Author: Uri Blass
Date: 20:43:58 04/02/03
Go up one level in this thread
On April 02, 2003 at 22:52:11, Mike Byrne wrote: >with some of the new engines, RR game/5" - that is the standard CM8K sitting >tied for first > >new_test Secane, PA USA 2003 > > > 1 Shredder 7 8.0/11 > > 2 CM8 The King 8.0/11 > > 3 Fritz 8 8.0/11 > > 4 Ruffian 1.0.1 7.5/11 > > 5 List 504 7.0/12 > > 6 Deep Junior 7 7.0/11 > > 7 Anaconda 1.0 6.5/12 > > 8 Pharaon 2.62 5.5/11 > > 9 AnMon5.21 5.5/11 > > 10 Deep Fritz 7 5.5/11 > > 11 Smar Think 0.16.b 5.5/11 > > 12 SOS.2 for Arena 5.0/12 > > 13 Wildcat 2.79 5.0/11 > > 14 The King ST 5.0/11 > > 15 Yace Paderborn 5.0/11 > > 16 Crafty 19.03 Capablanc 4.5/11 > > 17 Comet B60 4.0/11 > > 18 Aristarch 4.4 3.5/11 > > 19 Deep Sjeng 1.0 3.5/11 > > 20 Movei 804a 2.5/12 I thought the dead line already passed for sending a new version. Note that I am not surprised from the last place of movei and every result of more than 20% is good for me(espacially when only the strong sjeng participated). I already have a new version that is probably slightly better based on my tests(only difference is that I give smaller bonus for passed pawns that are blocked by pieces(idea taken from Crafty) and I got 69/120 against previous version in my tests at different time control(10 second per game,20 second per game and 1 1 time control). I have still a lot to improve in my pawn evaluation and movei still does not evaluate passed pawns relative to the king but I need first to define my plans and later to test them. 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.