Author: Dann Corbit
Date: 17:11:52 07/20/99
Go up one level in this thread
On July 20, 1999 at 20:05:01, Dann Corbit wrote: >Look Here: >ftp://38.168.214.175/pub/Public_CAP_Results/Apocalypse/Apocalypse.epd > >There are 7765 EPD positions from common test suites analyzed at between 8 hours >and 12 minutes of PII 300 MHZ CPU time (depending on position and whim). Here is an interesting summary: White(1): epdscor apocalypse.epd total: 7624 total acd: 104572 mean total acd: 13.72 solve: 5701 solve percent: 74.78 solve acd: 77195 mean solve acd: 13.54 unsol: 1923 unsol percent: 25.22 unsol acd: 27377 mean unsol acd: 14.24 Most (not all) of the unsolved EPD scores are actually bugs in the EPD suites themselves. At least 3/4 of them, maybe 9/10. The others are Zugzwang problems or deep positional concepts that are beyond most chess program's ability to solve.
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.