Author: Dann Corbit
Date: 16:22:59 01/06/00
Go up one level in this thread
On January 06, 2000 at 18:59:32, Andrew Walker wrote: >On January 06, 2000 at 14:25:10, Dann Corbit wrote: > >>The data can be retrieved in a side-by-side format from: >>ftp://38.168.214.175/pub/both.epd >> >>This is a comparison of data produced by PC's at very long time controls with >>data produced by a supercomputer. >> >>This is a very significant and important step for C.A.P. (and I think it will >>also be for chess theory). > >How were these positions selected? There were a number of alternatives explored. This particular set has the following characteristics: 0. All positions are from commonly known and named openings. 1. All positions in this set had an initial evaluation somewhere between being down one full pawn and being down two full pawns. This is a strange evaluation for an opening position since it *should* be about equal. Indeed, as others have noticed, the reanalyzed results are almost always closer to zero. >With the "Masked Marvel" computer it may be worthwhile to examine >the output to see the cutoff values produced for alternate moves. >These may indicate a bust to a known opening! I am sure that some of them will do exactly that. It will take some really smart people to know for sure though. >Also, would it be possible to produce a list of the most commonly occurring >positions in the King's Gambit? Some of these may be interesting to try... We have analyzed every opening position from a large set, and many KG positions are included. We are nearly due for a release of several hundred thousand positions for public consumption. The supercomputer will put a real shine on the apple, where it appears that we have a worm-hole right now.
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.