Author: blass uri
Date: 02:30:50 03/28/00
Go up one level in this thread
On March 28, 2000 at 05:13:32, Enrique Irazoqui wrote: >On March 28, 2000 at 04:39:32, Bertil Eklund wrote: > >>On March 28, 2000 at 03:20:18, blass uri wrote: >> >>>On March 28, 2000 at 01:24:18, Bertil Eklund wrote: >>> >>>>On March 27, 2000 at 09:53:57, Ed Schröder wrote: >>>> >>>>>On March 27, 2000 at 09:06:01, Robert Hyatt wrote: >>>>> >>>>>>On March 25, 2000 at 23:13:49, Tina Long wrote: >>>>>> >>>>>>>On March 25, 2000 at 14:28:13, James Robertson wrote: >>>>>>> >>>>>>>>On March 25, 2000 at 13:41:28, Roger wrote: >>>>>>>> >>>>>>>>>Would tablebases for Tiger have changed this result at all? >>>>>>>>> >>>>>>>>>Roger >>>>>>>> >>>>>>>>Maybe a quarter of a point.... My experience with tablebases is that if the >>>>>>>>program is moderately smart it doesn't benefit tremendously from them. >>>>>>>> >>>>>>>>James >>>>>>>> >>>>>>>Ed Schroder said about 6 months ago that Tablebases were worth about 10 points >>>>>>>on the SSDF scale. >>>>>>> >>>>>>>I'm 70% sure he said that! I'm 100% sure that Ed said once that something was >>>>>>>worth very little rating points. >>>>>>> >>>>>>>I'm glad I could add some real detail to this discussion. >>>>>>> >>>>>>>Tina Long >>>>>> >>>>>> >>>>>>Ed is wrong there. it is _amazing_ how many comp vs comp games end up in >>>>>>krp vs kr, with the side without tablebases losing most of those. There are >>>>>>other endings too (KQP vs KQ, see for example crafty vs nimzo in the ICCT >>>>>>tournament last month). >>>>>> >>>>>>The wrong way to test this is to play A with, vs A without. the right way to >>>>>>test this is A without vs B without, then A with vs B without. But A ought to >>>>>>be reasonably close to B without tablebases... >>>>> >>>>>Tablebases have a great future no doubt. But what is available at the >>>>>moment (4-5 pieces) its value for Rebel is not more than 5-10 elo I >>>>>would say because: >>>>> >>>>>a) most cases are simply covered by chess knowlegde; >>>>> >>>>>b) the loss of speed during search because of all the >>>>>disc access. >>>>> >>>>>So I don't think I am wrong when the subject is Rebel. >>>>> >>>>>Things might change dramatically when for instance the complete >>>>>6 pieces become available. +100 elo easily for chess programs. >>>>> >>>>>Ed >>>> >>>>Hi! >>>> >>>>I agreed totally with you some months ago but todays best programs uses the TBs >>>>in the search very efficient. In example Hiarcs and Nimzo began to find the Tbs >>>>with 10-11 pieces on the board (tournament time) but today Crafty, Fritz6, >>>>Junior6 and Shredder4 find the TBs with 15-16 pieces on the board. These four >>>>programs are probably the best in endgames, only Tiger without TBs come close. >>>> >>>>I guess the above programs earns 25-50 elo with TBs. >>>> >>>>Bertil >>> >>>1)The fact that they find the tablebases does not say in how many cases the >>>tablebases change the result and we cannot know from this about the elo >>>improvement. >>>The only way to know is by testing the program with tablebases and the program >>>without tablebases. >>> >>>2)It is known that crafty also could find the tablebases with 15-16 pieces some >>>monthes ago. >>> >>>Uri >>Hi! >> >>I have followed a lot of games, when the above mentioned programs wins the >>endgame. I don't think it's only a coincidence that these programs and maybe >>Tiger seems to play better endgames than i.e. Nimzo, Hiarcs, Rebel Genius Mchess >>and so on. >> >>2. Yes and Crafty plays very good endgames. >>3. Endgames becomes more and more important for todays programs, and I guess >>Tablebases is the most efficient way to go. > >I agree with you. I didn't count the games in which tablebases were decisive, >but I think they add at least half a point every 20 games or so. It is hard to count the number of games because we often do not know what was the result without tablebases. I think that tablebases are sometimes counter productive when you play against someone who does not use tablebases because the program may prefer a simple draw instead of going to a drawn KRP vs KR that it can practically win. The program may prefer a simple loss of KQ vs KQPP instead of going to a lost KQ vs KQP that it can practically draw. The only way to know the value of tablebases is by testing and you cannot know only by watching games. 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.