Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: To Vasik - What is the progress of MP Rybka ?

Author: Vasik Rajlich

Date: 05:18:35 01/19/06

Go up one level in this thread


On January 19, 2006 at 03:48:34, Tord Romstad wrote:

>On January 18, 2006 at 22:26:06, Jay Urbanski wrote:
>
>>On January 18, 2006 at 17:55:55, Dann Corbit wrote:
>>
>>>Executables are heavyweight processes that heavily consume resources and threads
>>>are lightweight processes which consume less resources.  But a threaded server
>>>needs any global variables to have write-access gated with a critical section
>>>whereas a global in a spawned server has no effect (since each server is a
>>>single process running a single thread of execution).
>>>
>>>It is better, but much harder, to write a threaded chess engine.
>>
>>
>>Very true but with nearly all processor improvements in the immediate future
>>coming from more cores/threads - it will likely be worth the effort to bite the
>>bullet and learn to write good threaded code at some point.
>
>Why?  Wouldn't a chess engine using multiple processes and
>shared memory be just as fast as a similar program with
>multiple threads on a computer with several CPUs (or CPUs
>with multiple cores)?  I thought the engine using processes
>would just consume a bit more memory, and not have any
>significant disadvantages apart from that.
>
>I'm not saying that you are wrong, of course.  I am totally
>ignorant about this subject, and I ask in order to learn more.
>
>Tord

Dann and Jay are I guess talking about the mp implementation of just sharing the
hash table. It's a very cheap way to get some performance benefit, and I am
thinking about it as a lazy way out for the time being.

Indeed I realize that multi-threaded apps are probably the future. It's just a
question of finding the time to do this (and taking this time away from other
tasks).

Vas




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.