Author: Robert Hyatt
Date: 20:15:03 03/27/00
Go up one level in this thread
On March 27, 2000 at 12:02:38, blass uri wrote: >On March 27, 2000 at 09:06:01, Robert Hyatt wrote: ><snipped> >>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. > > >It does not prove that it lost because of not having tablebases. Yes it does... because these programs are _losing_ the resulting krp vs kr games against crafty... and they are drawing won krp vs kr endings vs crafty, on the chess servers... Lonnie could recount dozens of such games he has played... > >I think that the side with tablebases sacrificed a queen to get p With computer vs computer, the single most common ending is krp vs kr. Because one side wins a pawn, but defends well. Many of these are _very_ tricky to defend. Sacrifices are not very common in them, but trading at the right instant is quick common. There is no good reason to not have them, since the 3+4+krpkr files are only 200 megs total... In the ICC ending with nimzo, it wasn't a sac at all, just kqp vs kq after the trades... a won kqp vs kq which is very difficult to win without tables, and very hard to draw with them. art of the KRP >vs KR wins from KRQP vs KR positions or did different sacrifices. > >The side with tablebases did not lose some KRP vs KR positions also because of >the fact that it prefered to lose other endgames like KRRP vs KR. > >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.