Yahoo Tuning Groups Ultimate Backup makemicromusic Re: [MMM] MIDI to WAV

Carl,

{you wrote...}
>These look like tools that capture the output of a synth on your PC and save it to disk...
>http://www.midi2wav.com/
>http://convert-midi.com/midi-to-wav.shtml
>http://www.maniactools.com/soft/midi_converter/index.shtml
>http://www.008soft.com/products/midi-to-wav-maker.htm

None of those mention anything about a synth - since they don't give any explicit information, it looks like they must just use a generic GM midi synth, no options for other sounds, soundfonts, etc.

>GigaStudio is another possibility. It's always crashed my machine and the interface is a study in obfuscation.

Yeah, not to mention expensive.

I don't know why people won't at least try Cakewalk's Music Creator Pro:
http://www.cakewalk.com/Products/MusicCreatorPro/default.asp

For only $40 you've got a large percentage of the capabilities of Sonar. Even if one *only* wanted to pour in a midi file and not do any editing, etc, you've at least got the options of external midi hardware, VST instruments and effects, and still have GM synth capabilities (if one wants that). But then with that minimal investment you've now got a platform to do a lot more editing of the musical material prior to rendering, which can't be a bad thing at all.

Hell, I might buy it as a test bed for things like this on a low-end PC and see what I can get out of it.

Cheers,
Jon

🔗Carl Lumma <ekin@...>

I just bought
http://www.flstudio.com/documents/SUB_25.html
but 'due to the enormous amount of fraud (65% of our
orders are with stolen credit cards), we have no choice
but to process each order manually, during EU business
hours'. So I won't have it until Monday, I guess.

Meanwhile, I found
http://homepage.ntlworld.com/jez.price/players.htm
and downloaded Sam!Solo and Font!SF2 to try out.

-C.

At 01:29 PM 4/8/2006, you wrote:
>I'll try to write this e-mail again, since it never showed up.
>
>These look like tools that capture the output of a synth on
>your PC and save it to disk...
>http://www.midi2wav.com/
>http://convert-midi.com/midi-to-wav.shtml
>http://www.maniactools.com/soft/midi_converter/index.shtml
>http://www.008soft.com/products/midi-to-wav-maker.htm
>
>WAVmaker is tuning-friendly and has its own synthesis
>capabilities, but it isn't clear what sample formats it
>supports. I've sent an email asking about sf2, but I'm not
>hopeful. Daniel Wolf used to use it... can you update us
>on this, Daniel?
>http://www.polyhedric.com/software/wavmaker/
>
>My old tool of choice, Audio Compositor, is still dead. :(
>
>GigaStudio is another possibility. It's always crashed my
>machine and the interface is a study in obfuscation.
>
>Edirol's Sound Canvas softsynths are another option. Herman
>Miller is using one of them. They don't support sf2 I don't
>think, but their sounds are I guess not too bad, and they
>do save WAV, and... what about MTS support...?
>
>Thanks,
>
>-Carl

🔗Carl Lumma <ekin@...>

At 01:59 PM 4/8/2006, you wrote:
>Carl,
>
>{you wrote...}
>>These look like tools that capture the output of a synth on your PC
>and save it to disk...
>>http://www.midi2wav.com/
>>http://convert-midi.com/midi-to-wav.shtml
>>http://www.maniactools.com/soft/midi_converter/index.shtml
>>http://www.008soft.com/products/midi-to-wav-maker.htm
>
>None of those mention anything about a synth - since they don't give
>any explicit information, it looks like they must just use a generic
>GM midi synth, no options for other sounds, soundfonts, etc.

They don't have a synth, they use existing ones on your machine.
I thought I said that. They use Windows MIDI devices, not VST or
anything...

>>GigaStudio is another possibility. It's always crashed my machine
>and the interface is a study in obfuscation.
>
>Yeah, not to mention expensive.

Some of us have industry copies. :)

>I don't know why people won't at least try Cakewalk's Music Creator Pro:
> http://www.cakewalk.com/Products/MusicCreatorPro/
>For only $40

Looks like $110. Music Creator 3 is $40, though. I wonder what
the difference is?

-C.

🔗Jon Szanto <jszanto@...>

Hi,

{you wrote...}
>They don't have a synth, they use existing ones on your machine. I thought I said that.

You said something like that, but I took a look at all four sites, and see nada about this. Does this mean that you've used them and found this out, or did you find other documentation? Not carping, I just don't see anything relating to using an existing synth.

>They use Windows MIDI devices, not VST or anything...

Hmmm...

>Some of us have industry copies. :)

Warez! Warez! :)

>Looks like $110. Music Creator 3 is $40, though. I wonder what the difference is?

My bad. It will still be around $100, as MC3 is 40 and you need the VST adapter (soft) if you want to go that route, which is 60. But the additional stuff bundled with the Pro I don't think warrants the additional expense, you'd be better off going just a tad futher and getting one of the Sonar home editions.

But since I've already got the adapter, I will probably pop for MC3 and give it a whirl. I'm sure just setting up a default template, into which you open a midi file, could give some very nice results just as a first step, and from there you could go even further with musical development.

I'll report back some point in the future...

Cheers,
Jon

🔗Carl Lumma <ekin@...>

At 02:13 PM 4/8/2006, you wrote:
>Hi,
>
>{you wrote...}
>>They don't have a synth, they use existing ones on your machine. I
>thought I said that.
>
>You said something like that, but I took a look at all four sites, and
>see nada about this. Does this mean that you've used them and found
>this out, or did you find other documentation? Not carping, I just
>don't see anything relating to using an existing synth.

I tried the first one, and that's how it works. Like it, the
others don't mention they come with a synth. If they do, they
would almost certainly mention it. One makes mention of requiring
a full duplex sound card. These programs are all clearly in the
same class -- loopback-type recorders. I'm happy to be proven
wrong.

>>Some of us have industry copies. :)
>
>Warez! Warez! :)

No, I worked in the industry. I have pretty much any piece of
music software around. Of course, before that, I had warez.

>>Looks like $110. Music Creator 3 is $40, though. I wonder what the
>difference is?
>
>My bad. It will still be around $100, as MC3 is 40 and you need the
>VST adapter (soft) if you want to go that route, which is 60. But the
>additional stuff bundled with the Pro I don't think warrants the
>additional expense, you'd be better off going just a tad futher and
>getting one of the Sonar home editions.

Does Pro come with the adapter? Then whatever else extra it came
with would be free.

I of course have Sonar, but I think it would be easier to be
elected governor than make music with it.

>But since I've already got the adapter, I will probably pop for MC3
>and give it a whirl. I'm sure just setting up a default template, into
>which you open a midi file, could give some very nice results just as
>a first step, and from there you could go even further with musical
>development.
>
>I'll report back some point in the future...

Please do.

-Carl

🔗David Beardsley <db@...>

🔗Gene Ward Smith <genewardsmith@...>

🔗Carl Lumma <ekin@...>

🔗Jon Szanto <jszanto@...>

{you wrote...}
>Where to begin... let's just say, they made sure I wasn't the one to review it.

That's good. Not that I would buy/not buy on a review alone (I like to demo software) it has been one of the most fun and pleasurable experiences I've had in many years of home recording. I can't believe what I can do with it, how easy it is, and there is a great community of users.

It doesn't matter, however, as all that counts is what works for each individual. Doesn't for you, it does for me, and that's that.

I might rather have you as governor, however. :)

Cheers,
Jon

🔗Jon Szanto <jszanto@...>

Gene,

{you wrote...}
>--- In MakeMicroMusic@yahoogroups.com, Jon Szanto <jszanto@...> wrote:
>> For only $40 you've got a large percentage of the capabilities of
>Sonar.
>
>For $40 dollars do I get something which supports microtonality?

I'm sorry, I'll hope you'll read on to the reply to Carl: it would really be a total of around $100, because you need a small software addition that allows you to use VST instruments (while I don't want to get too technical, you actually *could* do it with just the plain program, as long as you purchased the z3ta+ synth, which has a DXi version; you'd just have the one instrument - one Joe Pehrson uses - but I know you aren't a synth guy).

This should seem to be clear after the many times it has been discussed: for many people, the microtonality isn't in the host program, but in the instruments one uses. My setup is to use Sonar in conjunction with a variety of samplers and synths that *themselves* support microtonality. This way I'm not locked into pitch bending or relaying for microtonal support. Sonar, or any other host or sequencer that supports these VST instrument, doesn't have to know a single thing about tuning - it simply records and plays midi notes, and the synths are tuned themselves to whatever intonation you want.

Again: since there are a dozen or more software instruments (VSTi) that can accept a tuning of pretty much any manner, all they need is something to drive them. This is what the host and/or sequencer is for. In my little contrib piece, I used 3 instruments (not counting the percussion) - 1 sampler/synth and 2 synths - that all were tuned to a scale I created in Scala, and saved to a file(s). These tuning files (one uses .scl, the other two use .tun) were loaded into the instruments. I composed using them that way in realtime. If I were to save the result as a midi file, and you were to play it, it would come out 12tet, because there aren't any pitch bends in it.

Hope that makes sense...

Cheers,
Jon

🔗David Beardsley <db@...>

🔗Gene Ward Smith <genewardsmith@...>

--- In MakeMicroMusic@yahoogroups.com, Jon Szanto <jszanto@...> wrote:

> I'm sorry, I'll hope you'll read on to the reply to Carl: it would
really be a total of around $100, because you need a small software
addition that allows you to use VST instruments (while I don't want to
get too technical, you actually *could* do it with just the plain
program, as long as you purchased the z3ta+ synth, which has a DXi
version; you'd just have the one instrument - one Joe Pehrson uses -
but I know you aren't a synth guy).

I'm not a synth guy because they don't seem to be making synths with
me in mind. I want something which can work from an ascii score of
some kind, like Csound, or from a midi file, and which supports
microtonality. If I could get a collection of Csound intruments I
liked I'd be in business, perhaps.

> This should seem to be clear after the many times it has been
discussed: for many people, the microtonality isn't in the host
program, but in the instruments one uses. My setup is to use Sonar in
conjunction with a variety of samplers and synths that *themselves*
support microtonality.

The host program has the note numbers and instruments without the tunings?

> Again: since there are a dozen or more software instruments (VSTi)
that can accept a tuning of pretty much any manner, all they need is
something to drive them. This is what the host and/or sequencer is for.

And what drives the host--in other words, what is the score, if any?

🔗Carl Lumma <ekin@...>

>The host program has the note numbers and instruments without the tunings?

Right.

>And what drives the host--in other words, what is the score, if any?

The way these things work is, they're set up like a cross between
a mixing board and a MIDI sequencer. You have a timeline with tracks
that can be either audio or MIDI. The MIDI tracks go out through
synths, the audio through effects, and everything gets bussed back
into a mix of some kind. Signal flow in these things is horrendously
confusing.

-Carl

🔗Jon Szanto <jszanto@...>

Gene,

{you wrote...}
>I'm not a synth guy because they don't seem to be making synths with me in mind.

Right.

>I want something which can work from an ascii score of some kind, like Csound, or from a midi file, and which supports microtonality. If I could get a collection of Csound intruments I liked I'd be in business, perhaps.

I'd think Csound would be a good environment for you, but might be more "getting your hands dirty" than you'd care to deal with (or have time). Do you like Prent's use of samples in Csound? Since I know you would like to emulate acoustic instruments, it seems like his work is some of the most successful of that strategy in Csound I've heard.

>The host program has the note numbers and instruments without the tunings?

Ah, the light bulb goes on! Yes, that is it exactly. For instance, let us say I want to write a single line in 15tet. I start up the sequencer (Sonar), insert an instrumental track (lets say an instrument called Rhino). When I do this there is now a track ready for me to record some midi music. I open Rhino and load a Scala .tun file for 15tet. Now, whenever I play my 61-note keyboard, Rhino is playing notes in 15tet. I can now enter music by

- playing in real time
- step entering notes (this is non-realtime entry from the keyboard)
- use a piano roll
- I could even, if I had a midi file laying around, simply 'import' the midi file

Hit the "Play" button and it plays the music in 15tet. It really is pretty simple. One of the beauties of this is that since it is the instrument that encompasses the tuning, and not the midi file, I could have any number of instruments tuned to differing tunings. Not everyone would do this, but for instance: someone working in a large-number ET for accuracy might actually be composing parts with small subsets. You could create tuning files for some of those subscales and load those individually.

I'm not saying everyone should work this way, and it may very well not be the best route for you. But it really works well for me right now, and is getting better all the time. One of the things about the current MMM Day stuff is that I've got a program that translates samples from/to many formats. What this means is that I can take, for instance, soundfonts and translate them to a format for an instrument called Wusikstation; this instrument tunes with .tun files, and so I now have access to some decent instruments (I used a Steinway soundfont in this piece) that can be microtuned. It is a lot less expensive than going the Kontakt 2 route - I still have more work to do to get the samples/patches sounding the way I want, but it may mean that I don't have to deal with Kontakt for the moment.

>And what drives the host--in other words, what is the score, if any?

The midi data that is either input by me, or imported into the sequencer from an existing midi file. One of the nice things is that you can immediately view the data as a list of events, piano roll, or actual staff notation (12tet only).

Cheers,
Jon

🔗Gene Ward Smith <genewardsmith@...>

🔗Joe <tamahome02000@...>

Ok, I'm pretty sure I saw an ad saying they'll give away Wusikstation
in the next Computer Music magazine (March?). It's one of those soft
synths that understands .tun scala files. I'm not sure if it creates
sounds from scratch, or uses samples though. So you could import a
midi without any pitch bends, and then set the tuning for them. Gene,
do you actually write those scala .seq files from scratch?

Joe

--- In MakeMicroMusic@...m, Jon Szanto <jszanto@...> wrote:
>
> Gene,
>
> {you wrote...}
> >I'm not a synth guy because they don't seem to be making synths
with me in mind.
>
> Right.
>
> >I want something which can work from an ascii score of some kind,
like Csound, or from a midi file, and which supports microtonality. If
I could get a collection of Csound intruments I liked I'd be in
business, perhaps.
>
> I'd think Csound would be a good environment for you, but might be
more "getting your hands dirty" than you'd care to deal with (or have
time). Do you like Prent's use of samples in Csound? Since I know you
would like to emulate acoustic instruments, it seems like his work is
some of the most successful of that strategy in Csound I've heard.
>
> >The host program has the note numbers and instruments without the
tunings?
>
> Ah, the light bulb goes on! Yes, that is it exactly. For instance,
let us say I want to write a single line in 15tet. I start up the
sequencer (Sonar), insert an instrumental track (lets say an
instrument called Rhino). When I do this there is now a track ready
for me to record some midi music. I open Rhino and load a Scala .tun
file for 15tet. Now, whenever I play my 61-note keyboard, Rhino is
playing notes in 15tet. I can now enter music by
>
> - playing in real time
> - step entering notes (this is non-realtime entry from the keyboard)
> - use a piano roll
> - I could even, if I had a midi file laying around, simply 'import'
the midi file
>
> Hit the "Play" button and it plays the music in 15tet. It really is
pretty simple. One of the beauties of this is that since it is the
instrument that encompasses the tuning, and not the midi file, I could
have any number of instruments tuned to differing tunings. Not
everyone would do this, but for instance: someone working in a
large-number ET for accuracy might actually be composing parts with
small subsets. You could create tuning files for some of those
subscales and load those individually.
>
> I'm not saying everyone should work this way, and it may very well
not be the best route for you. But it really works well for me right
now, and is getting better all the time. One of the things about the
current MMM Day stuff is that I've got a program that translates
samples from/to many formats. What this means is that I can take, for
instance, soundfonts and translate them to a format for an instrument
called Wusikstation; this instrument tunes with .tun files, and so I
now have access to some decent instruments (I used a Steinway
soundfont in this piece) that can be microtuned. It is a lot less
expensive than going the Kontakt 2 route - I still have more work to
do to get the samples/patches sounding the way I want, but it may mean
that I don't have to deal with Kontakt for the moment.
>
> >And what drives the host--in other words, what is the score, if any?
>
> The midi data that is either input by me, or imported into the
sequencer from an existing midi file. One of the nice things is that
you can immediately view the data as a list of events, piano roll, or
actual staff notation (12tet only).
>
> Cheers,
> Jon
>

🔗Gene Ward Smith <genewardsmith@...>

--- In MakeMicroMusic@yahoogroups.com, "Joe" <tamahome02000@...> wrote:

Gene,
> do you actually write those scala .seq files from scratch?

I screw around with various things which I put into Maple, which then
screws around with them a whole lot more, which then writes out the
first version of the Scala seq file to which I attach the correct
header stuff. Then I make a midi file which I torment further by means
of other programs such as Nytonyx, getting various kinds of data which
I sometimes go back to Maple with. At various stages you can listen to
the results and make further changes as a result.

The same basic approach works for Csound.

🔗Gene Ward Smith <genewardsmith@...>

🔗Jon Szanto <jszanto@...>

{you wrote...}
>--- In MakeMicroMusic@yahoogroups.com, "Gene Ward Smith"
><genewardsmith@...> wrote:
>Etc etc etc. This does not seem to be the same as what people call algorithmic composition, so I've quit saying that. It is interactive, between me and the computer, and computer aided, but the computer is used as a composition tool, not as the composer. Who else does this? The algorithmic composition group on Yahoo seems to be from another perspective.

What exactly DO you do? I have never been able to figure out just what it is you start with to get the whole ball rolling? Why not share the very first element in the process, so we could all brainstorm on it?

Cheers,
Jon

🔗Gene Ward Smith <genewardsmith@...>

--- In MakeMicroMusic@yahoogroups.com, Jon Szanto <jszanto@...> wrote:

> What exactly DO you do? I have never been able to figure out just
what it is you start with to get the whole ball rolling? Why not share
the very first element in the process, so we could all brainstorm on it?

I do different things. One approach is this: start with a tuning
system, and a scale to go with it, and do some one-fingered noodling
to get thematic material. Now deconstruct the notes into 4 or 5 equal,
plus a chord (NB: this is not a unique process so far as the chords
go), which at this point will just be complete tetrads or pentads. Now
take the 4 or 5 equal stuff, and start composing in it, using the
thematic material as reduced to 4 or 5 et. Get something which sounds
decent this way; the point is, in 4 or 5 everything harmonizes with
everything else, since the whole thing is really a chord, so you can
get your parts to move around freely. Now you've got some of a 4 and 5
equal piece, and when you add in chords for the parts which had chord
info attached, your theme is reconstituted. You can use the chord info
for the bass part, favoring root position.

The people composing music in 4 and 5 equal around here are really
missing a bet, in that it makes a great starting point, by the way.
That's what I did when I took Andrew Heathwaite's 5-et piece and
transmuted it; that could be done just as easily with Rosencrantz the
Sane's 4-et piece, or Hans Straub's 5-et piece. The final result
sounds noticably like the 4 or 5 et starting point, but the addition
of harmony makes an enormous difference; you can compare the two
versions of Pinta Penta to get one example.

Anyway, you now have a bunch of stuff with gaps in the chords, and you
fill in the gaps--I like to use chords with common notes; at this
point you also can start filling in chords which go beyond
otonal/utonal. Now use Maple to reconsitute music from 4 or 5 et
skeletons plus chords. Now listen to what you've got either as a midi
file or when rendered; and see where to go from there. You can grab
things and transform them in various ways, etc etc.

Or, you could simply write a piece from the get-go in 4 or 5 et, and
harmonize that. All these people demonstrating the musical
possibilities of 4 or 5 et are also demonstrating something else.

🔗Gene Ward Smith <genewardsmith@...>

--- In MakeMicroMusic@yahoogroups.com, "Gene Ward Smith"
<genewardsmith@...> wrote:

> The people composing music in 4 and 5 equal around here are really
> missing a bet, in that it makes a great starting point, by the way.

I've been trying to convince Robert Walker that an algorithmic
microtonal composition program which goes way beyond what he has now,
or anyone else has, is in reach from where he is now. What's required
for the first step is something which writes music in parts in 3, 4 or
5 et.
I suppose I could try my hand at writing code to do that myself, but
FTS has already gone so far. I also think Tonalsoft might be able to
have a composition system which would be a lot easier to use using
these ideas; as it is it makes the most sense for me to just use it to
write in 4 or 5 et, which is hard enough as it is with these cussed
piano rolls, and export that.

🔗Jon Szanto <jszanto@...>

Gene,

{you wrote...}
>--- In MakeMicroMusic@yahoogroups.com, "Gene Ward Smith"
><genewardsmith@...> wrote:
>What's required for the first step is something which writes music in parts in 3, 4 or 5 et.

How possible is it to write interesting music in 3/4/5 tet? Give yourself a test: write a melodic line - no accompanying parts whatsoever - in one of those, and see if you can come up with something that is memorable or compelling listening-wise.

I understand that something could be called a "melody" in definition, but if we are talking about making, in the end, music that people want to listen to, there has to be intrinsic interest in the materials. If all one is doing with an algo engine is spinning out raw material to hang harmonization on, then we are back in the realm of harmony experiments. Even the raw material should have some manner of musical interest to it!

I could very well be missing the boat on this one. If you know of examples of 3/4/5 that can bear extended and repeated listenings, I'd be curious. Whenever I've messed around with 5, it's lack of a personality makes for rather directionless music. It is quite safe, as everything sounds 'ok', but unless the object is to just float for long periods of time, I'm not getting it.

By the way, thanks for the exposition on your methodology. I'm curious: you say you "noodle around" to compose a melody. Is it possible to hear what that sounds like before you do anything else? I ask because I listened to Dreyfus again last night, and am curious to hear the initial steps.

Cheers,
Jon

🔗Gene Ward Smith <genewardsmith@...>

--- In MakeMicroMusic@yahoogroups.com, Jon Szanto <jszanto@...> wrote:

> How possible is it to write interesting music in 3/4/5 tet? Give
yourself a test: write a melodic line - no accompanying parts
whatsoever - in one of those, and see if you can come up with
something that is memorable or compelling listening-wise.

It doesn't need to be that transcendently interesting, because it
isn't going to stay that way if you do what I've suggested. What do
you think of the examples people post from time to time?

>If all one is doing with an algo engine is spinning out raw material
>to hang harmonization on, then we are back in the realm of harmony
>experiments. Even the raw material should have some manner of musical
>interest to it!

It's not to be expected that an algo engine cranks out Mozart--even
Cope, who basically cheats by feeding Mozart in at one end, doesn't
get Mozart out at the other.

> I could very well be missing the boat on this one. If you know of
examples of 3/4/5 that can bear extended and repeated listenings, I'd
be curious. Whenever I've messed around with 5, it's lack of a
personality makes for rather directionless music. It is quite safe, as
everything sounds 'ok', but unless the object is to just float for
long periods of time, I'm not getting it.

It's precisely the safety which is the point. But I'm not proposing
you stick there, it's just a starting point.

> By the way, thanks for the exposition on your methodology. I'm
curious: you say you "noodle around" to compose a melody. Is it
possible to hear what that sounds like before you do anything else? I
ask because I listened to Dreyfus again last night, and am curious to
hear the initial steps.

Well, Dreyfus actually has too many notes for the noodling to occur.
Noodling only works when you've got a small enough scale, and 72 is
too many.

> Jon
>