Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: HyperThreading & Comp v Comp matches

Author: Robert Hyatt

Date: 06:57:59 11/24/04

Go up one level in this thread


On November 24, 2004 at 01:50:18, Uri Blass wrote:

>On November 23, 2004 at 23:10:16, Robert Hyatt wrote:
>
>>On November 23, 2004 at 01:30:47, Uri Blass wrote:
>>
>>>On November 23, 2004 at 01:11:10, Ray Banks wrote:
>>>
>>>>On November 22, 2004 at 23:20:58, Robert Hyatt wrote:
>>>>
>>>>>I use ponder-on all the time, one cpu or not...  If both engines ponder, each
>>>>>gets 1/2 the processor, and that doesn't disrupt normal time allocation...
>>>>
>>>>Why does everyone use ponder off then on their tourneys?
>>>>
>>>>I've previously been told that with a single processor you can't guarantee each
>>>>engine will get 50%, therefore to use ponder off.
>>>
>>>engines that ponder often do not ponder all the time.
>>>They ponder only most of the time and they may not ponder when there is a single
>>>move or when they are in book and engines that ponder in these cases have bigger
>>>advantage relative to their advantage in real games.
>>>
>>>Uri
>>
>>
>>This goes back to the old days of Ed complaining about "doubles" in the SSDF
>>games.  I told him more than once "just solve it yourself and it will be solved
>>for all time.  If you wait for the SSDF to handle doubles, you will be waiting
>>for a long time."  Ditto for pondering.  It is _not_ hard to always have
>>something to ponder...
>
>
>I do not say that it is hard but it is not very important for playing strength
>and there are more important things that I prefer to fix.
>
>Pondering in most of the cases is enough for a lot of programmers.
>I do not ponder when movei has a single move in the pv or when it is in book
>and there are more important things to do to improve the engine.
>
>I have no problem that movei is slightly handicapped by it in real games but I
>do not like to get another handicap by playing on one cpu with ponder on.
>
>Uri


One cpu matches are here to stay.  You can either fix the problem so that you
always ponder, which I did 20 years ago, or you can "get another handicap" when
someone plays with ponder on.

Software engineering dictates that you write code that works for every
circumstance you could anticipate.  It is good for peace of mind.  Only other
thing I would suggest is not having a "ponder off" switch so that your engine
plays like it always plays, each and every time.



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.