Author: Harald Faber
Date: 06:16:32 09/01/99
Go up one level in this thread
On September 01, 1999 at 05:25:57, Frank Quisinsky wrote: >On September 01, 1999 at 04:44:38, Harald Faber wrote: > >>On August 31, 1999 at 19:09:50, Frank Quisinsky wrote: >> >>>Hello Robert, >>> >>>Result: >>>~~~~~~~ >>> >>>Time from Zarkov (WinBoard debug file) >>> >>>The game Zarkov - Crafty >>>See post from yesterday, PGN and log file from Crafty ! >>> >>>AMD K6-3 450 MHz, WinBoard 4.0.3, Ponder = OFF, >>>Crafty 24/8 MB Hash, Zarkov 8 MB Hash >>>40 moves in 40 minutes ! >>>Game from WT-5 tournament ! >>> >>>I replay the game (Crafty with ponder = on) >>>40 moves in 40 minutes ! >>> >>>Result: >>> >>>2 other moves ! >> >> >>*1* single different move can change the whole game and result! >> >> >>>Of course not enough analyses, but I make >>>this analyses with many programs and I can >>>not see that ponder is 50-100 ELO ! >>> >>>Not more than 20-40 ELO (ponder = off) and >>>10 ELO for an other time manangment ! >>> >>>Kind regards >>>Frank >> >>Really hard to say. >>1 different move cannot change a win into a double win. :-) >>It even can change it to a draw or loss like it can change a draw to loss or win >>or a loss into a draw or a win. >>Very time extensive to find that out. >>And the negative effect (played move X without ponder which is much better than >>move Y with ponder) is certainly not because of ponder but because of search >>depth+eval or didn't get Crafty to ply+n with ponder? >> >>BTW your English is really....improvable. ;-) > >I think, an good message in an bad english is better than an bad message in an >good english ! Sometimes it leads to misunderstandings. >I can make the experiment with games from the WinBoard tournament and I found >not an other result ! > >Kind Regards >Frank How many games did you replay on 2 machines?
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.