Author: Uri Blass
Date: 08:29:02 12/15/05
Go up one level in this thread
On December 15, 2005 at 10:56:28, Tord Romstad wrote: >On December 15, 2005 at 10:13:51, Uri Blass wrote: > >>I read that Dieter claimed that the ecmgcp test suite contains too many >>sacrifices and he prefers the arasan test suites so I wonder if you tried to >>compare results in this test. > >I just tried the Arasan 5 test suite, and got a result of 67/141 at 10 seconds >per position. This doesn't look good, but I have no idea how difficult these >positions really are. What scores are normal for other programs? I do not know what scores are normal for other programs because I test only movei in ecmgcp and arasan but when I compare with movei your result in arasan is relatively worse than your result in ECMGCP and movei needs more than 20 seconds to get your ECMGCP results in A3000 when it needs only slighly more than 10 seconds to get you arasan result. I do not know exact times because I considered solution as correct only if movei did not change it's mind later so the number of solutions in x seconds may be slightly higher if I give movei more time. For the record Movei in ECMGCP solved in less than 2000 solved=147 solved in less than 2050 solved=147 solved in less than 2100 solved=147 solved in less than 2150 45 solved=148 solved in less than 2200 26 solved=149 solved in less than 2250 solved=149 solved in less than 2300 31 solved=150 solved in less than 2350 solved=150 solved in less than 2400 149 solved=151 solved in less than 2450 solved=151 solved in less than 2500 solved=151 solved in less than 2550 142 solved=152 Movei in arasan(you need to add 1 to the number of solutions in arasan) because my program to calculate number of solution has a bug and do not consider bm with 2 solutions and considered 51 as wrong only because movei changed it's mind from Nexf6 to Ngxf6) 1n3rn1/3b4/rq1P1pkp/1p4p1/p1p1N1N1/2Q2B1P/1P3PP1/R3R1K1 w - - bm Nexf6 Ngxf6; id "arasan4.51"; solved in less than 900 124 solved=62 solved in less than 950 123 solved=63 solved in less than 1000 104 solved=64 solved in less than 1050 solved=64 solved in less than 1100 110 solved=65 solved in less than 1150 0 solved=66 solved in less than 1200 solved=66 solved in less than 1250 solved=66 solved in less than 1300 solved=66 solved in less than 1350 16 133 solved=68 solved in less than 1400 34 113 solved=70 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.