Author: Eelco de Groot
Date: 03:24:24 02/10/02
Go up one level in this thread
On February 10, 2002 at 05:34:23, Eelco de Groot wrote: >On February 09, 2002 at 23:09:07, Albert Silver wrote: > >>On February 09, 2002 at 19:55:58, Andrew Williams wrote: >> >>>Hi Ed, >>> >>>PostModernist says: >>> >>>EPD TEST SET: rebelhard.epd - 360 seconds per position >>> >>>ID rebelhard.epd.0001 - CORRECT in 6.05 seconds. nodes: 354557 >> >>Could you post the analysis and a few lines beyond? >> >> Albert >> >>>ID rebelhard.epd.0002 - WRONG. nodes: 35362560 (Nxh6) >>>ID rebelhard.epd.0003 - WRONG. nodes: 53597184 (Kd2) >>>ID rebelhard.epd.0004 - WRONG. nodes: 76704256 (Rf3) >>>ID rebelhard.epd.0005 - CORRECT in 273.02 seconds. nodes: 45357242 >>> >>>RESULT: 2 Correct (5 positions @ 360 seconds). >>> >>>NODES per second 163993 >>> >>> >>>This is on an Athlon 1200. >>> >>>Andrew > > >Me too! I would like to know what Troitzky's intended line was in his study. I >couldn't find the study on the Net, so far. > >Trying Q5T it won't play Rf7: > >[d]k2N2K1/8/8/8/5R2/3n4/3p4/8 w - - bm Rf7 (Troitzky); > >Game begin > >00:00 05.01 -3.22 1.Rf1 >00:00 07.01 -3.71 1.Rf1 >00:01 08.00 -4.01 1.Ra4+ Kb8 2.Ra1 Nc1 3.Rb1+ Kc8 4.Nf7 d1=Q >00:01 08.01 -3.79 1.Rf1 Ne1 2.Nf7 d1=Q 3.Rg1 dxe2 4.Rg7 >00:02 09.00 -3.97 1.Rf1 Ne1 2.Ne6 d1=Q 3.Rf8+ Kb7 4.Rf7+ >00:05 10.00 -4.06 1.Rf1 Ne1 2.Ne6 d1=Q 3.Rf8+ Kb7 4.Rf7+ >00:16 11.00 -4.40 1.Rf1 Ne1 2.Nc6 d1=Q 3.Rf8+ Kb7 4.Nd8 >01:35 12.00 -4.39 1.Rf1 Ne1 2.Nc6 d1=Q 3.Rf8+ Kb7 4.Nd8+ >04:05 13.00 -4.59 1.Rf1 Ne1 2.Nc6 d1=Q 3.Rf8+ Kb7 4.Nd8+ >22:10 14.00 -4.59 1.Rf1 Ne1 2.Nf7 d1=Q 3.Rg1 Ka7 4.Rg7 >13:11 15.00 -4.76 1.Rf1 Ne1 2.Nf7 d1=Q 3.Rg1 dxb3 4.Rg7 > >And after 1.Rf7: > >Rebel Engine for ECTool. (c) Ed Schröder > >Engine version : REBEL CENTURY 2 >Rebel personality : "Personality Q5T" >Opening book file : RebelC.mvs >Hash table size : 4 MB >Analysis mode : Analyzing next move >Refresh interval : 500 ms > >Last Move : 1.Rf7 (Black to play) > >00:00 05.00 4.54 1..d1=Q 2.Re7 Nc5 3.Ne6 dxg4+ 4.Ng7 >00:01 08.00 5.09 1..d1=Q 2.Re7 Nc5 3.Ne6 dxd5 4.Re8+ Kb7 >00:02 09.00 5.47 1..d1=Q 2.Re7 dxg4+ 3.Rg7 gxc8 4.Kh7 Nf4 >00:04 10.00 5.90 1..d1=Q 2.Ne6 dxg4+ 3.Ng7 Ne5 4.Rf5 g3 >00:06 11.00 5.91 1..d1=Q 2.Ne6 Ne5 3.Re7 dxg4+ 4.Ng7 gxg5 >01:16 12.00 6.55 1..d1=Q 2.Rf8 dxg4+ 3.Kh7 gxh4+ 4.Kg8 hxg5+ >03:43 13.00 6.15 1..d1=Q 2.Rf8 dxg1=Q+ 3.Kh7 gxh2+ 4.Kg7 hxe5+ >06:14 14.00 6.13 1..d1=Q 2.Rf8 dxg4+ 3.Kh7 gxh4+ 4.Kg7 hxg5+ >25:49 15.00 6.59 1..d1=Q 2.Rf8 dxg4+ 3.Kh7 gxh4+ 4.Kg8 hxg5+ >07:27 16.00 6.59 1..d1=Q 2.Rf8 dxg4+ 3.Kh7 gxh4+ 4.Kg8 hxg5+ > >Iteration 17 looks to be a fail-low but that could take quite a while. Celeron >500 MHz, clocking 8 hours so far on a reply to 1.Rf7. After 1.Rf7 d1=Q 2.Nc6 I get the following. No more phantom pieces taken by the queen, maybe that was just a glitch from retrieving lines from hash, but still 7 pawns difference from seeing it as a draw: Last Move : 2.Nc6 (Black to play) 00:04 07.00 9.30 2.Ne5 Kg7 3.Qd4 Ne6 4.Qg4 Kf6 5.Nc6 Ng5 6.Kb8 00:00 03.00 6.59 2..Qg4+ 3.Rg7 Qc8+ 4.Kh7 Qxc6 5.Kh8 Qh6 00:00 05.00 6.96 2..Qg4+ 3.Kh7 Qe4+ 4.Kg7+ Qxc6 5.Rf8+ Kb7 6.Rf7 Kc8+ 7.Kg8 Qe6 00:01 06.00 6.91 2..Qg4+ 3.Kf8 Qc8+ 4.Kg7 Qxc6 5.Rf8+ Kb7 6.Rf7 Kc8 7.Rf8+ 00:02 07.00 6.97 2..Qg4+ 3.Kf8 Qc8+ 4.Kg7 Qxc6+ 5.Rf8 00:04 08.00 6.92 2..Qg4+ 3.Kf8 Qc8+ 4.Kg7 Qxc6+ 00:14 09.00 7.05 2..Qg4+ 3.Kf8 Qc8+ 00:48 10.00 7.05 2..Qg4+ 3.Kf8 Qc8+ 4.Kg7 Qxc6+ 5.Rf8+ Kb7 6.Rf7+ Kc8 7.Rf8 Kd7 8.Rf7 04:07 11.00 6.88 2..Qg4+ 3.Kf8 Qc8+ 4.Kg7 Qxc6+ 5.Rf8+ Kb7 6.Rg8 Qe4 7.Kf7 Kc6 8.Re8 09:55 12.00 7.02 2..Qg4+ 3.Kf8 Is this ending Queen vs. Rook really a draw? Then I think RC4 will see that, even without Nalimov tablebases and the fact that RC2(Q5T) couldn't see it isn't really relevant anymore of course. So I'm interested what RC4 would make of it! Eelco
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.