[PlanetCCRMA] Sine waves in real time

Sun Oct 21 15:41:03 2007

On Sun, 2007-10-21 at 10:29 +0100, joey.a wrote:
> Message: 1
> > From: "Bill Schottstaedt" <bil@ccrma.Stanford.EDU>
> > Sent: Saturday, October 20, 2007 8:00 PM
> > Subject: [PlanetCCRMA] sine waves in real time
> >
> > > "Which processor and clock speed are you using?"
> >
> > 3.8 GHz Pentium 4.
> (Context: >> In the Scheme version of Snd, I get about 180 such sines)
> 
> Thanks. That is a fast clock. From this I get that your Scheme algorithm has
> ~ 12% the efficiency of the SuperCollider algorithm, if you are running the
> CPU flat out, or ~ 23% if you are referring to 50% CPU loading.
> Understandable, really, because a look up table should always be faster than
> computing a sin (or even a log) function.

Watch out... a P4 is not comparable to the newer generation of Core Duo
processors for Intel. Cycle for cycle the newer generations of
processors do a lot more work, so comparing GHz has no real meaning. The
best would be to find some standard benchmark and run it on those
processors and then normalize for that (same with Intel vs. AMD, etc). 

-- Fernando

> > > Although I don't yet understand the syntax you are all using, I don't
> think the
> > > 44100 KHz referred to below, has anything to do with the above, which
> > > will largely dictate how well any given software is going to perform.
> >
> > oops - 44.1 KHz -- a typo.
> 
> That sounds more like it! Standard audio sampling rate ;-)
> 
> Message: 2
> From: "Kjetil S. Matheussen" <kjetil.matheussen@notam02.no>
> Cc: <joey.a@accelerators.co.uk>
> Sent: Saturday, October 20, 2007 12:28 PM
> >
> > Its a 4 year old amd xp2800. I could run about 140 voices in realtime at
> > 80% cpu usage,
> 
> Interesting. Your clock is ~ half the speed of Bill's, yet you appear get ~
> the same performance, from the same software.
> 
> Does this mean that AMD is ~twice as efficient as Intel, in music
> applications, or that Bill has specified for only 50% CPU loading?
> 
> Bill?
> 
> > and about 60 voices in non-realtime 100% cpu usage.
> 
> Sorry, this does not compute. Which is the typo?
> 
> _______________________________________________
> PlanetCCRMA mailing list
> PlanetCCRMA@ccrma.stanford.edu
> http://ccrma-mail.stanford.edu/mailman/listinfo/planetccrma