Author: Mike S.
Date: 17:25:39 04/27/03
Go up one level in this thread
On April 27, 2003 at 19:58:37, paul bedrey wrote: >If so could you post the results? I ran it in Fritz8 gui using the analyse >testset option. I set the extra ply option to 4 ply. Since the definition of the WM test is *20 minutes per position*, and the solving time is the (last) decision time after which the solution is kept until 20:00, maximum time should be set to 1200 seconds and extra ply to 99, to ensure that the full 20 minutes are used. This may take a while :o)) 33 hours... 4 extra plies may be too less, because i.e. the engine may show the intended solution move from ply 3 to 7, which is within a few seconds, but could have "changed it's mind" after some minutes later, thus missing the correct solution... It's a common observation in the WM test, that engines are changing to the correct move, then again to another move, etc. back and forth during the testing time. >The result of 61 solved on >my P-4 1800 seemed reasonable but the average time for the 61 positions was >about 2:23 which seemed to good to be true. Can anyone verify this result? Manfred Meiler has reported 64 solutions for an Athlon 1400: http://f23.parsimony.net/forum50826/messages/62277.htm Seems reasonable... I think, an Athlon@1400 MHz may have a slightly better chess performance than a P4/1800 MHz. Regards, M.Scheidl
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.