Author: Jorge Pichard
Date: 02:24:56 01/24/03
Go up one level in this thread
On January 24, 2003 at 04:22:22, Jorge Pichard wrote: >On January 24, 2003 at 03:35:15, Uri Blass wrote: > >>On January 24, 2003 at 02:41:22, Jorge Pichard wrote: >> >>>On January 24, 2003 at 02:00:00, Peter Skinner wrote: >>> >>>>On January 24, 2003 at 00:54:39, Nat Larson wrote: >>>> >>>>>Bareev is only playing against a pent 2.4 gig, why the overkill vs kramnick, >>>>>hell bareev is not that much weaker. >>>> >>>>No reason as to the overkill vs Kramnik, but I am guessing it was a marketing >>>>tactic for the computer supplier. >>>> >>>>Really I don't think Bareev is going to do much vs Hiarcs on a 2.4Ghz. I would >>>>suspect a draw, and wins for Hiarcs. >>>> >>>>All the hardware in the world would not matter as Kramnik showed in the first >>>>few games vs Fritz. >>>> >>>>Programs are still weak when it comes to positional chess, and Kramnik showed >>>>that. No hardware would have been able to save Fritz in those games. >>> >>> >>>That is correct as far as strategical positions are played against Deep Fritz or >>>Deep Junior, but when it comes to tactical positions the extra Ghz can make a >>>world of a difference. Deep Fritz would have done terribly if a mere Pentium >>>500 Mhz was used instead of the 8x 1000 Mhz system, since Kramnik would have >>>outplayed it tactically as well. >>> >>>Pichard >> >>I do not agree. >> >>I believe that the reason that Deep Fritz did not lose the match is not the >>hardware. >> >>I also believe that better hardware may help to get better positional decisions. > >Yes, better hardware also help the program to get a better position much quicqer >within the same time control allowed than using a slower processor, but certain >horizon effect which some program don't have any clue of how to deal with the >position, or blocked positions with pawns becomes useless regardless of the >speed of the hardware. But when dealing with tactical positions the faster >hardware is capable of solving the tactical shot much faster within the same >time control, and this could make a world of a difference when playing against >humans. > PS: The only measure of performance that really matters is the amount of time it takes to execute a given position (tactical or strategical) within the same time control. Certain tactical positions could be solved with a mere 50 Mhz, but if you change the hardware using the same program it will be solved faster. Therefore, in standard time control the faster the hardware the better. Pichard
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.