Pure synthesis for microtonality (was Re: Communications with NI re. FM7)

> {you wrote...}
>> I don't think sample-playback synths are good for
>> microtonality, which (I
>> believe) by it's nature requires pure synthesis - FM,
>> additive, physical
>> modeling, analog, whatever. My perception is that it's purer from
>> fundamental to overtones.
>
> While 'pure' synthesis solves some of the problems you mention,
> I think you
> need to open up to the wide possibilities of music that can be done
> microtonally. And I say that specifically because we've had
> examples from
> people on this list that extensively use samples, both for emulation of
> real instruments (Prent Rodgers) and for morphing into new territories
> (Bill Sethares / Jacky Ligon) and everything in between.

No, no, I agree ! Big open mind here. Perhaps too open, neurons
sometime seem to fire into empty space ;)

I said, "more realistically, like actual acoustic instruments,
which is a good thing if you happen to write for acoustic
instruments".

Meaning that in an ideal world, using a (good) physical model,
or other pure synthesis emulation, of an acoustic/real
instrument would be preferable to using samples of said
instruments.

But, sample-playback "synthesis" cannot go away soon enough, as
far as I'm concerned.

That would leave sampling as the tool of choice for Musique
Concrète related & other inspirations involving the manipulation
of real-world sounds. This includes Wavetable synthesis, as used
in Waldorf's Microwave machines, or the more recent, Granular
synthesis, etc.

Anyone curious about Wavetable synthesis look here:
<http://www.harmony-central.com/Synth/Articles/Wavetable_101/>,
and if you're not yet curious, listen to Waldorf's demos on the
product pages at <http://www.waldorf-gmbh.de>. Relevant to
microtonality however, I don't know if any of the Microwaves
(other than possibly the "Wave" beastie) allow for the
remapping of samples.

> They are all tools, and some may require more work, but in the
> right hands
> music will occur.
Very True, and my intention is to neither demean nor look down
on people who get the job done with the tools they have access
to. I use a Korg NS5R for my humble sonic scribbling. But, I
plan to replace it with a used TX-802, *if* NI doesn't get
128-note full MIDI tuning tables into FM7 in the mean time.

I recently came across one of the most idiotic articles I have
ever seen by Tom White (President & CEO of the MIDI
Manufacturers Association, MMA). Dating from 1996, the editorial
is entitled, "FM Synthesis is dead". Perhaps we have myopic
twits like him to thank for the MIDI Tuning Standard being such
a well kept secret. <http://www.midi.org/newsviews/fm.htm> Ooh,
look where it's hosted !

Too many stupid articles seem to imply that FM is mostly "good
for making noises".

One can no more exhaust the potential of FM synthesis than you
can exhaust the musical potential of the violin. Even
experimenting with 2-operator software implementation can
demonstrate this.

Anyone with a Mac should try this delightful little stand-alone
application (built in Max/MSP) I came upon while looking up info
on Max/MSP. I don't have the page URL at hand, but the file is
at <http://oak.cats.ohiou.edu/~phillipm/E-A_MusicPrimer.sit>,
about 532kb. It demonstrates FM, Additive, etc. For anyone who
hasn't experienced first hand what these synthesis methods are
potentially capable off, I strongly recommend you try this out.

> Cheers,
> Jon

Cheers,
Joel

Joel wrote: "
>But, sample-playback "synthesis" cannot go away soon enough,
> as far as I'm concerned.

I don't understand the complaint. I have found that good samples open
up a whole new world of expression, variability, and nuance that make
it simple to create very interesting microtonal music. There is a
terrific range of sounds in some good sample libraries. They include
the complex "noises" that musicians have perfected over the years to
communicate. Examine closely the starting transients of the guitar or
violin samples in the McGill University Master Samples and hear how
complex a good sample library can be.

I admit that I have not spent any time working with physical modeling
synthesizers. Do you have example music I could listen to? Csound
supports this, and many Csound fans have asked me to move to Csound
orchestras that don't require integration and purchase of the sample
libraries I use. I have no idea where to start, or the confidence
that it would be worth the effort when I have a good sample library
now.

Prent Rodgers
Freezing in Seattle

On Friday, June 7, 2002, at 05:09 , MakeMicroMusic@yahoogroups.com wrote:

> Pure synthesis for microtonality (was Re: Communications with > NI re. FM7)
>
> Joel wrote: "
>> But, sample-playback "synthesis" cannot go away soon enough,
>> as far as I'm concerned.
>
> I don't understand the complaint. I have found that good samples open
> up a whole new world of expression, variability, and nuance that make
> it simple to create very interesting microtonal music. There is a
> terrific range of sounds in some good sample libraries. They include
> the complex "noises" that musicians have perfected over the years to
> communicate. Examine closely the starting transients of the guitar or
> violin samples in the McGill University Master Samples and hear how
> complex a good sample library can be.

I tried to make my observations clear in both posts on the subject. I don't discount for one second the achievements made in the sample-playback arena. It's not so much a complaint as much as it is an observation. Sample-playback "synthesis" for the purpose of emulating acoustic & electric instruments is a fudge.

Would anyone looking for organ & analog synth sounds choose samples over currently available modeling instruments ? There are people here or perhaps on the *other* list that use the VL-70m & the Oasys PCI, perhaps they can offer their opinion on the physical modeling sounds available to them.

I have briefly encountered IRCAM's Modalys (as yet) non-real-time physical modeling synthesis software for the Mac OS. It makes incredible sounds and is a great deal more advanced & complex in comparison to the method discovered/invented by Stanford & used by Yamaha & Korg. If & when Modalys becomes a real-time system, it will have enormous impact. See <http://www.ircam.fr/>

Wouldn't you like to be able to compose manipulating a virtual piano in which you could whimsically manipulate the material the strings are made of, their dimensions, their tension, the wood the piano is built of, the tuning of each virtual string's fundamental & harmonics, the material the hammers are made of, where the microphones are placed, etc. ?

Can you imagine how handy this could be for someone designing an acoustic or electric microtonal instrument ?

Modalys can do this right now, but not in real-time, unfortunately. It's amazing. I remember when I first tried it out, I was trying to make a drum sound, any drum sound & it wouldn't work. I e-mailed IRCAM, & someone responded telling me that my microphone & beater placement were physically *impossible* & would not make a sound, so I changed a wee bit of code & the most incredibly realistic badly designed & tuned bass drum sound came rumbling thourgh my crappy speakers.

Physical modeling holds great promise, which is being held back, perhaps by processor speeds & misguided commercial interests. Even some publications which should know better routinely judge synths on polyphony & multitimbrality & I'm sure this puts pressure on manufacturers to keep churning out generic character-less instruments with ever more pointlessly impressive specs.

I'm sure many people expected great things after Yamaha's VL-1 appeared on the scene, but after all these years all that remains is the VL-70m (@ US$500 I think). Korg built the Z-1, then the Oasys PCI card, & now all that remains is their MOSS board for their overpriced Triton instruments. I think Yamaha has a software physical modeling synth available for the Windows OS. Try <http://www.yamaha.co.uk/> for Yamaha info & <http://www.korg.co.uk> for Korg info.

> I admit that I have not spent any time working with physical modeling
> synthesizers. Do you have example music I could listen to? Csound
> supports this, and many Csound fans have asked me to move to Csound
> orchestras that don't require integration and purchase of the sample
> libraries I use. I have no idea where to start, or the confidence
> that it would be worth the effort when I have a good sample library
> now.

Can't think of anything off-hand, sorry. There must be demos at Yamaha & Korg's websites though. Composers have used Modalys. In my limited experience Csound doesn't lend itself to intuitive musical use. It is brilliant to teach/learn DSP techniques though. Perhaps you should try & get your paws on a Yamaha VL-70m & see what happens.

>
> Prent Rodgers
> Freezing in Seattle

Sorry about the length of this. Well there isn't much traffic on the list right now...

Cheers,
- Joel