Author: blass uri
Date: 16:00:20 02/05/99
Go up one level in this thread
On February 05, 1999 at 18:37:30, James T. Walker wrote: >On February 05, 1999 at 17:39:42, Robert Hyatt wrote: > >>On February 04, 1999 at 23:05:04, James T. Walker wrote: >> >>>On February 04, 1999 at 22:05:26, Robert Hyatt wrote: >>> >>>>On February 04, 1999 at 19:55:03, James T. Walker wrote: >>>> ><<<<<< snip >>>> >>>> >>>>I'm not sure what you are testing, but _no_ 16.x version is 100 points better >>>>or worse than another. That is a _serious_ difference, which is lacking in >>>>these. >>>> >>>>By my testing, 16.4 is quite a bit better than prior versions, based on a _lot_ >>>>of engine vs engine games on the chess servers... >>> >The games were played at G/10 or G/5. I do this with early versions just to get >a quick read on how it is doing. If you rember, when I asked you earlier about >where you thought Crafty was at it's best you effectively said to try anything I >wanted. I've tested Nimzo vs Fritz and Junior and they seem to play about as >close at G/10 as they do at 40/90 with maybe Fritz being better at G/5 or G/10 >than the others. Anyway I moved the time up to G/25 and so far Crafty is >leading Nimzo by +9 -6 =7. That's quite a turn around from 25-7 favor Nimzo. I think that crafty as an engine for fritz/Junior is not identical to the real crafty. I found that crafty as an engine for Junior5 is using its time to search even when there is a mate in 1. It is not a practical problem because crafty wins in both cases but I do not like it. I do not know if there are other changes in the crafty as engine for... and I do not know if crafty as an engine for Junior is weaker or stronger than the real crafty. I guess it is not stronger. I am interested to know what is the result of the real not parallel crafty against the same Nimzo with the same openings in 25 min/game. Uri
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.