Author: Robert Hyatt
Date: 21:27:54 06/18/99
Go up one level in this thread
On June 18, 1999 at 05:05:12, Harald Faber wrote: >On June 18, 1999 at 04:51:54, Timothy J. Frohlick wrote: > >>>I'm sure that this has been posted before, but I couldn't find it. >>>So my question is: >>>Does anybody know why CM6000 and CS Tal are not present in Paderborn >>>at the WCCC ? >>>Kind regards >>> >>>K.E. >> >> >>Klaus, >> >>CM6000 has a lousy opening book and wouldn't stand a chance in hell against the >>fierce Paderborn competition. CS Tal II is a lot weaker than any of the >>contestants and with an elo rating of 2300 would be creamed. > >CSTal is said to be in testing phase so no chance to test it in Paderborn... >Don't know why Mindscape didn't manage to go there but certainly they would have >been able to manage a better opening book e.g. Who knows. In fact they don't >need this event to sell their copies. > >>Remember that we are not dealing with SSDF here. We are dealing with machines >>that put out at +1,000,000 nps and have multi-gigabyte endgame tablebases. A >>Volkwagen can go faster than me on my bicycle but it aint no match for a >>Ferrari. > >Therefore isn't it amazing that very slow programs like Lambchop win against the >Ferraris like it did against PConners? >Conclusion: A Ferrari isn't worth anything if it drives into the wrong >direction. However, don't conclude that "fast -> stupid". It is possible to be fast _and_ play good moves. Pconners, Cilkchess, Zug and such have one serious problem that they don't recognize. They don't play enough real games on the hardware they will use. Cilk played a stupid rook-lift type move in the first two games I watched, and suffered as a result. Playing a few dozen games on ICC would have isolated this glitch quickly. You can _not_ test on X, then run on Y, if Y is much faster than X. It just doesn't work. I've been burned by it too many times over the years...
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.