Author: Peter McKenzie
Date: 11:42:13 12/03/01
Go up one level in this thread
On December 03, 2001 at 13:33:06, Dieter Buerssner wrote: >On December 03, 2001 at 03:49:49, Peter McKenzie wrote: > >>Version one of my endgame test suite (50 positions) is finally ready for public >>consumption. It can found at: >> >>http://homepages.caverock.net.nz/~peter/eg_test/eg_test.htm > >Thanks Peter, for the testsuite. I found two small formal errors. positions >20 and 23 are missing a semicolon after the move. OK, thanks for picking that up. I'll fix it tonight & upload a new version. > >One thing, that I would find nice: If you could add a c0, saying something >like "only drawing move", "only winning move" and for avoid move positions >something like "loses", "only draws" or similar. A c0? Sorry for my ignorance, is that part of the epd standard? Perhaps you could give me an example? Thanks. > >Is the test suite designed, so that when there is a winning move, that there >should be no alternate winning moves, even when they are not that convincing? Yes that is the general idea. There should be no alternate winning moves. Or drawing moves if the solution is to draw. > >Yace results, on AMD K6-2 475, 20M hash, all 4-men, few 5-men TBs: > > 1s 5s 20s 60s st >PET001: Pawn en: 1 (e4) 1 (e4) 1 (e4) 1 (e4) 0.053 >PET002: Pawn en: 0 (Kg3) 0 (Kg3) 0 (Kg3) 1 (e4) 29.616 >PET003: Pawn en: 0 (Kxb3) 0 (Kxb3) 0 (Kxb3) 0 (Kxb3) nf >PET004: Pawn en: 1 (Kb1) 0 (Kb2) 1 (Kb1) 1 (Kb1) 5.161 >PET005: Pawn en: 0 (Ke1) 0 (Ke1) 1 (b3) 1 (b3) 49.639 >PET006: Pawn en: 0 (Kd5) 0 (g5) 0 (a5) 1 (b3) 32.001 >PET007: Pawn en: 0 (Kc5) 0 (Kc5) 0 (Ke4) 0 (Ke4) nf >PET008: Pawn en: 0 (a4) 0 (a4) 0 (a4) 1 (g4) 33.410 >PET009: Pawn en: 1 (g4) 1 (g4) 0 (a3) 0 (g3) nf >PET010: Pawn en: 0 (axb5) 0 (axb5) 0 (axb5) 0 (axb5) nf >PET011: pawns v: 0 (g6) 0 (g6) 0 (g6) 0 (g6) nf >PET012: pawns v: 0 (Be4) 0 (h3) 0 (h3) 0 (h3) nf >PET013: pawns v: 0 (Rxc5+) 0 (Rxc5+) 0 (Rxc5+) 1 (a4+) 23.756 >PET014: Knight : 0 (Ne4) 0 (Ne4) 0 (h3) 0 (h3) nf >PET015: Knight : 0 (Nd8) 0 (Nd8) 0 (Nd8) 0 (b4) nf >PET016: bishop : 0 (Be3) 0 (Be3) 0 (Be3) 0 (Be3) nf >PET017: bishop : 0 (gxh4) 0 (gxh4) 0 (gxh4) 0 (gxh4) nf >PET018: opposit: 1 (b4) 1 (b4) 0 (a4) 0 (a4) nf >PET019: opposit: 0 (a3) 0 (a3) 0 (a3) 0 (a3) nf >PET020: opposit: 0 (Bxe3) 0 (Bxe3) 0 (Bxe3) 0 (Bxe3) nf >PET021: bishop : 1 (Bh3) 1 (Bh3) 1 (Bh3) 1 (Bh3) 0.007 >PET022: bishop : 0 (a5) 0 (a5) 0 (a5) 1 (Bd8) 39.225 >PET023: bishop : 0 (Be3) 0 (Be3) 0 (Bc1+) 0 (Bc1+) nf >PET024: bishop : 0 (g5) 0 (g5) 0 (g5) 1 (e6) 38.712 >PET025: BN vs B: 1 (Bf6) 1 (Bf6) 1 (Bf6) 1 (Bf6) 0.293 >PET026: rook vs: 1 (g1=B) 1 (g1=B) 1 (g1=B) 1 (g1=B) 0.053 >PET027: rook vs: 0 (Rxd2) 1 (Rf5+) 1 (Rf5+) 1 (Rf5+) 1.720 >PET028: rook en: 0 (g3) 0 (a5) 1 (g4) 1 (g4) 8.893 >PET029: rook en: 0 (Rb3) 0 (Rb3) 0 (Rb3) 0 (Rb3) nf >PET030: rook en: 0 (Rd7) 0 (Rd7) 0 (Rd7) 0 (Rd7) nf >PET031: rook en: 1 (f4) 1 (f4) 1 (f4) 1 (f4) 0.065 >PET032: rook en: 0 (Rd5) 0 (Rd5) 0 (Rd5) 0 (Rd5) nf >PET033: rook en: 0 (b5) 0 (b5) 0 (b5) 0 (b5) nf >PET034: rook en: 0 (Rb8) 0 (Rb8) 0 (Rb8) 0 (Rb8) nf >PET035: rook en: 0 (axb5) 0 (Kc7) 0 (axb5) 0 (axb5) nf >PET036: rook en: 1 (Re7) 1 (Re7) 0 (Rb4) 0 (Rb4) nf >PET037: RN vs R: 0 (Rd6) 0 (c3) 0 (Bd8) 0 (c3) nf >PET038: RB vs R: 0 (Ra3) 0 (Ra3) 0 (Ra3) 1 (c4+) 55.860 >PET039: RB vs R: 0 (Rf8+) 0 (g5) 0 (g5) 0 (g5) nf >PET040: RB vs R: 0 (exf6+) 0 (exf6+) 0 (exf6+) 0 (exf6+) nf >PET041: RR vs R: 0 (Rb4) 0 (Rb4) 0 (Rb4) 0 (Rb4) nf >PET042: double : 0 (Rg2) 0 (Rg2) 0 (Rg2) 1 (f7) 30.265 >PET043: double : 0 (Rb8) 0 (Rc3) 1 (Rxa3) 1 (Rxa3) 15.470 >PET044: RRN vs : 0 (Kf1) 0 (Kf1) 0 (Kf1) 0 (Nb3) nf >PET045: Queen e: 0 (Qd3) 0 (Qd3) 0 (Qd3) 0 (Qd3) nf >PET046: Q+B vs : 0 (Qxf3) 1 (Qe5+) 1 (Qe5+) 1 (Qe5+) 3.638 >PET047: Q+B vs : 1 (Bf4) 1 (Bf4) 1 (Bg5) 1 (Bg5) 0.003 >PET048: Q+B vs : 0 (Qxh7+) 0 (Qxh7+) 1 (cxd5) 1 (cxd5) 11.470 >PET049: QNN vs : 1 (Qg3) 1 (Qg3) 1 (Qg3) 1 (Qg3) 0.421 >PET050: QNBR vs: 0 (Bxg6) 0 (Bxg6) 0 (Bxg6) 0 (Bxg6) nf > > Summary: 11 12 14 22 379.731 > >I did not calculate the points due to your suggestion yet. > >>The web pages need a bit of work, but hopefully they will do for now > >What would be nice, if all the analysis could be viewed on one page, instead of >a page for each position. Perhaps even a txt-only version without diagrams, just >the FENs and the comments. OK, I'll look at doing something like that since Peter Kappler also suggested something similar. I was planning to put next/previous links on the position pages too. > >Regards, >Dieter Peter
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.