Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Chessmaster's defects

Author: Roy Eassa

Date: 13:38:00 03/07/02

Go up one level in this thread


On March 07, 2002 at 16:21:01, John Merlino wrote:

>On March 07, 2002 at 15:38:59, Roy Eassa wrote:
>
>>On March 07, 2002 at 14:58:10, John Merlino wrote:
>>
>>>As for the manual (actually, you mean the FAQ), all it explains is that chess
>>>engines will try to get as much cpu speed as Windows will give them; EVERY
>>>decent chess engine does this, not just The King. If you are TRULY concerned
>>>about engines taking more than their share of the CPU, then just run comp vs.
>>>comp games with ponder off.
>>
>>
>>Are you saying that, if a $20 program hogs the CPU more than other competing
>>programs, the user should resolve the problem by running out and spending $1000
>>for a second computer?
>
>Not at all. I said that if you are concerned about unequal distribution/usage of
>the CPU (whether your concerns are valid or not), then just run comp vs. comp
>games with ponder off. That solves the problem and puts both engines on
>identical footing (assuming that both engines use a negligible amount of the CPU
>when idle).
>
>Nor am I agreeing with the accusation that The King "hogs the CPU". Nobody has
>conclusively shown this to be the case.
>
>jm


I shouldn't have used the term "hogs the CPU" as that has a negative implication
(sorry).  How about, "uses substantially more CPU time than other chess
programs," or something like that?  It's not a BAD thing, and you're right -- it
may not even be true.  But if it IS true, I hope there would be a cheaper way to
resolve it in engine-engine matches than having to purchase an entire additional
computer!  People are clever beasts and I have a strong feeling that this issue,
if true, can be resolved a different way.



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.