Author: John Merlino
Date: 16:16:54 06/06/02
Go up one level in this thread
On June 06, 2002 at 18:22:46, Kevin Strickland wrote: >I have been running engine matches, as well as an account on ICC with different >engines and have noticed somethings. > >1. Crafty greatly improves in the endgame with the tablebases in use. It plays >fine without them but does incredibly better with them. This goes for almost all >freeware/shareware programs. > >2. The commercial programs I have tested are Fritz, Junior, Shredder, Chess >Tiger, Gambit Tiger. With these programs they do extremely well without the >tablebases. In fact I have seen that the two Tigers play with almost no >difference at all. The commercial programs just seem to "know" the endings I >have seen. The Tigers are definately proof of this. > >3. The only commercial program that seems to "need" the tablebases is Nimzo. For >whatever reason it just plays endgames horribly without them. > >4. Of the commercial programs that _don't_ use the tablebases it is a saw off of >which is better. Goliath Blitz and Rebel seem to play endgames very well without >them. The winboard Goliath Blitz does not do well, but in the Chess Academy >interface it does very well. I have it outscoring Fritz 7 23-17 thus far in 40/2 >matches. I would be interested to see the results of your endgame tests with The King engine from Chessmaster. It does not use tablebases, but I believe it will play the endgames quite well. >5. The program that benefits the most out of all the engines that I tested is >without question Yace. The difference in play with/out them is great. I am >surprised by this. > >In short why is it that the commercial programs with the exception of Nimzo do >not seem to greatly improve with tablebases? Fritz 5.32 is a monster is does not >use them yet is just as good as most commercial counter-parts. Yet the freeware >programs increase greatly in strength with using them. > >Does a commercial program contain more knowledge in the endgame due to lack of >way to improve the opening/middlegame so the authors concentrate on the endgame >due to the amount of knowledge there is to put in there? > >This is just my findings and I am sure that others will disagree, but I think my >two cents count as well. > >Kevin.
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.