I don't have specific experience with double processor systems and Csound,
but I am a long time Csound user and contributor, and I work as a
programmer. I think the story would be that you will not see large
improvements in performance using Csound on a multi-processing system
because Csound is a single-threaded application. You would see a slight
improvement because one processor could devote itself mostly to running
Csound, and the other could handle the other things happening on the
machine, but this would probably do nothing like double the speed, and might
not even be noticeable.

Csound could be rewritten to take advantage of multiprocessing, but the
rewrite would be difficult and at a low level; it would involve creating a
pool of threads, from which new instrument instances would receive one. This
would improve performance only if the number of instrument instances was
fairly large, because there would some additional processor overhead
incurred by requiring each thread to synchronize with the ksmps period, and
to manage the threads.

I find that with a Pentium II 450 MHz and 128 MB RAM, either plain old
Csound or my AXCsound will do things in real time that used to take 4 to 5
times real time. Simple textures and scores are definitely rendered in much
better than real time. Complex things with adsyn take a reasonable time,
whereas they used to be a pain.

-----Original Message-----
From: Sean Costello 
To: CSOUND 
Date: Friday, March 12, 1999 4:49 PM
Subject: Csound Performance on Multiprocessor Intel Systems?


>Hi all:
>
>Does anyone know about the performance of the various Csound types on
>multiprocessor Intel systems?  I am frustrated with making a change that
>requires 30 seconds of typing, and waiting 20+ minutes to hear the
>results.  Yes, I know that this is a modern miracle, that back in the
>old days you had to walk 10 miles in the snow to put the punchcards into
>the IBM 360, then a week later drive 40 miles to the AD convertors, etc
>etc. - but if I could get an improvement over what I currently have, it
>would drastically improve my productivity, as I could shape compositions
>much quicker.
>
>Would standard Csound's performance be improved with a multiprocessor
>system, or would it have to be rewritten?  Can any of the other Csound
>variants like DirectCsound or the Linux Developers Group Csound take
>advantage of a dual processor system? I don't have any experience with a
>dual processor system, so I don't know if there is a general increase in
>performance, or if the program and/or operating system have to be
>(re)written to utilize the dual processors.
>
>I am not necessarily talking about real-time stuff, which seems best
>left to a Macintosh (or perhaps the BeOS at some point in the future).
>I would like to simply cut down the compiling time for orc/sco
>combinations that use computationally intensive opcodes (sndwarp, pvoc).
>Does anyone on the list have experience with a dual processor system, in
>either Windows or Linux?
>
>Thanks,
>
>Sean Costello (who has stayed up until 4:30 am two nights in a row
>making tiny changes, and waiting 20+ minutes for things to compile)