carldaman wrote:Hi Geert. I would expect EXchess v6.50Bx64 to perform clearly better than its published 2481 rating.
Cheers,
CL
Hello Carl,
My eloratings are not the most relevant ones. The time control of my tournament is rother long so the number of games played by an engine is still very low. EXChess has played 44 games so far from which 50% has been played by version 6.10b.
EXChess started the tournament with 2 draws: against Crafty which I think is really good, and against the surprisingly strong GnuChess. Its next opponent will be ... Komodo .
Komodo is the USA champion! Xpdnt and RomiChess demote to the first division.
3 time losses for Daydreamer. ExChess produces strange evaluation scores. Completely won positions are evaluated with a score of -270, and after 3 such scores, EXChess resigns the game. Even a correct matescore in the wb2uci.eng file (MateScore =10000000) doesn't solve the problem. The results are corrected manually though.
My first post of a tournament always contains the list of participants; also the number of processors is announced there; max number = 6. Crafty, Zappa, Arasan and Dorky made use of this advantage.
ExChess produces strange evaluation scores. Completely won positions are evaluated with a score of -270, and after 3 such scores, EXChess resigns the game. Even a correct matescore in the wb2uci.eng file (MateScore =10000000) doesn't solve the problem. The results are corrected manually though.
Thanks for testing EXchess and for being careful to correct EXchess' scores in these positions. This is a strange result because EXchess should never resign on its own.... I have not coded this as a possibility, so if a resignation is happening, the GUI must be doing it. Also, such strange scores should not appear, can you send me the PGN of one of those games so I can run a few positions to test?