Cameleon 5000 and additive synthesis

Has anybody seen this, or know about its microtuning
support?

http://www.camelaudio.com/productsCA5000.htm

I downloaded all the sound samples, and I was fairly
impressed. I've been looking for a good additive synth
for a long time.

In particular, I'm interested in being able to specify
the tuning of the partials, Sethares-style (hi Bill!).
And in particular, a crucial feature would seem to be
a 'partial-tuning map' abstraction, which would allow all
the the patches on the instrument to be quickly 'retuned'
together.

I tend to doubt anything does this out of the box, though
I guess it could be done in CSound, Max, or Kyma. If
anybody knows of such a thing I'd like to hear about it.
Cameleon does advertise "64 detunable partials", and I've
downloaded the demo. I'll write back when I've tried it,
and I'd love to hear from others who have done the same.

The VirSyn Cube was mentioned here recently...

http://www.virsyn.com/en/E_Home/e_home.html

...but at that time version 1.5 (which does resynthesis
out of the box) was not yet available. I see it's now
available, and I'm off to download the sound samples
and demo.

I have here the Feb. issue of Electronic Musician magazine,
which has a review of Cube 1.01. That version does support
inharmonic spectra, but apparently in a very limited way
which would not be useable for Sethares-style synthesis.
However, maybe 1.5 has expanded here... VirSyn's site lists
"Each partial has internal envelopes for Frequency" as a
new feature in 1.5. . .

-Carl

Hello Carl,

> Has anybody seen this, or know about its microtuning
> support?
>
> http://www.camelaudio.com/productsCA5000.htm

They have a forum at:

<http://www.kvr-vst.com/forum/viewforum.php?f=32>

Ben, the technical director at Camel Audio, is replying to all demands and questions.
I just created a topic called "Alternative tunings" and Ben replied 2 days later.
Feel free to create a new one for a partials tuning system! :-)

Kind regards,
Philippe

I tried out the demo a while back. It's pretty nice. Cube had a few more of the
features I was interested in . Reviews here: http://www.kvr-vst.com/#news1533
all give it fairly high ratings. Especially for sound quality. I don't know
anything about the microtuning capabilities.

On Sun, 15 Feb 2004 14:45:39 -0800
Carl Lumma <ekin@...> wrote:

> Has anybody seen this, or know about its microtuning
> support?
>
> http://www.camelaudio.com/productsCA5000.htm
>
> I downloaded all the sound samples, and I was fairly
> impressed. I've been looking for a good additive synth
> for a long time.
>
> In particular, I'm interested in being able to specify
> the tuning of the partials, Sethares-style (hi Bill!).
> And in particular, a crucial feature would seem to be
> a 'partial-tuning map' abstraction, which would allow all
> the the patches on the instrument to be quickly 'retuned'
> together.
>
> I tend to doubt anything does this out of the box, though
> I guess it could be done in CSound, Max, or Kyma. If
> anybody knows of such a thing I'd like to hear about it.
> Cameleon does advertise "64 detunable partials", and I've
> downloaded the demo. I'll write back when I've tried it,
> and I'd love to hear from others who have done the same.
>
> The VirSyn Cube was mentioned here recently...
>
> http://www.virsyn.com/en/E_Home/e_home.html
>
> ...but at that time version 1.5 (which does resynthesis
> out of the box) was not yet available. I see it's now
> available, and I'm off to download the sound samples
> and demo.

:} Next time I come up with spare cash... I'm buying Tera as well.
{Provided that {spare cash} actually happens.}

> I have here the Feb. issue of Electronic Musician magazine,
> which has a review of Cube 1.01. That version does support
> inharmonic spectra, but apparently in a very limited way
> which would not be useable for Sethares-style synthesis.
> However, maybe 1.5 has expanded here... VirSyn's site lists
> "Each partial has internal envelopes for Frequency" as a
> new feature in 1.5. . .
>
> -Carl
>
>
>
> [MMM info]------------------------------------------------------
> More MMM music files are at http://www.microtonal.org/music.html
> ------------------------------------------------------[MMM info]
> Yahoo! Groups Links
>
>
>
>
>
>

--
See these tears so blue. An ageless heart that can never mend.
Tears can never dry. A judgement made can never bend.
...
And I've been putting out fire with gasoline...
{Dave}

--- In MakeMicroMusic@yahoogroups.com, Carl Lumma <ekin@l...> wrote:

> I tend to doubt anything does this out of the box, though
> I guess it could be done in CSound, Max, or Kyma.

It can easily be done in Csound. What I haven't found so easy is
getting a really nice, musical sound.

G,

{you wrote...}
>It can easily be done in Csound. What I haven't found so easy is getting a >really nice, musical sound.

That is one of the things that impresses me about Prent Rodgers work - the musicality of the samples (and sample manipulation) he utilizes. Maybe Prent could shed light on this, but I also know that the quality of the samples is paramount, and in this regard it might mean looking for a good (and possibly pricey) sample library.

Cheers,
Jon

>> I tend to doubt anything does this out of the box, though
>> I guess it could be done in CSound, Max, or Kyma.
>
>It can easily be done in Csound. What I haven't found so easy is
>getting a really nice, musical sound.

Aren't there nice, musical additive patches/code available for
CSound? Then, if it has this functionality, one would simply
apply the desired tuning map, and it shouldn't change the sound
much unless you're going for a macrotemperament.

-Carl

>>It can easily be done in Csound. What I haven't found so easy is
>>getting a really nice, musical sound.
>
>That is one of the things that impresses me about Prent Rodgers
>work - the musicality of the samples

It isn't hard to get samples to sound roughly like musical instruments.

>(and sample manipulation) he utilizes. Maybe
>Prent could shed light on this, but I also know that the quality of
>the samples is paramount, and in this regard it might mean looking
>for a good (and possibly pricey) sample library.

He discusses some of his tools on his blog. But nowhere have I
heard in Prent's music the ultra-realism for which the quality
of samples is paramount.

-Carl

C,

{you wrote...}
>It isn't hard to get samples to sound roughly like musical instruments.

I don't care about roughly. I mean *really* musical, and I may be more concerned about it than other people. I have two colleagues that are active composers in the film/television world, where sampled instruments are used a lot (as I'm sure you know already). The amount of not only care spent on the samples themselves, but *how* they are used (both in the playing/programing, and the choices) is night and day. I've heard people use Larry's gear to do their own demos, and it sounded like they tossed out all the good stuff and brought in junk - all due to not having his skill with those sounds/tools.

>He discusses some of his tools on his blog.

In the archives of MMM somewhere is a slightly more detailed discussion of how he worked with the samples. I'm pretty sure they were from the McGill library, and I think Prent downplays his use of them more than I tend to give him credit! :)

>But nowhere have I heard in Prent's music the ultra-realism for which the >quality
>of samples is paramount.

Well, our ears are slightly different, aren't they. Besides, Gene was talking about musicality, not realism - two dissimilar pieces of produce. OTOH, Gene usually is mimicing (sp?) real-world acoustic instruments, (I remember his search for a great piano), so again it is worth investigating good sample manipulation.

Cheers,
Jon

>>But nowhere have I heard in Prent's music the ultra-realism for
>>which the quality of samples is paramount.
>
>Well, our ears are slightly different, aren't they.

Very.

>Besides, Gene was talking about musicality, not realism - two
>dissimilar pieces of produce. OTOH, Gene usually is mimicing (sp?)
>real-world acoustic instruments, (I remember his search for a great
>piano), so again it is worth investigating good sample manipulation.

Gene and I were talking about additive synthesis, not sampling.

-Carl

--- In MakeMicroMusic@yahoogroups.com, "Jonathan M. Szanto"
<JSZANTO@A...> wrote:
> G,
>
> {you wrote...}
> >It can easily be done in Csound. What I haven't found so easy is
getting a
> >really nice, musical sound.
>
> That is one of the things that impresses me about Prent Rodgers
work - the
> musicality of the samples (and sample manipulation) he utilizes.
Maybe
> Prent could shed light on this, but I also know that the quality of
the
> samples is paramount, and in this regard it might mean looking for
a good
> (and possibly pricey) sample library.

Samples is a whole different topic.

--- In MakeMicroMusic@yahoogroups.com, Carl Lumma <ekin@l...> wrote:

> Aren't there nice, musical additive patches/code available for
> CSound?

What CSound has are "instruments"; normally these are of baroque
complexity, but can be simplified in practice by setting some of the
control parameters to fixed values. Probably I should try CSound
again; I was trying to produce what I thought were nice-sounding,
additively synthesized instruments which could be adjusted to the
temperament, and was using nonkleismic and starting with my own
instruments from scratch. I eventually gave up in disgust after
writing some nonkleismic CSound music.

Then, if it has this functionality, one would simply
> apply the desired tuning map, and it shouldn't change the sound
> much unless you're going for a macrotemperament.

Tuning is not a problem for CSound--you simply tell something how
many Hertz, and it goes and does it.

--- In MakeMicroMusic@yahoogroups.com, "Jonathan M. Szanto"
<JSZANTO@A...> wrote:

> OTOH, Gene usually is mimicing (sp?) real-world acoustic
instruments, (I
> remember his search for a great piano), so again it is worth
investigating
> good sample manipulation.

I use acoustic-instrument samples mostly, but not exclusively.
Electronic samples are hard-pressed to match acoustic ones in audible
interest value.

>> Then, if it has this functionality, one would simply
>> apply the desired tuning map, and it shouldn't change the sound
>> much unless you're going for a macrotemperament.
>
>Tuning is not a problem for CSound--you simply tell something how
>many Hertz, and it goes and does it.

Yes, but can you write code that allows the kind of modularity
required for the feature I suggested? In your first reply, you
seemed to think this was no problem. If this is the case, then
you could simply download somebody else's instruments and retune
their partials.

-Carl

Carl,
What methdology are you looking to use? There are many descriptions
of using Additive Synthesis with Csound. Most describe a set of
oscillators, each at a different frequency, amplitude, and envelope.
See for example:
http://www.sfu.ca/sca/Manuals/Csound/Hamel/Week4.html
http://www.uweb.ucsb.edu/~johnt/csound.html
http://www.web42.com/crenz/en/data/additive_synthesis.html

You can hand code the parameters in csound scores and orchestras or
write a program to generate them based on the methodology you
propose.

I have chosen to start with samples to try to create realistic
instruments, mostly to avoid sounding electronic. I have varying
degrees of success in this. I still have to make dozens of tweaks
to get an interesting sound.

I am always amazed at how dull the samples sound by themselves. Its
almost as if the musicians are trying to sound dull just to ensure
their role in the future of music. But in reality, a good musician
makes thousands of decisions for each note, most of them
unconscious, and all based on the musical context. Context is a
tough one to code in csound.

My experience with additive synthesis, from the 1970's using Timbre
Tuning on a PDP-11 at UCSD, is that the output sounded very
electronic. I am sure people have gone beyond that by now. Do you
have any examples on the web I could listen to?

>
> Yes, but can you write code that allows the kind of modularity
> required for the feature I suggested? In your first reply, you
> seemed to think this was no problem. If this is the case, then
> you could simply download somebody else's instruments and retune
> their partials.
>
> -Carl

Prent Rodgers

--- In MakeMicroMusic@yahoogroups.com, Carl Lumma <ekin@l...> wrote:

> Yes, but can you write code that allows the kind of modularity
> required for the feature I suggested? In your first reply, you
> seemed to think this was no problem. If this is the case, then
> you could simply download somebody else's instruments and retune
> their partials.

Most people's instruments are not using additive synthesis of sine
waveforms.

Heya Prent,

>What methdology are you looking to use?

See:
/makemicromusic/topicId_5757.html#5757

>There are many descriptions
>of using Additive Synthesis with Csound. Most describe a set of
>oscillators, each at a different frequency, amplitude, and envelope.
>See for example:
>http://www.sfu.ca/sca/Manuals/Csound/Hamel/Week4.html
>http://www.uweb.ucsb.edu/~johnt/csound.html
>http://www.web42.com/crenz/en/data/additive_synthesis.html

Thanks for the pointers!

>You can hand code the parameters in csound scores and orchestras or
>write a program to generate them based on the methodology you
>propose.

Good, that's as I thought.

>Context is a tough one to code in csound.

It can be used to 'render' MIDI files, though, correct? IIRC
there's a MIDI->CSound (orc?) utility around (that Bill Alves
has used) for doing this.

I'm also interested in realtime performance, which short of
something like a Creamware card (which I don't think exists
anymore) isn't possible with CSound to my knowledge (or are
the machines these days fast enough for limited applications?).

>My experience with additive synthesis, from the 1970's using Timbre
>Tuning on a PDP-11 at UCSD, is that the output sounded very
>electronic. I am sure people have gone beyond that by now. Do you
>have any examples on the web I could listen to?

So far I've used samples to do all the MIDI 'rendering' I've done,
as I currently don't have an additive solution on the PC.

One way around the electronic-sounding problem is to do resynthesis.
You might listen to the Cameleon demos and see what you think.

I do have a Kawai K5000s additive synth, which I've played a
few gigs with. It is fantastic for all manners of organs
(naturally), though it also has a great clarinet. Unforch, it
isn't multitimbral or microtunable.

-Carl

Carl said:

> See:

> /makemicromusic/topicId_5757.html#5757
For this set of ideas, you would definitely need some time with
Csound and an external program to take the tuning and impose it on
the orchestra. This is a few dozen hours of work, at least. But it
would definitely give you an opportunity to hear what is possible.
I've always been amazed at what people can do with Csound.

> It can be used to 'render' MIDI files, though, correct? IIRC
> there's a MIDI->CSound (orc?) utility around (that Bill Alves
> has used) for doing this.

Csound now reads Midi instruments and midi file directly, with some
opcodes. I haven't tried that. Look up "midi csound" on google for
examples. Some have reported serious problems with the support, but
others use Csound for midi all the time. It depends what you are
trying to do.

>
> I'm also interested in realtime performance isn't possible with
> CSound to my knowledge (or are the machines these days fast enough
> for limited applications?).
>

Most machines today of 1.5 gHz and up can do realtime. There is a
Csound benchmark around that creates some very complex instruments
one at a time until the sound breaks up. My laptop died at 14 notes,
others have gone to 20 or more. It's not infinite, or even
predictable, but it is used by many people.

Prent

--- In MakeMicroMusic@yahoogroups.com, "Prent Rodgers"
<prentrodgers@c...> wrote:

> Csound now reads Midi instruments and midi file directly, with some
> opcodes. I haven't tried that.

Eh? When in the world did it start doing that? A year ago you needed
to convert the midi file to Csound format, which left off a lot of
information.

By the way, if you want to use Setherized partials, you can't just
use a single periodic waveform of the type you get from adding
partials (using Csound GEN10.) You need to take sine waves from GEN10
and add them together yourself. No need to worry about cosine waves,
since the phase relations will shift anyway. However you can steal
someone else's set of overtone amplitudes if you think those are
nifty.

Got to go and check what's up with Csound and midi...