comp.lang.idl-pvwave archive: archive » Re: large matrix operations

Home » Public Forums » archive » Re: large matrix operations

Show: Today's Messages :: Show Polls :: Message Navigator
E-mail to friend

Re: large matrix operations [message #62886 is a reply to message #62872]

Mon, 13 October 2008 15:14

Michael Galloy
Messages: 1114
Registered: April 2006

Senior Member

On Oct 13, 3:17 pm, Vince Hradil <vincehra...@gmail.com> wrote:
> Thanks for cleaning that up. I think the conclusions are the same -
> matrix_multiply does not speed things up. I wonder what happens if we
> really push the envelope wrt RAM and matrix size. Unfortunately, I
> don't have a lot of time to play with this.

MATRIX_MULTIPLY is about 15% faster in my tests, but I guess you were
wanting a *very* significant speedup i.e. like the speedup you get
when vectorizing code?

Mike
--
www.michaelgalloy.com
Tech-X Corporation
Software Developer II

Report message to a moderator

[Message index]

		Re: large matrix operations By: Vince Hradil on Tue, 14 October 2008 06:10
		Re: large matrix operations By: Michael Galloy on Mon, 13 October 2008 15:14
		Re: large matrix operations By: Vince Hradil on Mon, 13 October 2008 14:17
		Re: large matrix operations By: Michael Galloy on Mon, 13 October 2008 13:51
		Re: large matrix operations By: David Fanning on Mon, 13 October 2008 10:39
		Re: large matrix operations By: Vince Hradil on Mon, 13 October 2008 10:07
		Re: large matrix operations By: Vince Hradil on Mon, 13 October 2008 09:00
		Re: large matrix operations By: Vince Hradil on Mon, 13 October 2008 08:08
		Re: large matrix operations By: David Fanning on Mon, 13 October 2008 08:04

Previous Topic:	multislice
Next Topic:	Catalyst Library Going Open Source

-=] Back to Top [=-

[ Syndicate this forum (XML) ] [

] [

]

Current Time: Fri Oct 17 21:30:57 PDT 2025

Total time taken to generate the page: 0.30804 seconds