Expressive sound synthesis?

[Max M.]

but rather they've been abstracted to meaningful parameters such as "material", "stiffness" in this synth:.

Exactly. This is just because most of physical modeling algorithms are made to directly derive "the coefficients" from things like "material", "stiffness" or "dump".
Contrary, nothing in an FM or PWM synth can be thought of as a "material". It's not a problem to make some FM patch where certain parameter (or set of parameters) can be mapped to a knob named "material", but this will apply to this particular patch only, thus for this thing to make sense you'll have to hide anything else and limit tweaking to those predefined patched with predefined "macro" parameters. It's nothing wrong with such approach and such instruments were made from time to time, but apparently they just did not get popular.

Actually, even for a Physical Modeling synths, those parameters are not always as friendly you could expect (unless the synth is intentionally limited to two or three knobs and the rest of parameters are totally hidden from you). For example, in the Chromatophone choose any "more abstract" sound preset (anything that does not sound like bell/drum) and you'll find that neither "material" nor "stiffness" parameters has anything to do with "material"/"stiffness" anymore.

[aMUSEd]

I'd like to think about synthesis more in the terminology of "sound expression", rather than in parameter or scientific language.

There is a UK group developing free plugins based a similar idea. "The plug-ins allow you to save and load parameter settings semantically, meaning you can type things like warm, fuzzy and bright into the text box and the computer will try and figure out what you mean!
http://www.semanticaudio.co.uk

Hah. Great. Never heard of these guys. Thanks.

Only problem is these terms are so subjective, for example some people refer to 'dark' sounds in terms of tonal qualities (opposite of 'bright') while others use a more emotional definition, as in 'dark' ambient.

http://tamingthesaxophone.com/bright-sound-dark-sound

[Michael L]

Only problem is these terms are so subjective

Totally. They are crowdsourcing the definitions.
Personally, I am more curious about the Chair-to-Computer modulation parameters:
https://www.youtube.com/watch?v=JnLylwvWIpE

[Sendy]

I'd like to think about synthesis more in the terminology of "sound expression", rather than in parameter or scientific language.

There is a UK group developing free plugins based a similar idea. "The plug-ins allow you to save and load parameter settings semantically, meaning you can type things like warm, fuzzy and bright into the text box and the computer will try and figure out what you mean!
http://www.semanticaudio.co.uk

Hah. Great. Never heard of these guys. Thanks.

Only problem is these terms are so subjective, for example some people refer to 'dark' sounds in terms of tonal qualities (opposite of 'bright') while others use a more emotional definition, as in 'dark' ambient.

http://tamingthesaxophone.com/bright-sound-dark-sound

Exactly. In the future maybe the computer will be able to read our mind and construct the sounds we hear in our heads, but at that point, the computer will be working harder than the "imaginer". Part of the process of creating ideas in the real world is that you have to make sacrifices - there's no 100% analog (no pun intended) of that one sound you're hearing in your head, in the real world. By adapting your thoughts through the lense of tehnical parameters you're exploring the boundary between the objective and subjective, and that's what makes art worthwhile in my opinion.

Learning the technical stuff may seem like boring trial and error, but I've found it to be the most fulfilling part about creating electronic music.

[aMUSEd]

Only problem is these terms are so subjective

Totally. They are crowdsourcing the definitions.

I'm not sure they are, that's the problem. I pointed this out in the original thread on these, the way they work is they crowdsource what people call a particular sound, but if different people are using a different definition of say 'warm' there is no way for them to sort those out, so you just end up with a definition by committee, which will be a bit of everything.

[Sendy]

It may even be a bit more insidious than that, because to become fluent with this new technology, you'll have to relearn how you think about sound and effectively "normalize" (again, no pun intended ) your thought process, losing all the associations and quirks that make your perception of sound unique. That doesn't sound like a noble aim for an artist.

[Ridan]

While this is rather old news, i'm reminded of Alvin Lucier's experiment with brain waves and amplification. Wiki - "EEG electrodes attached to the performer’s scalp detect bursts of alpha waves generated when the performer achieves a meditative, non-visual brain state. These alpha waves are amplified and the resulting electrical signal is used to vibrate percussion instruments distributed around the performance space. Other important early pieces include Vespers (composition)|Vespers (1968), in which performers use hand-held echolocation devices to locate the approximate physical center of a room, to deepen their understanding of acoustical perception, and to reveal the elements of environmental space through non-visual means."

https://www.youtube.com/watch?v=bIPU2ynqy2Y

[soundmodel]

Exactly. In the future maybe the computer will be able to read our mind and construct the sounds we hear in our heads, but at that point, the computer will be working harder than the "imaginer". Part of the process of creating ideas in the real world is that you have to make sacrifices - there's no 100% analog (no pun intended) of that one sound you're hearing in your head, in the real world. By adapting your thoughts through the lense of tehnical parameters you're exploring the boundary between the objective and subjective, and that's what makes art worthwhile in my opinion.

Learning the technical stuff may seem like boring trial and error, but I've found it to be the most fulfilling part about creating electronic music.

But I'm also thinking that there might just be a "gap" between some engineers designing the instruments and the musicians making music with them. I've come across some plug-ins that have really seemed like they were designed like a scientific instrument, rather than a musical apparatus.

Musical apparatuses benefit from being adjusted to musicality. That's why many people use presets as well, because they express "ideas about music". Even a compressor plug-in can be made to express musical ideas, but one needs to constrain its parameters in such way they are "musically meaningful", rather than scientific.

[soundmodel]

Or e.g. why can't sounds be designed like an engineer designs a machine. A machine is probably more well defined than a sound, it has more theory and it's more concretely conceivable.

Try writing on a paper what a certain FM synth sound consists of.

[Sendy]

Oh, for sure. I appreciate plugins that give readouts in musical notes as opposed to frequencies, and beat values instead of milliseconds. Patch morphing is great (though of course you still need to specify the states to morph to via the usual means - and knowing how best to specify these tehnically will allow you to create morphs with the most impact or virtuosity). I used to be against automatic beat slicers, preferring to do things by hand. But at the end of the day, sometimes using tools helps you do more with your time - provided you don't let the tools numb your awareness such that you become dependent upon them.

I just think there's a danger of taking layers of interface comfort too far. There's a talk by chiptune artist Danimal Cannon that explores the area of using technicality to aid musicality really well, and he touches on this oversimplification in DAWs towards the end of the talk quite well.

https://www.youtube.com/watch?v=_7k25pwNbj8

[pottering]

Historically electronic(/electric/mechanized) music instruments were created by performers and composers who felt limited by the existing instruments, they felt they had reached the limits of what could be done with them.

They wanted new instruments that could put into reality new music ideas that couldn't be performed by human hands. Some wanted super-human precision, others super-human speed, and so on.

ref:
http://www.luminosoa.org/site/books/det ... new-music/

Nothing bad about more ways to control parameters in a computer, of course.

[Max M.]

there might just be a "gap" between some engineers designing the instruments and the musicians making music with them.

Nope. You barely find any engineer in this area who is not a musician himself. Actually the gap is between "less-technically-advanced" users and "more-technically-advanced" ones. It's a well-known fact that 99.9% of synth/sampler users never tweak anything but use only presets (sometimes with a minimal tweaking), and still the synths with more tweaking stuff (no matter how musical or technical it is) are more useful for "preset users", simply because they can have more presets created by another ("more-technically-advanced") users, while with a "non-tweaking" instrument they are limited to presets provided by the manufacturer...
All above of course is in very simplified/rough words (not counting a lot of other not less important factors like usability etc. etc.).

In other words, I'd say it's not really about how "musical" or "scientific" is a name of a basic parameter (because in both cases it's usually too abstract anyway), but more about providing some additional tweaking/abstraction control on top of the basic parameters (e.g. "morphing" like mentioned in prev. posts), and then about making all this as friendly/usable as possible (i.e. it's more about generic "usability" of a product, and not about "material" vs. "osc a frq cv" lingvo).

[Ridan]

It's a well-known fact that 99.9% of synth/sampler users never tweak anything

I don't really have a solid opinion on this one way or another, but am curious where you found these statistics?

[Max M.]

It's a well-known fact that 99.9% of synth/sampler users never tweak anything

I don't really have a solid opinion on this one way or another, but am curious where you found these statistics?

This statistic is not that magic to estimate for one being in the industry long enough (I remember the times when most of musicians had nothing to do with sound at all :)

[Michael L]

It's a well-known fact that 99.9% of synth/sampler users never tweak anything but use only presets

Hmmm.... A major plugin developer emailed me, "It's interesting how half of the audio community are geeks and the others are, well, basic users" and this 50/50 also seems closer to what I gather from blogs.

Some wanted super-human precision, others super-human speed, and so on.
ref:http://www.luminosoa.org/site/books/det ... new-music/

That is a very interesting free downloadable book. It concludes,
Writing at the turn of the millennium, the American composer Ron Kuivila proposed three strategies for coping with the destabilizing ephemerality of musical technology: getting “under” it by working with basic acoustic principles, looming “over” it by devising works abstractly, or going “into” it by probing the latent potentials residing in familiar objects and ostensibly obsolete technologies.

.. and then - speaking of ostensibly obsolete - presents this historically accurate 1950s electroacoustic studio app (mac only/no presets/irony):

Berna2.jpeg