Hmmm ... maybe I misinterpreted something in one of your papers. I'll have
to do a web search now and find the exact quotation. In any event, CSound is
certainly established and there is a wealth of existing CSound code
unmatched by SAOL. And there's nothing in the SAOL world (yet -- sly grin
:-) to match Extended CSound, with the ability to use offboard DSP chips.
I've been saying all along that whether you use CSound, SAOL or any other
*pure software* synthesizer, there are real limits to the kind of music you
can create even in the fastest PCs, especially if you want to perform in
real time. In other words, it's best to think of even a GHz PC as a control
processor, and invest in some kind of offboard DSP hardware. I now own three
synthesizers: a Yamaha VL70-m, which uses physical modeling and is
monophonic, and two Alesis NanoSynths, which use sample tables and are
polyphonic. The beauty of an arrangement like this is:
1. Low cost -- the NanoSynths are about $150 street price (if you can find
one :-) and the VL70-m is about $600. DSP-based synths like the Kyma run
about $2500.
2. The PC only has to generate control-rate signals. In the case of the
VL70-m, essentially all that has to be sent is the parameters of the
physical model, and some pitch bend data in the case of microtonal music. I
can do this in almost any language that can send a MIDI stream to my USB
port. The Nanos are a little more complex; they're essentially General MIDI
renderers as shipped, but they can be patched on the fly to some extent.
The big win, though, is that I don't *have* to do any DSP programming. I am
certainly *capable* of doing DSP programming, but it is time-consuming and
without a large library, like the one built into CSound, it's a thankless
task. And since the synths are designed from the start to be musical
instruments, with a human musician driving them from a keyboard or, in my
case, a wind controller, they're already performance-ready!
The goal of my efforts is to produce microtonal and xentonal musical works
that can be performed by a soloist and accompanied by a computer in a
real-time performance setting. When the soloist is playing an electronic
instrument, the computer, in addition to providing the accompaniment, is
responsible for holding the solo instrument to the particular tuning scheme
desired. I can *code* all this in SAOL or CSound, but can the computer keep
up with me if it also has to generate the samples for the accompaniment in
real time? Somehow, for *interesting* accompaniments (I'm not talking
electric bass and a drum kit here :-) I rather doubt it.
(Well, maybe a two-processor 1 GHz system running Linux, with one processor
handling the OS and the MIDI streams and the other one working out of
locked-down memory generating the sample stream would have a shot at it. But
now we're talking $2000 right there for the server, plus many days of
coding, debugging, version controlling, testing and other real-time systems
and applications software engineering tasks. If the end result doesn't sound
a *lot* better than the VL70-m and the NanoSynths, what then? :-)
As I mentioned before, I'm on another list devoted to microtonal and
xentonal music. Those folks recognize the power of CSound and SAOL to
generate the precise control of pitch and timbre this style of music
requires. Many are mathematically inclined, many have the software
engineering skill set to pull off a work of this nature. But a lot of them
are musicians ... they don't want to code, they don't really even want to
push buttons, mouse around on a GUI or twiddle knobs. They want to play a
keyboard or a wind controller or wave a Buchla Lightning wand around on a
stage in front of a live audience. I think for these folks, Extended CSound,
with a suitable wrapper for non-programmers would be just the ticket.,An
"Extended SAOL" wouldn't have any significant advantage, since they wouldn't
care about the underlying syntax or semantics. But ask them to spend weeks
in a studio producing a seven-minute canned tape piece, with minutes or
hours between a parameter change and hearing the music produced by that
change, and you lose them.
-- M. Edward (Ed) Borasky, Chief Scientist, Borasky Research http://www.borasky-research.com http://www.aracnet.com/~znmeb mailto:znmeb@borasky-research.com mailto:znmeb@aracnet.com"There's no fuel like an old fuel" -- the National Coal Institute
> At 05:54 PM 1/19/01 -0800, znmeb@teleport.com wrote: > >I do not actively use CSound, and in > >fact Barry Vercoe, the CSound inventor, has said that he believes CSound > >users should migrate to SAOL. If it's good enough for Barry Vercoe, it's > >good enough for me :-). > > I don't believe I actually said that. But I am willing to wait & > see whether > SAOL develops the music/audio processing maturity of say Extended > Csound, on which I've spent most of my time these past four years .... > > -- Barry
This archive was generated by hypermail 2b29 : Mon Jan 28 2002 - 11:46:41 EST