Author: Robert Hyatt
Date: 10:49:07 01/21/03
Go up one level in this thread
On January 21, 2003 at 07:29:12, Uri Blass wrote: >On January 20, 2003 at 20:56:00, Robert Hyatt wrote: > >>I had posted this, but somehow it failed. I am redoing it as best I >>can. >> >>My overall impression was "Crafty played just fine" in general. It had >>book problems, but no instant book wins or book losses, just some not-so-good >>positions it was able to overcome. >> >>The dual xeon 2.8 was pretty fast. Crafty searched somewhere around 2.5M >>nodes per second, typically, and was generally searching 13-14 plies deep in >>the middlegame, sometimes even deeper. Overall the hyper-threading and xeon >>stuff worked as advertised and produced no problems or quirks. >> >>Speed helps, but it isn't yet absolute, as the game vs Ferret shows. But it >>certainly helps, although a dual 2.8 is not exactly blazingly fast compared to >>some machines like the quad 2.0 machine from Dell, not to mention the itanium2 >>(mckinley) boxes that are around. But it was "sufficient" at the time, of >>course. :) >> >>I went into the event not thinking that there were any "forced losses" but also >>there would be no "easy wins" either. And that turned out to be true. I don't >>think there was an engine present that I could not beat a reasonable number of >>games. IE I would be happy to play anybody there with the full expectation that >>I could win as many as I would lose, if not better... > >I think that the only candidate to beat crafty out of the participants in a long >number of games is ruffian. > >I see no participant except you who used 2.8 gh even on one processor when you >used 2.8 gh with 2 processors so I guess that you had hardware advantage >relative to Ruffian and Yace even if we ignore the fact that you used 2 >processors. > >Uri Hossa ran on a single 2.8ghz PIV for example. I'd bet there were others. Bruce ran on a dual 1.4 athlon the second day. On the first day he ran on something different. There were quite a few "hot boxes" present. And I'd bet that more than one ran on something _faster_ than they claimed they were using too. It _always_ happens.
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.