sat, Jun 21, 2008 at 10:36 PM, Graham Breed <gbreed@...> wrote:
> Steve Morris wrote:
>
> > Carl is particularly on point when he says that music notation is a
> > language. It is not just LIKE a language, it IS a language. Like math
> > it is a specialized language.
>
> Sure, or maybe it's a language like dance steps are a
> language. It isn't a language like English is a language
> and it isn't anywhere near as difficult to learn.

Not!

Music notation is closer to written English than you choose to admit.
It is symbol based and has grammar and syntax. Perhaps I wasn't clear
enough. I meant language as a linguist would define it and I believe
that Carl was using the word in the same sense. Linguists consider
both math notation and music notations (and written choreography for
that matter) to be language. (I did not mean language in the fuzzy
right brain definition of anything that communicates, as in dance
steps.) These specialty languages are processed in the same parts of
the brain as written English is and by roughly the same rules.

Music notation is quite difficult to learn, something largely
forgotten by people who learned it so many years ago that it has
become second nature. As proof many times many people stop studying
music because they can not get through the language barrier. Learning
it is a boot camp grunt work kind of arbitrary memorization effort, a
barrier that insiders (musicians) are secretly pleased to have gotten
past and often equally pleased that others can't because it proves
their own superiority. If you want to know if a language is hard to
learn you don't ask the people who are fluent. In my experience it is
only the fluent sight reading musician that seriously suggests that
music notation is easy to learn which you have to admit is a bit self
serving. It is equivalent to saying "I got here and I say it was easy,
you didn't which makes you stupid or lazy."

The agreed fact that it isn't as difficult to learn as English doesn't
prove anything about difficulty. Evidence suggests that most high
school grads never learn to write or read English adequately. Writing
(and reading) fluent English is a tough learn. There are lots of
things that are easier to learn than English that are plenty hard to
learn.

By any objective definition standard music notation is a horrible
language poorly suited to discuss music theory. (It is barely adequate
for performance.) It might even be as horrible as English itself
although that's probably a stretch.

Fortunately the brain is a wonderful thing and used to working around
such barriers.

Regards,

Barbershop Steve

Steve Morris wrote:
> On Sat, Jun 21, 2008 at 10:36 PM, Graham Breed <gbreed@...> wrote:
>> Steve Morris wrote:
>>
>>> Carl is particularly on point when he says that music notation is a
>>> language. It is not just LIKE a language, it IS a language. Like math
>>> it is a specialized language.
>> Sure, or maybe it's a language like dance steps are a
>> language. It isn't a language like English is a language
>> and it isn't anywhere near as difficult to learn.
> > Not!

Oh, this thread's back.

> Music notation is closer to written English than you choose to admit.
> It is symbol based and has grammar and syntax. Perhaps I wasn't clear
> enough. I meant language as a linguist would define it and I believe
> that Carl was using the word in the same sense. Linguists consider
> both math notation and music notations (and written choreography for
> that matter) to be language. (I did not mean language in the fuzzy
> right brain definition of anything that communicates, as in dance
> steps.) These specialty languages are processed in the same parts of
> the brain as written English is and by roughly the same rules.

Written English has a lot more than grammar and syntax. It has vocabulary -- the largest of any language, as it happens -- and semantics, among other things. Hence written English is vastly more complex than musical notation and comparisons between the two aren't really helpful for making microtonal music.

Now, you mention linguistics, which is an interesting sidetrack. Do you have a reference for this consideration? Without trying very hard I found a passage written by linguists that's much more to my liking:

"...language is a uniquely human phenomenon. It is lodged in human brains; it is passed on from one generation to the next; it is intimately bound up with the forms of human thought. Unlike a specialized system like arithmetic, it serves a vast range of communicative needs—from getting your neighbor to keep the weeds down, to reporting simple facts, telling jokes, making declarations of love, or praying to a deity. And of course it functions in the midst of complex societies, not just as a means of communication, but as a marker of social identity—a sign of membership in a social class, ethnic group, or nation."

-- Geoff Nunberg and Tom Wasow
http://www.lsadc.org/info/ling-fields-overview.cfm

Some of that does apply to music, even written music, but not to music notation.

> Music notation is quite difficult to learn, something largely
> forgotten by people who learned it so many years ago that it has
> become second nature. As proof many times many people stop studying
> music because they can not get through the language barrier. Learning
> it is a boot camp grunt work kind of arbitrary memorization effort, a
> barrier that insiders (musicians) are secretly pleased to have gotten
> past and often equally pleased that others can't because it proves
> their own superiority. If you want to know if a language is hard to
> learn you don't ask the people who are fluent. In my experience it is
> only the fluent sight reading musician that seriously suggests that
> music notation is easy to learn which you have to admit is a bit self
> serving. It is equivalent to saying "I got here and I say it was easy,
> you didn't which makes you stupid or lazy."

I knew nothing about numerical notation (简谱) three years ago and I haven't made any special effort to learn it but I still say it's easy to learn.

Some people may find music notation of whatever kind difficult to learn. Probably those people wouldn't be looking for a score editor. Some people do find reading their mother tongue difficult to learn. Maybe they're thick, lazy, badly taught, or dyslexic. People interested in composition software using staff notation for microtonal music are likely smart, self-motivated, and already comfortable with staff notation. I'd expect them to pick up a new set of pitches relatively quickly (certainly within a few years) without needing a fancy all-singing, all-dancing score editor with a GUI and some kind of audio support. Some grunt work may be required -- like ear training or counterpoint exercises.

> The agreed fact that it isn't as difficult to learn as English doesn't
> prove anything about difficulty. Evidence suggests that most high
> school grads never learn to write or read English adequately. Writing
> (and reading) fluent English is a tough learn. There are lots of
> things that are easier to learn than English that are plenty hard to
> learn.

That music notation isn't as difficult to learn as English proves that music notation isn't as difficult to learn as English. Therefore the original comparison of English and music notation in terms of being difficult to learn is flawed.

The point isn't about learning music notation, anyway. Only learning a new way of writing pitch in a notation system you already know. That's actually very simple. How simple depends on the pitch system you're notating. Of course, Carl makes it look harder by mixing up the notation with the pitch system itself.

> By any objective definition standard music notation is a horrible
> language poorly suited to discuss music theory. (It is barely adequate
> for performance.) It might even be as horrible as English itself
> although that's probably a stretch.

What do you mean by "standard music notation"? The original proposal was specifically about a score editor for staff notation. If you think this about staff notation, do you agree that it isn't the best choice of notation for composition software? The proposal (which seems to have gone off-list) was to write a score editor from scratch to be a microtonal composition tool. One reason I caution against this is that staff notation's a very difficult system to typeset. Somehow the complexity of staff notation has become an argument in favour of it.

If you have a better proposal then we can look at it. Currently the standard way of making microtonal music with a computer is to use a sequencer with a piano roll view, or some other system that doesn't assume a particular scale. Blue already has a microtonal piano roll. Chuck's JI software has more chance of achieving usability than a brand new microtonal score editor.

Incidentally, there's a score editor called NoteEdit that's part of the KDE project. I don't know how good it is but maybe it's a place to start. It'll likely become cross platform along with the rest of KDE.

> Fortunately the brain is a wonderful thing and used to working around
> such barriers.

There are reasons for using custom notations, including variants of staff notation. There are enough people out there whose brains have worked around the barriers and I, at least, can't think of a better system. I maintain that limited applications like MicroABC are good enough for now. Better microtonal support from score editors will come when there's better microtonal music to be supported.

Graham

Graham wrote:
>Written English has a lot more than grammar and syntax. It
>has vocabulary -- the largest of any language, as it happens
>-- and semantics, among other things. Hence written English
>is vastly more complex than musical notation

You can hardly measure the difficulty of a language by the
size of its vocabulary (most English speakers use about the
same number of words and speakers of any other major
European language). How do you even define a 'word' of
notated music?

>Some people may find music notation of whatever kind
>difficult to learn.

If we're talking about sightreading, it has a lower success
rate than natural languages or natural second languages
(and it is also easiest to learn at the same age as natural
language literacy).

>The point isn't about learning music notation, anyway. Only
>learning a new way of writing pitch in a notation system you
>already know.

And rhythm. I believe I started this thread, and I believe
I was talking about a generalized notation editor, not strictly
a microtonal one.

>That's actually very simple.

Let's see and hear the music, then.

>If you have a better proposal then we can look at it.
>Currently the standard way of making microtonal music with a
>computer is to use a sequencer with a piano roll view, or
>some other system that doesn't assume a particular scale.
>Blue already has a microtonal piano roll.

Piano rolls don't assume particular scales -- so how is
Blue's microtonal?

>> Fortunately the brain is a wonderful thing and used to working around
>> such barriers.
>
>There are reasons for using custom notations, including
>variants of staff notation. There are enough people out
>there whose brains have worked around the barriers and I, at
>least, can't think of a better system. I maintain that
>limited applications like MicroABC are good enough for now.

Good enough for whom?

-Carl

--- In MakeMicroMusic@yahoogroups.com, Carl Lumma <carl@...> wrote:
> Good enough for whom?

No one that I work or perform with, that's for sure.

--- In MakeMicroMusic@yahoogroups.com, "Jon Szanto" <jszanto@...> wrote:
>
> --- In MakeMicroMusic@yahoogroups.com, Carl Lumma <carl@> wrote:
> > Good enough for whom?
>
> No one that I work or perform with, that's for sure.

I'll second that. I really would like to see a microtonal Sibelius or
Finale... please!
>

--- In MakeMicroMusic@yahoogroups.com, "Robin Perry" <jinto83@...> wrote:
> I'll second that. I really would like to see a microtonal Sibelius or
> Finale... please!

Unfortunately, "please" (or pleas) won't cut it. It's going to take
money and a viable user base. I liken it to the dilemma facing
hydrogen vehicles: they won't build hydro stations until there are
customers to use them, and it hardly makes sense to have a vehicle
that can't find any place to fuel up.

We're in the Middle Ages of the microtonal revolution. Each man/woman
to his own.

Cheers,
Jon

Robin Perry wrote:
> --- In MakeMicroMusic@yahoogroups.com, "Jon Szanto" <jszanto@...> wrote:
>> --- In MakeMicroMusic@yahoogroups.com, Carl Lumma <carl@> wrote:
>>> Good enough for whom?
>> No one that I work or perform with, that's for sure.
> > I'll second that. I really would like to see a microtonal Sibelius or
> Finale... please!

In what sense are Sibelius and Finale not microtonal?

Graham

Jon Szanto wrote:
> --- In MakeMicroMusic@yahoogroups.com, Carl Lumma <carl@...> wrote:
[MicroABC]
>> Good enough for whom?
> > No one that I work or perform with, that's for sure.

Wow, that's a real kick in the teeth!
Thanks! Now I know my effort's been wasted. Do you have any more constructive criticism?

Graham

--- In MakeMicroMusic@yahoogroups.com, Graham Breed <gbreed@...> wrote:
> Wow, that's a real kick in the teeth!

Not meant to be in the least. More after the cut...

> Thanks! Now I know my effort's been wasted. Do you have
> any more constructive criticism?

Well, I sure as hell hope I haven't missed a *whole* lot of
developments, and correct me (and castigate me) if necessary. I was
under the impression that MicroABC was an ascii/text-based notation
system, and put out 'scores' or 'parts' that consisted of a
language/code made up of ascii characters. I am under the impression
that it does *not* utilize anything resembling standard musical staff
notation.

If my assumptions are correct, then I stand by my statements: I work
in music, and have all my life. I expect that myself, and all my
professional colleagues, will be looking at some variant of
musical/staff notation, not some text-based data. Within widely
ranging parameters, I've had to accomodate a LOT of differing
notations, but there is no way that you will ever get a large body of
practicing players to switch to a notation system that is both alien
and quite at a disadvantage for reading.

I realize that some variants of this work in folk music circles and
similar situations. I don't think it is comparable to what passes for
modern music notational systems.

Again, Graham, it wasn't meant as an insult, but a very plain way of
saying that it is far, far from adoptable by the large body of
practicing, performing, working musicians. And I am very willing to
accept the arrows and swords if somewhere along the line I've missed
an important development of MicroABC into staff notation.

I also think Saggital is never going to fly. What do I know?

Best,
Jon

Jon Szanto wrote:
> --- In MakeMicroMusic@yahoogroups.com, Graham Breed <gbreed@...> wrote:
>> Wow, that's a real kick in the teeth!
> > Not meant to be in the least. More after the cut...

I'm trying to focus this thread on something practical. It didn't seem to be going anywhere last time round but now it's back so let's see what we can make of it.

>> Thanks! Now I know my effort's been wasted. Do you have >> any more constructive criticism?
> > Well, I sure as hell hope I haven't missed a *whole* lot of
> developments, and correct me (and castigate me) if necessary. I was
> under the impression that MicroABC was an ascii/text-based notation
> system, and put out 'scores' or 'parts' that consisted of a
> language/code made up of ascii characters. I am under the impression
> that it does *not* utilize anything resembling standard musical staff
> notation.

MicroABC as a format is an ASCII-based notation system. MicroABC as an application produces staff notation in PostScript form. So yes, if you were under the impression that staff notation didn't come out, you were wrong. It isn't quite standard because obviously it's microtonal. But it is pretty good.

> If my assumptions are correct, then I stand by my statements: I work
> in music, and have all my life. I expect that myself, and all my
> professional colleagues, will be looking at some variant of
> musical/staff notation, not some text-based data. Within widely
> ranging parameters, I've had to accomodate a LOT of differing
> notations, but there is no way that you will ever get a large body of
> practicing players to switch to a notation system that is both alien
> and quite at a disadvantage for reading.

Right, so they'll be looking at staff notation. The composers may have to deal with the ASCII but the performers don't need to worry about that.

> I realize that some variants of this work in folk music circles and
> similar situations. I don't think it is comparable to what passes for
> modern music notational systems.

A lot more folk musicians are using numerical notation, and ABC is sort of like that but uses letters and some other differences that are either arbitrary or make it look more like staff notation. But I'm rambling ...

> Again, Graham, it wasn't meant as an insult, but a very plain way of
> saying that it is far, far from adoptable by the large body of
> practicing, performing, working musicians. And I am very willing to
> accept the arrows and swords if somewhere along the line I've missed
> an important development of MicroABC into staff notation.

Hopefully, the whole point of MicroABC being to produce staff notation and MIDI counts as such a development :-)

> I also think Saggital is never going to fly. What do I know?

Once software supports Saggital, especially if it's free software that you can go in and hack, it should be easy to adapt it to use different symbols. All you need is a different font and a mapping for it. Hudson's got decimal and tripod notation working for me so I expect he'll do the same for whatever accidentals you want if you have the font.

LilyPond only works with quartertones but some of us are looking at sagittal support. It may mean a custom build but I've looked at the code and it's not at all bad. So I can promise you that by the end of the summer. Again, once Sagittal works it's that bit easier for other fonts.

Graham

I haven't used Finale. I've only heard that it's as cumbersome as
Sibelius is. I don't like to have to mess around with midi channels
to get a score to output correctly. Just don't wanna!

--- In MakeMicroMusic@yahoogroups.com, Graham Breed <gbreed@...> wrote:
>
> Robin Perry wrote:
> > --- In MakeMicroMusic@yahoogroups.com, "Jon Szanto" <jszanto@> wrote:
> >> --- In MakeMicroMusic@yahoogroups.com, Carl Lumma <carl@> wrote:
> >>> Good enough for whom?
> >> No one that I work or perform with, that's for sure.
> >
> > I'll second that. I really would like to see a microtonal Sibelius or
> > Finale... please!
>
> In what sense are Sibelius and Finale not microtonal?
>
>
> Graham
>

Carl Lumma wrote:
> Graham wrote:
>> Written English has a lot more than grammar and syntax. It >> has vocabulary -- the largest of any language, as it happens >> -- and semantics, among other things. Hence written English >> is vastly more complex than musical notation
> > You can hardly measure the difficulty of a language by the
> size of its vocabulary (most English speakers use about the
> same number of words and speakers of any other major
> European language). How do you even define a 'word' of
> notated music?

Indeed you can't. I'm not the one calling music notation a language so I don't need to think about what the words are.

>> Some people may find music notation of whatever kind >> difficult to learn.
> > If we're talking about sightreading, it has a lower success
> rate than natural languages or natural second languages
> (and it is also easiest to learn at the same age as natural
> language literacy).

I'm not talking about sight reading. If you have some references to support that statement they might be interesting to look at. Sight reading's off topic for composition software, though.

>> The point isn't about learning music notation, anyway. Only >> learning a new way of writing pitch in a notation system you >> already know.
> > And rhythm. I believe I started this thread, and I believe
> I was talking about a generalized notation editor, not strictly
> a microtonal one.

I don't believe you did start the thread. You mentioned a rhythmic notation, and that's been noted. It's not directly relevant to the topic of making microtonal music.

>> That's actually very simple.
> > Let's see and hear the music, then.

There are a few of us who've worked with alternative notations. You know where you can hear my music. I've photographed some notation so you can see it:

http://x31eq.com/music/wisofl.jpg

About a megabyte.

>> If you have a better proposal then we can look at it. >> Currently the standard way of making microtonal music with a >> computer is to use a sequencer with a piano roll view, or >> some other system that doesn't assume a particular scale. >> Blue already has a microtonal piano roll.
> > Piano rolls don't assume particular scales -- so how is
> Blue's microtonal?

Piano rolls do assume a particular scale -- usually 12 notes to the octave. I don't know what Blue does. Shall we look at the website?

http://www.csounds.com/stevenyi/blue/

The feature point is "Microtonal PianoRoll allows editing notes using any Scala scale".

>>> Fortunately the brain is a wonderful thing and used to working around
>>> such barriers.
>> There are reasons for using custom notations, including >> variants of staff notation. There are enough people out >> there whose brains have worked around the barriers and I, at >> least, can't think of a better system. I maintain that >> limited applications like MicroABC are good enough for now.
> > Good enough for whom?

Good enough for people who want staff notation with sagittal accidentals, or different numbers of notes to the octave, and some other variants of staff notation.

Graham

since there is music without melody or harmony, language is not whatmusic is. nor can we say that music is saying something, just becausepeople often extract contrary meanings out of it

,',',',Kraig Grady,',',',
'''''''North/Western Hemisphere:
North American Embassy of Anaphoria island
'''''''South/Eastern Hemisphere:
Austronesian Outpost of Anaphoria
',',',',',',',',',',',',',',',',',',',',',',',',',',',',',

-----Original Message-----
From: Graham Breed [mailto:gbreed@...]
Sent: Monday, July 7, 2008 12:01 AM
To: MakeMicroMusic@yahoogroups.com
Subject: Re: [MMM] Re: Composition Software...?

Carl Lumma wrote:
> Graham wrote:
>> Written English has a lot more than grammar and syntax. It
>> has vocabulary -- the largest of any language, as it happens
>> -- and semantics, among other things. Hence written English
>> is vastly more complex than musical notation
>
> You can hardly measure the difficulty of a language by the
> size of its vocabulary (most English speakers use about the
> same number of words and speakers of any other major
> European language). How do you even define a 'word' of
> notated music?

Indeed you can't. I'm not the one calling music notation a
language so I don't need to think about what the words are.

>> Some people may find music notation of whatever kind
>> difficult to learn.
>
> If we're talking about sightreading, it has a lower success
> rate than natural languages or natural second languages
> (and it is also easiest to learn at the same age as natural
> language literacy).

I'm not talking about sight reading. If you have some
references to support that statement they might be
interesting to look at. Sight reading's off topic for
composition software, though.

>> The point isn't about learning music notation, anyway. Only
>> learning a new way of writing pitch in a notation system you
>> already know.
>
> And rhythm. I believe I started this thread, and I believe
> I was talking about a generalized notation editor, not strictly
> a microtonal one.

I don't believe you did start the thread. You mentioned a
rhythmic notation, and that's been noted. It's not directly
relevant to the topic of making microtonal music.

>> That's actually very simple.
>
> Let's see and hear the music, then.

There are a few of us who've worked with alternative
notations. You know where you can hear my music. I've
photographed some notation so you can see it:

http://x31eq.com/music/wisofl.jpg

About a megabyte.

>> If you have a better proposal then we can look at it.
>> Currently the standard way of making microtonal music with a
>> computer is to use a sequencer with a piano roll view, or
>> some other system that doesn't assume a particular scale.
>> Blue already has a microtonal piano roll.
>
> Piano rolls don't assume particular scales -- so how is
> Blue's microtonal?

Piano rolls do assume a particular scale -- usually 12 notes
to the octave. I don't know what Blue does. Shall we look
at the website?

http://www.csounds.com/stevenyi/blue/

The feature point is "Microtonal PianoRoll allows editing
notes using any Scala scale".

>>> Fortunately the brain is a wonderful thing and used to working around
>>> such barriers.
>> There are reasons for using custom notations, including
>> variants of staff notation. There are enough people out
>> there whose brains have worked around the barriers and I, at
>> least, can't think of a better system. I maintain that
>> limited applications like MicroABC are good enough for now.
>
> Good enough for whom?

Good enough for people who want staff notation with sagittal
accidentals, or different numbers of notes to the octave,
and some other variants of staff notation.

Graham

[Non-text portions of this message have been removed]

--- In MakeMicroMusic@yahoogroups.com, Graham Breed <gbreed@...> wrote:
>
> Chuckk Hubbard wrote:

> > Not sure, I haven't had much luck in the past. Csound does it all
> > internally and just puts out the sound, so good for it. I'll have a
> > default soundfont instrument for those who want to use soundfonts and
> > not bother with Csound code, and then the option to manually edit
> > one's own Csound orchestra. Anyway, the soundfont specification would
> > take months to sort out and code into my program, so, since I'm using
> > Csound for audio anyway, I'll be using it to load soundfonts too.
>
> Er, if the Csound folks can embed FluidSynth can't you embed
> it directly as well?

Heh, well, surely it's possible, but there are lots of things the
Csound folks have done that I don't care to invest my time in doing.
I'm more interested in being a musician than a programmer, especially
when it comes to functions I don't expect to use much myself, like
soundfonts.

-Chuckk

music is perhaps the uber-gibberish, that can somehow seem laden with emotional or intellectual (or billboard-traffic-sign?) content...

whether that content is supplied by the musician or the listener is one of the unanswerable questions.

music notation is a convenience of bureaucrats.

hee.

now back to that long overdue Finale document I've been slaving over...what a nightmare! I could write it by hand in one tenth the time, but no one would read it, because it doesn't look "official". come to think of it, no one reads it, anyway...

hee.

kraiggrady@... wrote:
since there is music without melody or harmony, language is not whatmusic is. nor can we say that music is saying something, just becausepeople often extract contrary meanings out of it

,',',',Kraig Grady,',',',
'''''''North/Western Hemisphere:
North American Embassy of Anaphoria island
'''''''South/Eastern Hemisphere:
Austronesian Outpost of Anaphoria
',',',',',',',',',',',',',',',',',',',',',',',',',',',',',

-----Original Message-----
From: Graham Breed [mailto:gbreed@...]
Sent: Monday, July 7, 2008 12:01 AM
To: MakeMicroMusic@yahoogroups.com
Subject: Re: [MMM] Re: Composition Software...?

Carl Lumma wrote:
> Graham wrote:
>> Written English has a lot more than grammar and syntax. It
>> has vocabulary -- the largest of any language, as it happens
>> -- and semantics, among other things. Hence written English
>> is vastly more complex than musical notation
>
> You can hardly measure the difficulty of a language by the
> size of its vocabulary (most English speakers use about the
> same number of words and speakers of any other major
> European language). How do you even define a 'word' of
> notated music?

Indeed you can't. I'm not the one calling music notation a
language so I don't need to think about what the words are.

>> Some people may find music notation of whatever kind
>> difficult to learn.
>
> If we're talking about sightreading, it has a lower success
> rate than natural languages or natural second languages
> (and it is also easiest to learn at the same age as natural
> language literacy).

I'm not talking about sight reading. If you have some
references to support that statement they might be
interesting to look at. Sight reading's off topic for
composition software, though.

>> The point isn't about learning music notation, anyway. Only
>> learning a new way of writing pitch in a notation system you
>> already know.
>
> And rhythm. I believe I started this thread, and I believe
> I was talking about a generalized notation editor, not strictly
> a microtonal one.

I don't believe you did start the thread. You mentioned a
rhythmic notation, and that's been noted. It's not directly
relevant to the topic of making microtonal music.

>> That's actually very simple.
>
> Let's see and hear the music, then.

There are a few of us who've worked with alternative
notations. You know where you can hear my music. I've
photographed some notation so you can see it:

http://x31eq.com/music/wisofl.jpg

About a megabyte.

>> If you have a better proposal then we can look at it.
>> Currently the standard way of making microtonal music with a
>> computer is to use a sequencer with a piano roll view, or
>> some other system that doesn't assume a particular scale.
>> Blue already has a microtonal piano roll.
>
> Piano rolls don't assume particular scales -- so how is
> Blue's microtonal?

Piano rolls do assume a particular scale -- usually 12 notes
to the octave. I don't know what Blue does. Shall we look
at the website?

http://www.csounds.com/stevenyi/blue/

The feature point is "Microtonal PianoRoll allows editing
notes using any Scala scale".

>>> Fortunately the brain is a wonderful thing and used to working around
>>> such barriers.
>> There are reasons for using custom notations, including
>> variants of staff notation. There are enough people out
>> there whose brains have worked around the barriers and I, at
>> least, can't think of a better system. I maintain that
>> limited applications like MicroABC are good enough for now.
>
> Good enough for whom?

Good enough for people who want staff notation with sagittal
accidentals, or different numbers of notes to the octave,
and some other variants of staff notation.

Graham

[Non-text portions of this message have been removed]

[Non-text portions of this message have been removed]

Graham wrote:
>> If we're talking about sightreading, it has a lower success
>> rate than natural languages or natural second languages
>> (and it is also easiest to learn at the same age as natural
>> language literacy).
>
>I'm not talking about sight reading. If you have some
>references to support that statement they might be
>interesting to look at. Sight reading's off topic for
>composition software, though.

I'd be happy to provide references, but we won't make any
progress as long as you are believing stuff like this last
sentence.

>>> The point isn't about learning music notation, anyway. Only
>>> learning a new way of writing pitch in a notation system you
>>> already know.
>>
>> And rhythm. I believe I started this thread, and I believe
>> I was talking about a generalized notation editor, not strictly
>> a microtonal one.
>
>I don't believe you did start the thread. You mentioned a
>rhythmic notation, and that's been noted. It's not directly
>relevant to the topic of making microtonal music.

Mike and I started this thread. The point is to provide whatever
features microtonal composers may want. The best way to do that
is to allow them to specify a notation system.

>>> That's actually very simple.
>>
>> Let's see and hear the music, then.
>
>There are a few of us who've worked with alternative
>notations. You know where you can hear my music. I've
>photographed some notation so you can see it:
>
> http://x31eq.com/music/wisofl.jpg
>
>About a megabyte.

Nice! Where can I hear it? I don't see it on your music
page. By the way, every page on x31eq.com is currently
flashing a bunch of casino stuff immediately after
loading (FF3).

>>> If you have a better proposal then we can look at it.
>>> Currently the standard way of making microtonal music with a
>>> computer is to use a sequencer with a piano roll view, or
>>> some other system that doesn't assume a particular scale.
>>> Blue already has a microtonal piano roll.
>>
>> Piano rolls don't assume particular scales -- so how is
>> Blue's microtonal?
>
>Piano rolls do assume a particular scale -- usually 12 notes
>to the octave.

Huh?

>I don't know what Blue does. Shall we look
>at the website?
>
>http://www.csounds.com/stevenyi/blue/
>
>The feature point is "Microtonal PianoRoll allows editing
>notes using any Scala scale".

I can do that with any piano roll. Many of them may be limited
to 128 notes; if that's a limitation that matters Blue may be
worth looking at. I installed it several years ago and I thought
it was a flaming piece of junk, but maybe it's been improved
since then.

>>> I maintain that
>>> limited applications like MicroABC are good enough for now.
>>
>> Good enough for whom?
>
>Good enough for people who want staff notation with sagittal
>accidentals, or different numbers of notes to the octave,
>and some other variants of staff notation.

If these systems were good enough, people wouldn't spend
hundreds of dollars per license on software like Finale and
Sibelius, which you admit to never having used. We've already
heard from some folks here about how they want microtonal
versions of these programs.

-Carl

Kraig wrote:
>since there is music without melody or harmony, language is not
>whatmusic is. nor can we say that music is saying something, just
>becausepeople often extract contrary meanings out of it

We use many of the same processes in our brains to understand
music and language. One can say music is a synthetic device
that stimulates these areas, much as cocaine stimulates areas
designed to help us learn positive behaviors. Frank Herbert
was very perceptive with his "semuta music".

More germane to the thread here, music notation is a language
with reading and writing skills, though Graham doesn't seem to
think the former are important.

-Carl

> I haven't used Finale. I've only heard that it's as cumbersome as
> Sibelius is. I don't like to have to mess around with midi channels
> to get a score to output correctly. Just don't wanna!

Try using Fractal Tune Smithy to do the MIDI management for you. I've been
using it successfully with Finale, etc for quite a while to do microtonal
work. You don't need to send your micro output on different channels.

Rick

Rick,

I've never used Fractal Tune Smithy and a brief look at its home page does
not explain why a fractal algorithm generator would help fix finale midi
output. Could you elucidate please? This suggestion is a little too terse
for my tiny brain to grasp.

Steve

On Mon, Jul 7, 2008 at 12:02 PM, Rick McGowan <rick@...> wrote:

> > I haven't used Finale. I've only heard that it's as cumbersome as
> > Sibelius is. I don't like to have to mess around with midi channels
> > to get a score to output correctly. Just don't wanna!
>
> Try using Fractal Tune Smithy to do the MIDI management for you. I've been
> using it successfully with Finale, etc for quite a while to do microtonal
> work. You don't need to send your micro output on different channels.
>
> Rick
>
>

--
Steve Morris
barbershopsteve@...
Bass: Boston Wailers
Bass: Sounds Of Concord
Motto: Old age and treachery will always prevail over youth and skill

[Non-text portions of this message have been removed]

Steve,

> I've never used Fractal Tune Smithy and a brief look at its home page
> does not explain why a fractal algorithm generator would help fix
> finale midi output. Could you elucidate please? This suggestion
> is a little too terse for my tiny brain to grasp.

FTS is much more than just a fractal algorithm generator... It can be used
as a sophisticated MIDI relay, and it has the ability to use ".tun" tuning
files to re-tune channels via pitch-bend. It has remarkably flexible
handling of the channels.

See here:
http://www.robertinventor.com/software/tunesmithy/music.htm

Specifically "microtonal features for composers",
http://www.robertinventor.com/software/tunesmithy/composing.htm

The FTS page also has a pointer to my page on using FTS with GPO and Finale:
http://www.robertinventor.com/software/tunesmithy/composing.htm#large_orchestras

Rick

--- In MakeMicroMusic@yahoogroups.com, "Jon Szanto" <jszanto@...> wrote:
>
> --- In MakeMicroMusic@yahoogroups.com, Carl Lumma <carl@> wrote:
> > Good enough for whom?
>
> No one that I work or perform with, that's for sure.
>

I've never had any problem with any 'limitations' of the abcm2ps
family of programs, and I've used them in countless professional music
situations.

Until you can point out some true limitation of abc in your
experience, I'm afraid I'll have to conclude that you're being an
irrational snob. Ditto Carl.

-A.

Aaron wrote:
>> > Good enough for whom?
>>
>> No one that I work or perform with, that's for sure.
>>
>
>I've never had any problem with any 'limitations' of the abcm2ps
>family of programs, and I've used them in countless professional music
>situations.
>
>Until you can point out some true limitation of abc in your
>experience, I'm afraid I'll have to conclude that you're being an
>irrational snob. Ditto Carl.

I've already explained at some length its limitations in this
thread. Want to make a wisecrack without reading it? I'd say
you're the snob.

-Carl

--- In MakeMicroMusic@yahoogroups.com, Carl Lumma <carl@...> wrote:

> >>> I maintain that
> >>> limited applications like MicroABC are good enough for now.
> >>
> >> Good enough for whom?
> >
> >Good enough for people who want staff notation with sagittal
> >accidentals, or different numbers of notes to the octave,
> >and some other variants of staff notation.
>
> If these systems were good enough, people wouldn't spend
> hundreds of dollars per license on software like Finale and
> Sibelius, which you admit to never having used. We've already
> heard from some folks here about how they want microtonal
> versions of these programs.

In other words--if microtones are so cool, how come no one I know uses
them? ;)

This is an equivalent statement of 'value'.

Maybe the same people who shell out dough for big programs don't know
of, or aren't in the culture of, free or open-source software; maybe
they need or want a really flashy GUI environment and have prejudice
about working in a text-environment, maybe they are buying because
they feel swept up in all the professional hype surrounding a 'product'.

Thankfully, there are enough people who don't get swept up in
statements designed to make people march lock-step to someone's, or
some large group's, idea of a 'standard'--like Harry Partch, to name one.

It seems awfully co-dependant to criticise someone's toolset,
especially when people are getting solid work done with that toolset,
and especially given that these same expensive programs that you
insist are the best don't have the proper functionality yet, while
sometimes their FREE cousins do.

Incidentally, arguably the best music typesetting available by
computer is Lilypond, whose syntax I find horrible, but it looks great
and is free nonetheless.

-AKJ.

--- In MakeMicroMusic@yahoogroups.com, Carl Lumma <carl@...> wrote:
>
> Aaron wrote:
> >> > Good enough for whom?
> >>
> >> No one that I work or perform with, that's for sure.
> >>
> >
> >I've never had any problem with any 'limitations' of the abcm2ps
> >family of programs, and I've used them in countless professional music
> >situations.
> >
> >Until you can point out some true limitation of abc in your
> >experience, I'm afraid I'll have to conclude that you're being an
> >irrational snob. Ditto Carl.
>
> I've already explained at some length its limitations in this
> thread. Want to make a wisecrack without reading it? I'd say
> you're the snob.

I didn't mean it as a wisecrack, but a statement of how it seems you
and in particular, Jon, were disparaging something you've never really
had first hand use of. But, I'll retract it--sorry I name-called and
said you and Jon were snobs. I meant to say "don't be a dum-dum head".

I find it interesting that people focus on attacking tools around
here. Why not celebrate that people can be productive in various
non-standard ways, that people _create_ their own tools? I never
understood the 'herd mentality' of 'no-one who is a REAL musician
doesn't use TOOL-X'. Ridiculous and co-dependant is what it is. Use
whatever the f^&% you want to make or notate music--why should anyone
give a damn? The music at the other end is what matters.

Ok, getting back to the thread--all I could find, Carl, is that you
found microabc limited because it didn't have a GUI and because it
wasn't a real-time system. Fair enough...if those things are what you
need, then don't use microabc.

In my book, abcm2ps (I don't personally use microabc, but it comes
from that) is fine, scales reasonably well, and I have the syntax
pretty well mastered by now that I'm pretty quick at writing
scores---I'm willing to bet I can do it at speed as fast or faster
than some GUI notation software of your choice. I also like that it's
not a huge piece of bloatware.

Also, for my purposes, I'm using it the way Bach used paper and
quill--for, of all crazy things, *notating music*. I don't need a GUI,
and I don't need real-time, but, your mileage may vary.

-AKJ.

--- In MakeMicroMusic@yahoogroups.com, "Aaron Krister Johnson"
<aaron@...> wrote:
> Maybe the same people who shell out dough for big programs don't know
> of, or aren't in the culture of, free or open-source software; maybe
> they need or want a really flashy GUI environment and have prejudice
> about working in a text-environment, maybe they are buying because
> they feel swept up in all the professional hype surrounding a 'product'.
>
> Thankfully, there are enough people who don't get swept up in
> statements designed to make people march lock-step to someone's, or
> some large group's, idea of a 'standard'--like Harry Partch, to name
one.

Aaron, with utmost respect, your position comes with it's own set of
biases and agendas. I, for one, don't value one particular set of
tools over another in general, only that each person find the toolset
that is most applicable to them.

You, and many others, not only have the understanding but the
enthusiasm to work at code level, to get down and play around with the
myriad of install options on the underlying platforms, and more than
happy to put in the time and effort to get all of these pieces to work
together in something that resembles harmony.

Many others would simply like to load an application, get up to speed,
and start working. And, yes, some people like the interface model that
incorporates visual elements that perhaps mimic something they do, or
have done, in the real world, like placing notes on a piece of
manuscript paper. I'm happy that there are ways to do that for them,
and I don't consider them robots or lesser beings because they choose
a GUI-driven path to their muse.

Lastly, touting the free to low-cost nature of the beast is all well
and good. I, myself, have no issue with purchasing a product that has
been crafted with care and includes support when I run into trouble.
IIRC, I managed to find one such product, but damned if I can remember
what it was! :) And I seem to remember that you, yourself, tend to not
give away your music freely, but encourage people to support it
financially, feeling that artists should be compensated for their
work. I concur, at least in some scenarios.

Open-source, price-negligible options are a great thing. They also
aren't the only way to go.

Cheers,
Jon

Yes, I tried the recent free trial of the updated FTS. It can work,
but, it's still too cumbersome for me. I'm a lazy SOB.

--- In MakeMicroMusic@yahoogroups.com, Rick McGowan <rick@...> wrote:
>
> > I haven't used Finale. I've only heard that it's as cumbersome as
> > Sibelius is. I don't like to have to mess around with midi channels
> > to get a score to output correctly. Just don't wanna!
>
> Try using Fractal Tune Smithy to do the MIDI management for you.
I've been
> using it successfully with Finale, etc for quite a while to do
microtonal
> work. You don't need to send your micro output on different channels.
>
> Rick
>

Jon, see my notes below....

--- In MakeMicroMusic@yahoogroups.com, "Jon Szanto" <jszanto@...> wrote:
>
> --- In MakeMicroMusic@yahoogroups.com, Graham Breed <gbreed@> wrote:
> > Wow, that's a real kick in the teeth!
>
> Not meant to be in the least. More after the cut...
>
> > Thanks! Now I know my effort's been wasted. Do you have
> > any more constructive criticism?
>
> Well, I sure as hell hope I haven't missed a *whole* lot of
> developments, and correct me (and castigate me) if necessary. I was
> under the impression that MicroABC was an ascii/text-based notation
> system, and put out 'scores' or 'parts' that consisted of a
> language/code made up of ascii characters. I am under the impression
> that it does *not* utilize anything resembling standard musical staff
> notation.

Correction---abcm2ps, abc, microabc, etc. are ascii-to PostScript
score programs---the input is a text representation of the score, with
a nice syntax at that, and the output is a _printable standard staff
notation score_ in PostScript format (which can be converted to PDF as
well)

> If my assumptions are correct, then I stand by my statements: I work
> in music, and have all my life. I expect that myself, and all my
> professional colleagues, will be looking at some variant of
> musical/staff notation, not some text-based data. Within widely
> ranging parameters, I've had to accomodate a LOT of differing
> notations, but there is no way that you will ever get a large body of
> practicing players to switch to a notation system that is both alien
> and quite at a disadvantage for reading.

Jon, all due respect, but, wow--you should really do more research
next time before attacking what you don't understand.

Considering that you should be encouraging, as 'MMM list mom', people
to use available tools, including free ones, to actually further the
cause of MakingMicroMusic, this is an unfortunate oversight that you
are pushing people into expensive corporate software when people can
start doing this for cheap. Not that one should *avoid* commercial
software either, but abcm2ps and microabc are free tools with
micro-capability, available NOW for free!!!

> I realize that some variants of this work in folk music circles and
> similar situations. I don't think it is comparable to what passes for
> modern music notational systems.

> I also think Saggital is never going to fly. What do I know?

I wouldn't bet so either--I think the font is beautiful, and the work
done on it has been heroic and extraordinary, but I find it too large
and unwieldy (yes, even the reduced set is too large) for the systems
I tend to work with---I use standard sharps, flats, half-sharps and
half-flats so far when I'm notating, and haven't yet run into a problem.

But I want to be careful not to discourage its use....heck, it's there
if you are the kind of composer who needs it, and it looks pretty
complete. I'd say, however, that I'm practical,and I don't think it's
likely that a whole bunch of people are going to actively get
proficient at being trained to sight-read in Sagittal, for instance.
So I wouldn't myself print a Sagittal score. What's wrong with doing
what Johnny Reinhard does, or using HEWM, when you need that kind of
thing (by that I mean 13-limit accidentals and such)

-AKJ

--- In MakeMicroMusic@yahoogroups.com, "Jon Szanto" <jszanto@...> wrote:
>
> --- In MakeMicroMusic@yahoogroups.com, "Aaron Krister Johnson"
> <aaron@> wrote:
> > Maybe the same people who shell out dough for big programs don't know
> > of, or aren't in the culture of, free or open-source software; maybe
> > they need or want a really flashy GUI environment and have prejudice
> > about working in a text-environment, maybe they are buying because
> > they feel swept up in all the professional hype surrounding a
'product'.
> >
> > Thankfully, there are enough people who don't get swept up in
> > statements designed to make people march lock-step to someone's, or
> > some large group's, idea of a 'standard'--like Harry Partch, to name
> one.
>
> Aaron, with utmost respect, your position comes with it's own set of
> biases and agendas. I, for one, don't value one particular set of
> tools over another in general, only that each person find the toolset
> that is most applicable to them.
>
> You, and many others, not only have the understanding but the
> enthusiasm to work at code level, to get down and play around with the
> myriad of install options on the underlying platforms, and more than
> happy to put in the time and effort to get all of these pieces to work
> together in something that resembles harmony.
>
> Many others would simply like to load an application, get up to speed,
> and start working. And, yes, some people like the interface model that
> incorporates visual elements that perhaps mimic something they do, or
> have done, in the real world, like placing notes on a piece of
> manuscript paper. I'm happy that there are ways to do that for them,
> and I don't consider them robots or lesser beings because they choose
> a GUI-driven path to their muse.

I have no problem with GUIs, and often prefer them. I've never
disparaged them or their users as robots, either. I just think it's
important not to go the other way with the prejudice---sometimes text
systems are the most lean and elegant solution to a given problem.

> Lastly, touting the free to low-cost nature of the beast is all well
> and good. I, myself, have no issue with purchasing a product that has
> been crafted with care and includes support when I run into trouble.

Yeah, me too...especially true of instruments.

And having good support and docs are a good thing too....the level of
quality of software support and documentation varies greatly in open
source from piss-poor to excellent (ditto the software itself)

Ironically, the same thing can be said for commerical products (like
my Linksys wireless printing router---terrible piece of junk, poorly
documented and poorly made, and expensive to boot)

Software, like anything, ought be evaluated on a case-by-case basis.

> And I seem to remember that you, yourself, tend to not
> give away your music freely, but encourage people to support it
> financially, feeling that artists should be compensated for their
> work. I concur, at least in some scenarios.

Right now, my music is in a state most resembling 'freeware', meaning,
it's freely downloadable, but I appreciate cash.

Most open source licenses are more virulently anti-capital and
anti-'ownership'. It's good and bad, like anything.

> Open-source, price-negligible options are a great thing. They also
> aren't the only way to go.

Agreed....so let's not dismiss them either, especially not when we
don't really understand them, eh? ;)

> Cheers,
> Jon
>

--- In MakeMicroMusic@yahoogroups.com, "Aaron Krister Johnson"
<aaron@...> wrote:
>
> Jon, see my notes below....
>
> --- In MakeMicroMusic@yahoogroups.com, "Jon Szanto" <jszanto@> wrote:
> >
> > Well, I sure as hell hope I haven't missed a *whole* lot of
> > developments, and correct me (and castigate me) if necessary. I was
> > under the impression that MicroABC was an ascii/text-based notation
> > system, and put out 'scores' or 'parts' that consisted of a
> > language/code made up of ascii characters. I am under the impression
> > that it does *not* utilize anything resembling standard musical staff
> > notation.
>
> Correction---abcm2ps, abc, microabc, etc. are ascii-to PostScript
> score programs---the input is a text representation of the score, with
> a nice syntax at that, and the output is a _printable standard staff
> notation score_ in PostScript format (which can be converted to PDF as
> well)

Yes. You'll note that I explicitly indicated that I was not clear on
the developments for output from abc to staff notation. I explicitly
suggested castigation, and you've supplied it in a couple messages.
Try to keep in mind, however, that discussion/criticism != disparagement.

> Jon, all due respect, but, wow--you should really do more research
> next time before attacking what you don't understand.

A. it wasn't an attack
B. not knowing all the details is not equivalent to "don't understand"

> Considering that you should be encouraging, as 'MMM list mom', people
> to use available tools, including free ones, to actually further the
> cause of MakingMicroMusic, this is an unfortunate oversight that you
> are pushing people into expensive corporate software when people can
> start doing this for cheap.

A. if you read my further message, you realize that I now, as I always
have, advocate a pluralistic viewpoint, deeply rooted in "to each his own"
B. I'm "pushing" no one, and I was the first person on this list to
put together a scenario for creating microtonal music, on your
computer, with entirely free components. And that was a long time ago.
C. the "corporate software" rants got old a long time ago, Aaron.
Since I don't proselytize for said products, pleez to not paint me wif
dat brush. ;)

> Not that one should *avoid* commercial
> software either, but abcm2ps and microabc are free tools with
> micro-capability, available NOW for free!!!

Commendable.

> I wouldn't bet so either--I think the font is beautiful, and the work
> done on it has been heroic and extraordinary, but I find it too large
> and unwieldy (yes, even the reduced set is too large) for the systems
> I tend to work with---I use standard sharps, flats, half-sharps and
> half-flats so far when I'm notating, and haven't yet run into a problem.
>
> But I want to be careful not to discourage its use....

Oh, same here! I really think it all boils down to whether or not
people feel that a universal, one-size-fits-all solution is possible,
or even desirable. I tend to think not; others, obviously, believe
otherwise.

Look, this is all discussion, and airing the viewpoints of our
individual needs, uses, and wishes. No need for acrimony, yeah?

Cheers,
Jon

--- In MakeMicroMusic@yahoogroups.com, "Aaron Krister Johnson"
<aaron@...> wrote:
> I have no problem with GUIs, and often prefer them. I've never
> disparaged them or their users as robots, either.

You described users as "swept up in statements designed to make people
march lock-step to someone's, or some large group's, idea of a
'standard'". To me it seemed an apt description - robot. Sorry if I
read something into your phrase that was not intended.

As to the rest, general agreement.

Cheers,
Jon

Well, just to chime in for my part, I do find that it is much easier
to compose when I can hear what I'm doing. Notating via paper and
quill was probably easy for Bach as he had thoroughly explored that
style and could to some extent "hear it" in his head. I can notate
normal scores that way, but when we're getting into 6 part 11-limit
voice leading on I generally need to hear what my chord voicings
actually sound like to write anything meaningful. As time goes on, I
expect I'll start to get the hang of that to where I'll be able to
write without hearing, but it is a matter of where I'm at.

Aaron, you obviously have been doing this for quite some time. You
have a good idea of how septimal chords function in a compositional
context, you have a very intuitive grasp of advanced concepts of
harmony and melody... But I don't, yet. And neither do -ANY- of the
newbies to this field, by virtue of their being newbies. So for now, I
really do need to hear what I'm doing, which is why I proposed the
writing of software to do just that.

On the other hand, I do envision a time when I will be able to hear my
ideas so effectively that it will be much FASTER for me to write using
lilypond or microabc or something, and I won't have the actual audio
of my entering each note interfering with my musical ideas. So to some
extent, both are necessary.

Aaron wrote:
>> >>> I maintain that
>> >>> limited applications like MicroABC are good enough for now.
>> >>
>> >> Good enough for whom?
>> >
>> >Good enough for people who want staff notation with sagittal
>> >accidentals, or different numbers of notes to the octave,
>> >and some other variants of staff notation.
>>
>> If these systems were good enough, people wouldn't spend
>> hundreds of dollars per license on software like Finale and
>> Sibelius, which you admit to never having used. We've already
>> heard from some folks here about how they want microtonal
>> versions of these programs.
>
>In other words--if microtones are so cool, how come no one I know uses
>them? ;)
>
>This is an equivalent statement of 'value'.

Huh?

>Maybe the same people who shell out dough for big programs don't know
>of, or aren't in the culture of, free or open-source software; maybe
>they need or want a really flashy GUI environment and have prejudice
>about working in a text-environment, maybe they are buying because
>they feel swept up in all the professional hype surrounding a 'product'.

Huh??

>Thankfully, there are enough people who don't get swept up in
>statements designed to make people march lock-step to someone's, or
>some large group's, idea of a 'standard'--like Harry Partch, to name one.

Oh yes, let's pat ourselves on the back for using free software.
Let's turn this into an opportunity to discuss the pros and cons
of free and open source software. By all means, let's descend
into outright Linux advocacy. Hey Aaron, the X-Men are discriminated
against because they're different!

>It seems awfully co-dependant to criticise someone's toolset,

Graham said his tools were good enough for anyone who wants
"staff notation with sagittal accidentals, or different numbers
of notes to the octave, and some other variants of staff
notation". It's hard to say exactly what those things entail,
but Graham's tools are not good enough for me, so I'm living
proof by contradiction. So who's "criticizing someone's toolset"?
Fill us in, Aaron.

>especially when people are getting solid work done with that toolset,
>and especially given that these same expensive programs that you
>insist are the best

Where did I say that?

-Carl

As for Sagittal notation, I do find it a great idea, but I do also
find it a bit unwieldy to read. I thought that HEWM and Sims-Maneri (I
forget which is which) notation were both great ideas as well. I
prefer them to Sagittal notation because I find the symbol set easy to
adjust to. Using +/- for an 81/80 adjustment is much easier for me to
get used to than using different kinds of slashes and hooks. I do find
the concept of using "7" to go down a septimal comma and an upside
down 7 "L" shape to go up a septimal comma to be a little weird, but I
find it easier to read than I do the Sagittal equivalent.

I would like to see a version of Sagittal that has a focus on making
the symbol set more intuitive and easier to read... Although there are
those that think it already is intuitive and easier to read, so
perhaps apples and oranges are at play here.

I will say this though: the reason that +/- works so well for the
comma adjustment for me, and that 7/L does not, has to do with the
former using some kind of intuitive, synesthetic paradigm, and the
latter using a scientific, physical paradigm.

Think about it: The simplest accidental there is, the flat and sharp
signs, are both synesthetic terms: we are either "sharpening" or
"flattening" a note. Even though this isn't the original intended
usage for those terms, it stuck in the popular consciousness because
it makes intuitive, synesthetic sense, which is a phenomenon worth
noticing.

The original terms for sharp and flat were "soft B" for Bb and "hard
B" for B natural, and the symbols for flat and natural evolved out of
drawing a "softened" and a "hard" B in that way. Even using the "soft"
and "hard" paradigm to describe this phenomenon is an intuitive,
synesthetic term for it.

In fact, if you want to get into it further, we're using both of these
terms to describe a "raising" and a "lowering" of pitch, do we not?
The terms raising and lowering of pitch are also synesthetic! The
point is that these terms are much better to describe that than
describing a decrease or an increase of wavelength (which would be
backwards from what we're used to) or a "speeding up" or "slowing
down" of frequency. And this is the problem with the Sims-Maneri 7
flag - it has entirely to do with math and theory, and does not
describe the EXPERIENCE of what 7-limit intervals sound like, as the
"soft" and "hard" monikers do for their function, or the "raising and
lowering" monikers for pitch, or even the "plus" and "minus" for very
small comma adjustments.

I think that to come up with a really useful naming system and a
really useful notation system, we need to do some kind of soul
searching and meditating to describe if there is some common thread to
the experience of a 4:7 or 6:7 or 7:9 interval and then use that term
to describe those intervals. Then we'll apply that to whatever
temperament we're working in. For example, in 72-tet, we can say that
7/4 is C-Bb that's a third of a half step flat, but that doesn't
really mean anything musically, does it?

I was describing in the other thread that I heard intervals as having
various degrees of depth - 3/2 was the "shallowest," 5/4 was "deeper,"
and 7/4 was even deeper, etc. Describing an inflection of a 7-limit
interval as "deepening the sound" makes more MUSICAL sense than
describing it as moving by a septimal comma, although some layer of
accuracy is lost, as how "deep" are we deepening here? But I think
that as we explore the territory more and more, we'll find some
musical connotation with some of these higher-and-higher limit
intervals, and we can then give them those labels as "nicknames," if
you will. I do find that certain 7-limit chords have a very "mystical"
quality to them... but I'm not everyone. We don't want synesthetic
names that encourage TOO much as well - 7-limit intervals can be used
for a lot of non-mystical things. Just some simple notion like
"deeper" or "sharper" or something would do.

Or, to sum my position up nicely: I think that everyone can hear that
the experience of moving an interval from 16/9 to 7/4 does have a
quality to it that is different from just moving it down by 33 cents.
This is equivalent to saying that the periodicity mechanism in the
brain for determining intervallic relationships exists. What would be
interesting is if we found a term for that quality and made THAT the
7-limit moniker.

-Mike

On Mon, Jul 7, 2008 at 3:57 PM, Jon Szanto <jszanto@...> wrote:
> Oh, same here! I really think it all boils down to whether or not
> people feel that a universal, one-size-fits-all solution is possible,
> or even desirable. I tend to think not; others, obviously, believe
> otherwise.

As a newcomer here trying to put together a useful working set of
tools I would be happy with a one size fits lots solution. There seems
to be a lot of overlap, especially in my preferred open source world,
and no good way to learn strengths and weaknesses other than time
consuming download and try. A working set of tools that is known to
support exploration of a significant percentage of micro-tuning
sub-spaces would make me happy. On the other hand as a beginner I
probably focus on the obvious and miss where real interesting work is
happening so my judgement may not be relevant. As a newbie though it
does seem like a mess. Every time I tell someone I have settled on
some tools they send me pointers to wonderful stuff I probably should
be looking at instead. I used to waste my time writing tools that
never really got good enough to make me productive. Now I spend my
time trying to learn existing tools and my productivity has improved
but not by much.

Has anyone attempted a survey of existing tools with emphasis on
microtuning support.

Aaron wrote:

>>> Until you can point out some true limitation of abc in your
>>> experience, I'm afraid I'll have to conclude that you're being an
>>> irrational snob. Ditto Carl.
>>
>> I've already explained at some length its limitations in this
>> thread. Want to make a wisecrack without reading it? I'd say
>> you're the snob.
>
> I didn't mean it as a wisecrack, but a statement of how it seems you
> and in particular, Jon, were disparaging something you've never really
> had first hand use of.

It looks like Jon may have misunderstood the particulars of
microABC (or maybe not -- it's not entirely clear). But I think
his comment probably still holds. While we're on the subject,
have you ever achieved proficiency with a modern interactive score
editor? If so which one(s)?

>I find it interesting that people focus on attacking tools around
>here.

I think you perceived an attack on microABC where none was made.
MicroABC lacks realtime audio feedback and is therefore not even
remotely comparable to the kind of thing Mike, Steve, and I were
discussing. Graham is stating that is comparable and "good
enough" (for reasons that are completely opaque to me, alas).

>Ok, getting back to the thread--all I could find, Carl, is that you
>found microabc limited because it didn't have a GUI and because it
>wasn't a real-time system. Fair enough...if those things are what you
>need, then don't use microabc.

Fine by me. I want to discuss the design of a realtime system.
Graham's telling me I shouldn't bother.

>Also, for my purposes, I'm using it the way Bach used paper and
>quill--for, of all crazy things, *notating music*. I don't need a GUI,
>and I don't need real-time, but, your mileage may vary.

As I pointed out, Bach was using the man's notation. I want
to use a different, completely novel notation for each piece
of music I write. Good luck doing that with microABC. It's
not impossible (though some of the notations I want to use
probably aren't possible in microABC) but it would require
crazy-go-nuts pitch system learning skills.

-Carl

At 12:34 PM 7/7/2008, you wrote:
>
>Yes, I tried the recent free trial of the updated FTS. It can work,
>but, it's still too cumbersome for me. I'm a lazy SOB.

Robert is a genius but FTS is one of the hardest-to-use
applications I've ever seen.

-Carl

I'll chime in and say that almost none of the music posted here
since the list began (and I've heard it all) sounds like it was
through-composed. There are notable exceptions, but they're few
and far between.

-Carl

Mike wrote:
>Aaron, you obviously have been doing this for quite some time. You
>have a good idea of how septimal chords function in a compositional
>context, you have a very intuitive grasp of advanced concepts of
>harmony and melody... But I don't, yet. And neither do -ANY- of the
>newbies to this field, by virtue of their being newbies. So for now, I
>really do need to hear what I'm doing, which is why I proposed the
>writing of software to do just that.
>
>On the other hand, I do envision a time when I will be able to hear my
>ideas so effectively that it will be much FASTER for me to write using
>lilypond or microabc or something, and I won't have the actual audio
>of my entering each note interfering with my musical ideas. So to some
>extent, both are necessary.

At 01:37 PM 7/7/2008, Mike wrote:
>As for Sagittal notation, I do find it a great idea, but I do also
>find it a bit unwieldy to read. I thought that HEWM and Sims-Maneri (I
>forget which is which) notation were both great ideas as well. I
>prefer them to Sagittal notation because I find the symbol set easy to
>adjust to.

Me too!

>I would like to see a version of Sagittal that has a focus on making
>the symbol set more intuitive and easier to read... Although there are
>those that think it already is intuitive and easier to read, so
>perhaps apples and oranges are at play here.

It's probably as easy as possible to read given they intend to
distinguish some ridiculous number of commas entirely with different
kinds of arrows.

-Carl

Lots of my music is through composed, most using custom tools.

examples:

The Juggler
Adagio for Margo
Insect Ballet
Tarantella
Madwoman of Chaillot variations
my current project, Alleluia, for 3 voices in 17-equal
Melancholic

butyes, I also do lots of recorded improvs, and what jacob like to
call 'hyperimprovisations', meaning sped up MIDI improvs.

-AKJ.

--- In MakeMicroMusic@yahoogroups.com, Carl Lumma <carl@...> wrote:
>
> I'll chime in and say that almost none of the music posted here
> since the list began (and I've heard it all) sounds like it was
> through-composed. There are notable exceptions, but they're few
> and far between.
>
> -Carl
>
> Mike wrote:
> >Aaron, you obviously have been doing this for quite some time. You
> >have a good idea of how septimal chords function in a compositional
> >context, you have a very intuitive grasp of advanced concepts of
> >harmony and melody... But I don't, yet. And neither do -ANY- of the
> >newbies to this field, by virtue of their being newbies. So for now, I
> >really do need to hear what I'm doing, which is why I proposed the
> >writing of software to do just that.
> >
> >On the other hand, I do envision a time when I will be able to hear my
> >ideas so effectively that it will be much FASTER for me to write using
> >lilypond or microabc or something, and I won't have the actual audio
> >of my entering each note interfering with my musical ideas. So to some
> >extent, both are necessary.
>

--- In MakeMicroMusic@yahoogroups.com, Carl Lumma <carl@...> wrote:
>
> Aaron wrote:
>
> >>> Until you can point out some true limitation of abc in your
> >>> experience, I'm afraid I'll have to conclude that you're being an
> >>> irrational snob. Ditto Carl.
> >>
> >> I've already explained at some length its limitations in this
> >> thread. Want to make a wisecrack without reading it? I'd say
> >> you're the snob.
> >
> > I didn't mean it as a wisecrack, but a statement of how it seems you
> > and in particular, Jon, were disparaging something you've never really
> > had first hand use of.
>
> It looks like Jon may have misunderstood the particulars of
> microABC (or maybe not -- it's not entirely clear).

If you knew anythiing about microABC, you'd have known that his
description was pretty off, in that he thought it didn't even produce
a printed score.

> But I think
> his comment probably still holds.

Which one?

> While we're on the subject,
> have you ever achieved proficiency with a modern interactive score
> editor? If so which one(s)?

Haven't needed to, but then I'm not attacking them, so I don't see
your point. I'm defending against the trash talking pointed at tools
that I use and am productive with.

i don't see how they could be any harder than any other easy GUI
software I'm familiar with

> >I find it interesting that people focus on attacking tools around
> >here.
>
> I think you perceived an attack on microABC where none was made.
> MicroABC lacks realtime audio feedback and is therefore not even
> remotely comparable to the kind of thing Mike, Steve, and I were
> discussing. Graham is stating that is comparable and "good
> enough" (for reasons that are completely opaque to me, alas).

Well, I must have thrown you in the same kettle with Jon, sorry. Jon
clearly thought microabc was pointeless.

I would never claim microabc was useful for what you want to do.

All I'm sayiing is quite simple: it can produce very nice clean
scores, and it's free, which is a plus for some folks who value money.

> >Ok, getting back to the thread--all I could find, Carl, is that you
> >found microabc limited because it didn't have a GUI and because it
> >wasn't a real-time system. Fair enough...if those things are what you
> >need, then don't use microabc.
>
> Fine by me. I want to discuss the design of a realtime system.
> Graham's telling me I shouldn't bother.

Was he?

>
> >Also, for my purposes, I'm using it the way Bach used paper and
> >quill--for, of all crazy things, *notating music*. I don't need a GUI,
> >and I don't need real-time, but, your mileage may vary.
>
> As I pointed out, Bach was using the man's notation. I want
> to use a different, completely novel notation for each piece
> of music I write. Good luck doing that with microABC. It's
> not impossible (though some of the notations I want to use
> probably aren't possible in microABC) but it would require
> crazy-go-nuts pitch system learning skills.

I see this as a difficult problem for ANY software to solve, not just
abc and pals, yes?

-AKJ

--- In MakeMicroMusic@yahoogroups.com, Carl Lumma <carl@...> wrote:
>
> Aaron wrote:
> >> >>> I maintain that
> >> >>> limited applications like MicroABC are good enough for now.
> >> >>
> >> >> Good enough for whom?
> >> >
> >> >Good enough for people who want staff notation with sagittal
> >> >accidentals, or different numbers of notes to the octave,
> >> >and some other variants of staff notation.
> >>
> >> If these systems were good enough, people wouldn't spend
> >> hundreds of dollars per license on software like Finale and
> >> Sibelius, which you admit to never having used. We've already
> >> heard from some folks here about how they want microtonal
> >> versions of these programs.
> >
> >In other words--if microtones are so cool, how come no one I know uses
> >them? ;)
> >
> >This is an equivalent statement of 'value'.
>
> Huh?

By that I meant that your argument seemed to be a 'popular consensus
makes right' argument. i.e. 'everybody uses Finale, therefore anyone
who doesn't use Finale should hang their head in shame. Or that the
siimple fact that people are paying a lot of money means it's the best
solution.

>
> >Maybe the same people who shell out dough for big programs don't know
> >of, or aren't in the culture of, free or open-source software; maybe
> >they need or want a really flashy GUI environment and have prejudice
> >about working in a text-environment, maybe they are buying because
> >they feel swept up in all the professional hype surrounding a
'product'.
>
> Huh??

What's so difficult to understand about what I just said. Reread.
Saying 'huh' is lazy, or admitting you don't really have a strong
point to make in response.

> >Thankfully, there are enough people who don't get swept up in
> >statements designed to make people march lock-step to someone's, or
> >some large group's, idea of a 'standard'--like Harry Partch, to
name one.
>
> Oh yes, let's pat ourselves on the back for using free software.

you are reading far too much into what I'm saying--when have I patted
my own back here?

> Let's turn this into an opportunity to discuss the pros and cons
> of free and open source software.

Don't you mean "let's turn this into an oppurtunity to convince people
that the only solutions are ones we have to shell out big bucks for?"
Your above paragraph betrays that this is your POV on the matter.

> By all means, let's descend
> into outright Linux advocacy.

I haven't even mentioned Linux in this thread once. But you love the
strawman, don't you Carl, it's the top tool in your argument kit.

> Hey Aaron, the X-Men are discriminated
> against because they're different!

Now this deserves a "huh". huh?

> >It seems awfully co-dependant to criticise someone's toolset,
>
> Graham said his tools were good enough for anyone who wants
> "staff notation with sagittal accidentals, or different numbers
> of notes to the octave, and some other variants of staff
> notation". It's hard to say exactly what those things entail,
> but Graham's tools are not good enough for me, so I'm living
> proof by contradiction. So who's "criticizing someone's toolset"?
> Fill us in, Aaron.

I was more responding to Jon when I wrote that.

> >especially when people are getting solid work done with that toolset,
> >and especially given that these same expensive programs that you
> >insist are the best

> Where did I say that?

In the above paragraph about why people are buying expensive licensed
software. Ok, It's implied, not outright said.

-AKJ

Aaron wrote:
>> It looks like Jon may have misunderstood the particulars of
>> microABC (or maybe not -- it's not entirely clear).
>
>If you knew anythiing about microABC, you'd have known that his
>description was pretty off, in that he thought it didn't even produce
>a printed score.

I didn't get that from his message but I have actually used
microABC to produce printed music.

>> But I think
>> his comment probably still holds.
>
>Which one?

That it wouldn't be considered as an option by most professional
orchestral musicians.

>> >Ok, getting back to the thread--all I could find, Carl, is that you
>> >found microabc limited because it didn't have a GUI and because it
>> >wasn't a real-time system. Fair enough...if those things are what you
>> >need, then don't use microabc.
>>
>> Fine by me. I want to discuss the design of a realtime system.
>> Graham's telling me I shouldn't bother.
>
>Was he?

Yup.

>>> Also, for my purposes, I'm using it the way Bach used paper and
>>> quill--for, of all crazy things, *notating music*. I don't need
>>> a GUI, and I don't need real-time, but, your mileage may vary.
>>
>> As I pointed out, Bach was using the man's notation. I want
>> to use a different, completely novel notation for each piece
>> of music I write. Good luck doing that with microABC. It's
>> not impossible (though some of the notations I want to use
>> probably aren't possible in microABC) but it would require
>> crazy-go-nuts pitch system learning skills.
>
>I see this as a difficult problem for ANY software to solve, not just
>abc and pals, yes?

Sure. But with realtime audiovisual feedback, it gets exponentially
easier.

-Carl

Aaron wrote:

>> >Thankfully, there are enough people who don't get swept up in
>> >statements designed to make people march lock-step to someone's, or
>> >some large group's, idea of a 'standard'--like Harry Partch, to
>> >name one.
>>
>> Oh yes, let's pat ourselves on the back for using free software.
>
>you are reading far too much into what I'm saying--when have I patted
>my own back here?

If I gave you a nickel for every time you've delurked to deliver
your own patented blend of linux advocacy here, you could afford
to buy that copy of Finale.

>> Let's turn this into an opportunity to discuss the pros and cons
>> of free and open source software.
>
>Don't you mean "let's turn this into an oppurtunity to convince people
>that the only solutions are ones we have to shell out big bucks for?"
>Your above paragraph betrays that this is your POV on the matter.

Hardly.

-Carl

Aaron,

--- In MakeMicroMusic@yahoogroups.com, "Aaron Krister Johnson"
<aaron@...> wrote:
> > It looks like Jon may have misunderstood the particulars of
> > microABC (or maybe not -- it's not entirely clear).
>
> If you knew anythiing about microABC, you'd have known that his
> description was pretty off, in that he thought it didn't even produce
> a printed score.

Look, I'm going to back away from this for a while. You seem to have
completely ignored my follow-up messages, which took pains to point
out that I stated up-front that I might have been unaware of later
developments. And if one were being completely picky, microABC does
not produce notation - you need the additional app abc2ps for that.
I'd be pleased if someone could give me a timeline, because the last
time I was active on this list was when Hudson was working hard on
microABC, and I am wondering: did abc2ps, especially with microtonal
font support, come later? It would certainly go a ways to explaining
why I was unaware of at least limited support for printing.

Though that doesn't mean I "don't understand" it.

> Haven't needed to, but then I'm not attacking them, so I don't see
> your point. I'm defending against the trash talking pointed at tools
> that I use and am productive with.

If simply raising criticisms in a very benign manner constitutes
"trash talking", you've developed a very thin skin. Really, Aaron, I'm
glad it works for you, but I think that - in it's present state -
there are very few people who would choose this as a method,
*especially* if they are going to be presenting printed scores and
parts. Which was sort of my point, and I didn't at any juncture trash
the work that Hudson, et al, have done.

> Well, I must have thrown you in the same kettle with Jon, sorry. Jon
> clearly thought microabc was pointeless.

Not at all, and shame on you for characterizing my position that way.

> I would never claim microabc was useful for what you want to do.

Which was all *I* was saying as well - that it didn't meet my needs,
and, by extension, other musicians like me. Sheesh.

Regards,
Jon

I don't understand the argument here. MicroABC already exists. Some people
don't like it. Why are those people under attack? Jesus, if everyone second
guessed themselves every time they found some software interface hard to
use, human beings would have never developed the entire art of feng shui and
graphic design. Interface development is an academic field that is tied in
with accessibility and psychology.

-Mike

[Non-text portions of this message have been removed]

Mike,

--- In MakeMicroMusic@yahoogroups.com, "Mike Battaglia"
<battaglia01@...> wrote:
> MicroABC already exists. Some people
> don't like it. Why are those people under attack?

This is what pisses me off about Aaron's mis-characterizations of what
I wrote. Go back to my original message and read it in context:

/makemicromusic/topicId_18807.html#19213

I'm not going to explain myself again, but at no point did I "attack"
the authors, and I also took pains to offer that I might not have been
aware of notational developments.

Regards,
Jon

--- In MakeMicroMusic@yahoogroups.com, Carl Lumma <carl@...> wrote:
>
> Aaron wrote:
>
> >> >Thankfully, there are enough people who don't get swept up in
> >> >statements designed to make people march lock-step to someone's, or
> >> >some large group's, idea of a 'standard'--like Harry Partch, to
> >> >name one.
> >>
> >> Oh yes, let's pat ourselves on the back for using free software.
> >
> >you are reading far too much into what I'm saying--when have I patted
> >my own back here?
>
> If I gave you a nickel for every time you've delurked to deliver
> your own patented blend of linux advocacy here, you could afford
> to buy that copy of Finale.

I'll literally pay you a nickel for every post of mine advocating
Linux. Go ahead, find them. Show us all how much you are a blowhard.

If I paid you a nickel every time you exagerrated in a post, or used a
strawman argument, or posted at all, you'd have your own private island.

> >> Let's turn this into an opportunity to discuss the pros and cons
> >> of free and open source software.
> >
> >Don't you mean "let's turn this into an oppurtunity to convince people
> >that the only solutions are ones we have to shell out big bucks for?"
> >Your above paragraph betrays that this is your POV on the matter.
>
> Hardly.

You said "if these solutions were worth anything, noone would shell
out dough at all for Finale or Sibelius"...pretty self-evidently
dismissive of what you considered Graham's 'solution' to be.

-AKJ

--- In MakeMicroMusic@yahoogroups.com, "Mike Battaglia"
<battaglia01@...> wrote:
>
> I don't understand the argument here. MicroABC already exists. Some
people
> don't like it. Why are those people under attack? Jesus, if everyone
second
> guessed themselves every time they found some software interface hard to
> use, human beings would have never developed the entire art of feng
shui and
> graphic design. Interface development is an academic field that is
tied in
> with accessibility and psychology.

That wasn't my argument. In fact, I took great pains to say ABC wasn't
for everyone.

What bothered me was Jon dismissing it, when he clearly (at the time)
didn't even understand it could print music, of all things.

-AKJ

--- In MakeMicroMusic@yahoogroups.com, "Jon Szanto" <jszanto@...> wrote:
>
> Aaron,
>
> --- In MakeMicroMusic@yahoogroups.com, "Aaron Krister Johnson"
> <aaron@> wrote:
> > > It looks like Jon may have misunderstood the particulars of
> > > microABC (or maybe not -- it's not entirely clear).
> >
> > If you knew anythiing about microABC, you'd have known that his
> > description was pretty off, in that he thought it didn't even produce
> > a printed score.
>
> Look, I'm going to back away from this for a while. You seem to have
> completely ignored my follow-up messages, which took pains to point
> out that I stated up-front that I might have been unaware of later
> developments. And if one were being completely picky, microABC does
> not produce notation - you need the additional app abc2ps for that.
> I'd be pleased if someone could give me a timeline, because the last
> time I was active on this list was when Hudson was working hard on
> microABC, and I am wondering: did abc2ps, especially with microtonal
> font support, come later? It would certainly go a ways to explaining
> why I was unaware of at least limited support for printing.
>
> Though that doesn't mean I "don't understand" it.

Well, I'm confused by what you mean by 'developments'---abc, abcm2ps,
and microabc have from the beginning produced postscript scores.

>
> > Haven't needed to, but then I'm not attacking them, so I don't see
> > your point. I'm defending against the trash talking pointed at tools
> > that I use and am productive with.
>
> If simply raising criticisms in a very benign manner constitutes
> "trash talking", you've developed a very thin skin. Really, Aaron, I'm
> glad it works for you, but I think that - in it's present state -
> there are very few people who would choose this as a method,
> *especially* if they are going to be presenting printed scores and
> parts.

I guess I don't understand what this proves about anything. Very few
people write in 13-limit Just Intonation...let's not do that, then,
either.

In fact, in it's present state, microabc/abcm2ps can write very
complex and clean microtonal scores and parts. It's capable of nested
tuplets, beam-crossings, barless scores, note-heads only, percussion
staves, you name it.

> Which was sort of my point, and I didn't at any juncture trash
> the work that Hudson, et al, have done.
>
> > Well, I must have thrown you in the same kettle with Jon, sorry. Jon
> > clearly thought microabc was pointeless.
>
> Not at all, and shame on you for characterizing my position that
> way.

It seems logical to me anyway to assume you thought that if it didn't
print an actual score, it was useless in a sense.

> > I would never claim microabc was useful for what you want to do.
>
> Which was all *I* was saying as well - that it didn't meet my needs,
> and, by extension, other musicians like me. Sheesh.

Maybe I came into this late, but I missed the part where you were
talking about needing real-time notation. Sorry.

-AKJ.

Aaron wrote:
>> If I gave you a nickel for every time you've delurked to deliver
>> your own patented blend of linux advocacy here, you could afford
>> to buy that copy of Finale.
>
>I'll literally pay you a nickel for every post of mine advocating
>Linux. Go ahead, find them. Show us all how much you are a blowhard.
>
>If I paid you a nickel every time you exagerrated in a post, or used a
>strawman argument, or posted at all, you'd have your own private island.

I'll match you one for one. I'll go first:

"Why would I use that? That's not a Linux app."

Here's another:

"It seems that you and Jon have said some pretty nonsensical things
about Linux as an audio platform in the past, so this list was designed
to put the idea that one needs Windoze (a.k.a. crash box) or Apple to
rest."

Ok, I can't stop:

"Not to mention that Linux low-latency is *more mature* and lower
than anything Windows to date could do."

This one is killer:

/makemicromusic/topicId_18143.html#18149

Of course you have no problem comparing Charles to a "pastor in
a megachurch" for advocating his preferred platform (message
apparently deleted from archives), though I admit he's a far
greater offender than you.

>You said "if these solutions were worth anything, noone would shell
>out dough at all for Finale or Sibelius"...pretty self-evidently
>dismissive of what you considered Graham's 'solution' to be.

That's not what I said. Please don't use double quotes for stuff
you make up.

-Carl

Aaron,

--- In MakeMicroMusic@yahoogroups.com, "Aaron Krister Johnson"
<aaron@...> wrote:
> Well, I'm confused by what you mean by 'developments'---abc, abcm2ps,
> and microabc have from the beginning produced postscript scores.

Then I confess that I was not aware of the printing capability, and
must have only read about the microtonal additions that Hudson made to
ABC. As I stated in my original post, I was open to being corrected on
this, and I consider myself corrected with regard to printed output.

> I guess I don't understand what this proves about anything. Very few
> people write in 13-limit Just Intonation...let's not do that, then,
> either.

No, that is an exaggeration, and I never said people shouldn't do it.
What I *did* imply is that not very many people would choose to work
with a tool like this, which makes you conceive of a score with each
independent line being a separate section of the input text file. I
looked at a number of examples this afternoon, including a simply 4
voice Bach fugue in Hudson's example pdf. I'd shudder to think what it
would be like to compose a chamber ensemble of 10 or 20 parts in a
scenario such as this. ABC seems to have had it's roots in folk
traditions, and a lot of monophonic, melodic music. I remain
unconvinced that it can scale up to more complex scores.

> In fact, in it's present state, microabc/abcm2ps can write very
> complex and clean microtonal scores and parts. It's capable of nested
> tuplets, beam-crossings, barless scores, note-heads only, percussion
> staves, you name it.

This is good.

> It seems logical to me anyway to assume you thought that if it didn't
> print an actual score, it was useless in a sense.

Then you assumed wrong. It's not a binary choice, my friend, that to
fall short automatically means fail. It doesn't.

> Maybe I came into this late, but I missed the part where you were
> talking about needing real-time notation. Sorry.

I never did. But I'm also not about composing music by typing in lines
of ascii text.

Jon

--- In MakeMicroMusic@yahoogroups.com, Carl Lumma <carl@...> wrote:
>
> Aaron wrote:
> >> If I gave you a nickel for every time you've delurked to deliver
> >> your own patented blend of linux advocacy here, you could afford
> >> to buy that copy of Finale.
> >
> >I'll literally pay you a nickel for every post of mine advocating
> >Linux. Go ahead, find them. Show us all how much you are a blowhard.
> >
> >If I paid you a nickel every time you exagerrated in a post, or used a
> >strawman argument, or posted at all, you'd have your own private
island.
>
> I'll match you one for one. I'll go first:
>
> "Why would I use that? That's not a Linux app."