Author: Vincent Diepeveen
Date: 18:34:52 01/25/02
Go up one level in this thread
On January 24, 2002 at 18:28:04, Roy Eassa wrote: >On January 24, 2002 at 18:10:23, Pierre Bourget wrote: > >>On January 24, 2002 at 15:30:30, Roy Eassa wrote: >> >>>On January 24, 2002 at 15:28:24, Roy Eassa wrote: >>> >>>>On January 24, 2002 at 03:09:00, Kurt Utzinger wrote: >>>> >>>>>At first glance I thought: A too easy task for the programs of the new >>>>>generation. And in principle I was right. Somewhat disappointing however that >>>>>GambitTiger2 needs two minutes. Most programs solve the problem within seconds. >>>>>Kurt >>>> >>>> >>>>I don't think it's too easy ... some of the top engines, including Tiger and >>>>Fritz 7, need over 2 minutes to see that this line is winning. The fastest >>>>programs (for this puzzle) still take well over 10 seconds. To me, that makes >>>>this an excellent puzzle position. >>> >>> >>>Oops, just saw Chesster's time (2 seconds!). Well, kudos to chesster, but still >>>most other engines do have to think awhile on this position. >> >>Goliath Light 1.5 takes 10 secondes to find Bxc6+ on my Athlon XP 1600+. >> >>Pierre > > >I think a problem that takes some leading programs (e.g., Fritz, Tiger) several >minutes and some programs 10 seconds (on a very fast PC) and one or two programs >under 3 seconds is an IDEAL problem. It really differentiates one program from >another, much more so than most puzzles. it is not ideal unless you get a real huge score. just +1.x is not convincing. that could be just seeing 3 pawns versus piece. Only seeing more is convincing. Seems to me that the most speculative programs play this fastest. CST perhaps under 1 second? So quote the score please!
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.