Author: Christophe Theron
Date: 09:35:57 04/24/98
Go up one level in this thread
On April 24, 1998 at 07:28:29, Kevin Mulloy wrote:
>Gentlmen,
> This is my first trip to this site -- and I like what I've read so
>far! I have a question that one of you might be able to answer for me.
>I have the chance to purchase a dual 333 P2 for a very good price. My
>question is, "Would two 333 P2's give a strength advantage over a single
>processor if I were running one of the top programs on it?" I have most
>of the new programs that are available (my personal favorites are
>Hiarcs6! and Genius 5 -- followed by MCP7 and Fritz 5). I'm not very
>well versed on just how a dual processor set up works. If the resulting
>performance gains would be noticible (even 50 elo points would be an
>amazing gain -- in my opinion) I would be willing to "spring" for the
>extra processor and run some test. Does anyone have an opinion on this
>one? If the extra processor will or will not make a difference, what
>are the reasons? In advance, thanks for the input and help guys!
> Trapper
> trapper@netnitco.net
You will gain NOTHING from the additional processor. I don't mean "very
little", I really mean "nothing at all". Zero. Nada.
Current commercial programs are not designed to take advantage of
multiprocessing.
Multiprocessing has to be controlled by the program itself. The
operating system or the computer itself will not do the job of giving
each processor a little bit of the global work to speed things up.
So if your programs are not designed to split the work for 2 or more
processors, only one processor is going to be used, and you get NOTHING
MORE than what you get with a single processor computer.
I know only one available program that can use more than one processor:
Crafty in its latest versions. Because it has been especially designed
to do so.
Christophe
This page took 0 seconds to execute
Last modified: Thu, 15 Apr 21 08:11:13 -0700
Current Computer Chess Club Forums at Talkchess. This site by Sean Mintz.