Author: Robert Hyatt
Date: 13:47:14 08/30/99
Go up one level in this thread
On August 30, 1999 at 15:54:31, Frank Quisinsky wrote: >Hello Robert, > >>looking at the log files won't help. It is _not_ only the moves where it >>ponders correctly... those moves save time. And this time can affect >>_other moves_ that may or may not be pondered correctly.. because that >>time that is saved can be used in several ways when needed. Without pondering, >>this extra time is never accumulated, and is never able to influence the timing >>decisions... > >Yes I understand your message ! > >This theme is interesing for me. Today I looking in many log files from Crafty. >I see that Crafty in complicated positions calculate more time. Of course not >this what you mean ! > >Tomorrow I will replay the game Zarkov against Crafty (I postet the PGN and Log >File). I play the Zarkov moves in 1 minute and under 40/40 and in 30 sec under >40/20 (Crafty with ponder on under WinBoard). I think that Crafty not play many >better moves. > >This I will make with 5 Crafty games from the WinBoard 5 tournament (from time >to time). > >Robert, I think real that is not 50-100 ELO and will make experiments. But at >the moment I have one PC :-( > >Kind regards >Frank That is a better experiment... play the same zarkov moves, until crafty varies, then let them play the game out. That will tell you whether it finds better moves with extra time or not... But it will be very time consuming to test this carefully... and that is why I don't test with ponder=off at all...
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.