At 4:54 AM +0000 5/7/03, cormallen wrote:
> > Additive synthesis has existed for quite a while now- especially in
> > the academic sector - 'modulars' or software based ones for that
> > matter have been around for quite awhile. I've played one down at
> > Univ of IL back in the later seventies when I was going to go there
> > after HS to major in the veterinary school (love animals) and was
> > going to minor (diminish?) in Emusic. Its a LOT of wanking around
> > with minimal effect - my general proof: no major synth manuf. are
> > pursuing it anymore.
While additive synthesis is very old (I have schematics for a tube
additive synth from 1931), and even if you ignore organs with
drawbars, there have been many electronic additive synths, such as
the Technos Axcel, Crumar, DK Synergy, etc, and a number of
"workstations" that included additive as part of their arsenal --
Fairlight, PPG, Synclavier, Kyma, etc.
Afaik, no major synth manufacturer is making FM, DCO, or top-octave
synths either. So while I agree additive synthesis isn't that
practical ("a lot of wanking around with minimal effect"), I disagree
with your reasoning.
>I have a K5000 and it's a pretty cool synth, but for me the best part
>of it is the formant filter, not the classical additive stuff. Also,
>it's extremely time-consuming to program.
For a while, I was thinking of buying a K5000R, but yes, they are
difficult to program and imho the sound is kind of weak. I guess I
liked the idea that it did something different.
>That, I think, is the main problem with additive synthesis; there are
>too many 'knobs', none of which make much noticable difference to the
>sound. To effect a broad timbral change requires altering dozens of
>knobs at once.
I agree, and I feel the same way about Chowning FM. Even if you
took a DX7 and gave it a knob for every parameter, it still would be
a PITA to program.
I think both FM and additive synthesis are good candidates for software
control. Like a cerebellum, the software would translate a
high-level command into a multitude of low-level changes, so an
ergonomic interface with a manageable amount of knobs and switches
could control the actual synth parameters. Instead of having
hundreds of knobs for the actual parameters, it would have knobs for
qualities like Wood, Metal, Bright, Impulse, Shape, etc. This would
allow the development of different user-selectable "skins" for the
same synth.
>I've been intrigued lately by the Neuron for just this reason. It
>seems to be an attempt at a resynthesis scheme in which there are only
>a few knobs, but where those few really change the sound dramatically,
>and in quite novel ways. Also, it's always nice to see something
>genuinely original in synthesis...
I've read about it, but I haven't actually seen one. Aren't they
like $5K?? I'm extremely reluctant to spend that much $$$ on
anything digital.
>(I recently wrote a little Linux app which uses genetic programming
>techniques to 'breed' sounds in the fourier domain and which writes
>the results out as samples - cool, semi-random stuff for when I have
>∗no idea∗ what sound I want in a piece. The intention was to have it
>write out Halion multisamples, but Steinberg won't release the Halion
>file format. If anyone's interested, I can port it to Windows pretty
>easily...).
Porting it to OS X would be easier :)