Author: Robert Hyatt
Date: 21:36:08 05/29/00
Go up one level in this thread
On May 28, 2000 at 16:34:29, Jorge Pichard wrote: >On May 28, 2000 at 15:56:58, Jorge Pichard wrote: > >>On May 28, 2000 at 13:45:22, blass uri wrote: >> >>>On May 28, 2000 at 13:28:19, Jorge Pichard wrote: >>> >>>>On May 28, 2000 at 07:03:28, Michael Cummings wrote: >>>> >>>>>On May 27, 2000 at 18:04:09, robert michelena wrote: >>>>> >>>>>>Don't forget to adjust the CT.INI to the appropriate settings, and after >>>>>>starting Tiger, go to extra, options , advanced, contempt, -10. >>>>>> >>>>>>Tiger will maul CM7. >>>>> >>>>>Is it CM7000 or CM6000 >>>> >>>>My friend already adjusted the CT.INI by (NSEW=1 and moreSel=1) plus he also >>>>downloaded the learning patch CTFIX.ZIP offered in the Rebel Webpage. The >>>>CMQueen+ Settings was done using the CM6000 engine which we already tested by >>>>pitting it against the CM7000 engine in a 50 games match of 30 min per side, in >>>>which the CM6000 engine won by W25 L17 D8. >>> >>>I understood that the engine is the same and the only difference is that CM7000 >>>cannot use more than 1 Mbyte hash tables. >>> >>>Uri >> >>You are correct, but the test was done using CM7000 original settings vs the >>modified CMQueen+ settings using CM6000 engine and book. >> >>Pichard > >My friend and I can not believe how close this CMQueen+ settings is playing >against Tiger 12.Oe he has been with me ever since Yesterday when he recommended >me to use two Different computers allowing them 25 games each; we usually start >the game around 8:00 AM and run it until 8:00 PM. There was one game which we >discarded it game number 14 since we both saw that CMQueen+ was using the same >Opening with white and after move number 46 both programs have played identical >moves as game number 5, in which CMQueen+ won an endgame were two connected >pawns on the 6 rank were protected by the king in which Tiger lonely knight was >helpless since Tiger king was on the 4 rank unable to recapture the other pawn. This kind of "deletion" mars the science in such tests. Why would you discard such a game? Because one engine is not smart enough to avoid playing a losing line a second time? OK. Then why not discard games where one engine simply doesn't understand kingside attacks and gets attacked frequently. When playing "matches" I think you have to line them up, turn them on, and then let the chips fall where they may. If one engine attempts to repeat won games and the other doesn't attempt to not repeat, that is definitely an "engine weakness".
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.