|
These speeds sound not that different from the
results of existing Csound multi-threading by Steven Yi. What is the difference
in implementation? Your use of the adjective "reliably" indicates something, I
suspect.
Regards,
Mike
----- Original Message -----
Sent: Tuesday, March 24, 2009 7:22
AM
Subject: [Csnd] Re: Multicore computer
music
This sounds very exciting. I am really glad it's
happening.
----- Original Message -----
From: jpff <jpff@codemist.co.uk>
Date:
Tuesday, March 24, 2009 10:34 am
Subject: [Csnd] Multicore computer
music
To: csound@lists.bath.ac.uk
>
> In addition, multicore will become critical for computer
> music,
of that I am
> > sure, enabling more and more realistic physical
modeling at a
> bare minimum.
>
> VERY preliminary, and
not totally the same, but a student here has
> Csound running reliably
in two cores at faster speeds.
> This is his
>
undergraduate final year project so is still incomplete, but it looks
>
promising. Best so far is "Electric Priest" that does from
> 53s
to
> 34s; "Xanadu" is less good at only 10% gain. "Trapped.."
> only shows
> 16% gain. These results are before final
tuning of the
> algorithm. We
> hope to get access to a
quadcore over Easter.
>
> The mechanisms are written up in an
unpublished paper, and will
> be in
> his dissertation in
May/June.
>
> ==John ffitch
>
>
> Send bugs
reports to this list.
> To unsubscribe, send email
sympa@lists.bath.ac.uk with body
> "unsubscribe csound"
Dr
Victor Lazzarini, Senior Lecturer, Dept. of Music,
National University of
Ireland, Maynooth
|