Author: Uri Blass
Date: 10:17:14 09/01/03
Go up one level in this thread
On September 01, 2003 at 11:19:36, Steven J. Brann wrote: >On August 31, 2003 at 23:29:44, Slater Wold wrote: > >>On August 31, 2003 at 23:27:53, Slater Wold wrote: >> >>>Can you beat Ruffian's time? (P4 3.06Ghz - 100MB HTs) I am curious to see CM9k >>>results. >>> >>>[D]8/1P1bp1kp/3P1p2/R7/4Pb2/5P2/4KR2/6q1 b - - 0 44 >>> >>>00:00:00.0 -0.23 1 3 exd6 b8Q >>>00:00:00.0 6.83 1 6 Bxd6 >>>00:00:00.0 6.56 2 369 Bxd6 Ra7 >>>00:00:00.0 6.82 3 3601 Bxd6 Rf1 Qd4 >>>00:00:00.0 5.48 4 7116 Bxd6 Rd5 Bc6 Rxd6 Bxb7 >>>00:00:00.0 5.50 4 14145 Qb1 Rg2+ Kf7 Rh5 Qxb7 Rxh7+ >>>00:00:00.0 5.89 5 30785 Qb1 Rg2+ Kf7 b8Q Qxb8 dxe7 Kxe7 Rg7+ Kd6 Rxh7 >>>00:00:00.0 5.91 5 34945 Bxd6 Rd5 Bg3 Rxd7 Qxf2+ Kd3 Qxf3+ >>>00:00:00.1 6.55 6 67677 Bxd6 Rd5 Qb1 Rg2+ Kf8 Rh5 Bf4 b8Q+ Qxb8 Rxh7 >>>00:00:00.1 7.01 6 116475 Qb1 Rg2+ Kf7 b8Q Qxb8 dxe7 Bb5+ Kd1 Kxe7 >>>00:00:00.2 7.57 7 308845 Qb1 Rg2+ Kf7 Rh5 Bb5+ Rxb5 Qxb5+ Kf2 Qxb7 dxe7 Qa7+ Ke2 >>>00:00:00.9 8.08 8 1229946 Qb1 Rd5 Bb5+ Rxb5 Qxb5+ Kd1 exd6 Rg2+ Bg5 Kc2 Qxb7 Rg1 >>>00:00:01.4 8.21 9 1982474 Qb1 Rg2+ Kf7 Rh5 Bb5+ Rxb5 Qxb5+ Kf2 Bxd6 f4 Bc5+ Kf3 >>>Qxb7 >>>00:00:02.3 8.32 10 3295262 Qb1 Rg2+ Kf7 Rh5 Bb5+ Rxb5 Qxb5+ Kf2 Bxd6 f4 h5 Kg3 >>>Qxb7 >>>00:00:02.6 8.56 10 3684627 Bg3 >>>00:00:03.1 9.68 10 4400400 Bg3 b8Q Qxf2+ Kd1 Qe1+ Kc2 Qxa5 Qb2 Be5 Qb7 Qc3+ Kd1 >>>Qxf3+ Ke1 >>>00:00:07.5 13.74 10 11566347 Bh3 Kd3 Qxf2 b8Q Qd2+ Kc4 Qxa5 Qb7 Be6+ Qd5 Qa4+ >>>Kd3 Qa3+ Kc4 Bxd5+ exd5 Qxd6 >>>00:00:11.6 13.75 11 18315246 Bh3 Kd3 Qxf2 b8Q Qd2+ Kc4 Qxa5 Qb7 Be6+ Qd5 Qa4+ >>>Kd3 Qd1+ Kc4 Qe2+ Kd4 Be3+ Kc3 Bxd5 exd5 exd6 >>>00:00:13.7 14.10 12 21555471 Bh3 >>>00:00:44.5 M12 12 74559727 Bh3 Kd3 Qxf2 b8Q Bf1+ Kc3 Qe3+ Kb2 Bc4 Qf8+ Kxf8 >>>dxe7+ Kxe7 Ra7+ Kd6 Ra6+ Kc5 Ra5+ Kb6 Rb5+ Kxb5 e5 Qc1# >> >>00:02:07.2 Mate in 11 14 216200928 Bh3 Kd3 Qxf2 b8Q Bf1+ Kc3 Qe3+ Kb4 Qd4+ Kb3 >>Qc4+ Ka3 Bc1+ Qb2 Qc3+ Ka4 Qxb2 Rg5+ fxg5 Ka5 Qb5# >> >>I know it's no shorter than 11...I had Chest do through M10. > > >Slater, I let CM9K run for over an hour on my 3.0G P4 HT machine with a custom >personality (SS=16) and no Mate in 11 was found. Then forced the line you have >for a Mate in 11 (see below) with a strange result. Normally CM9K will find a >mate and then improve on it to fewer moves if a solution is possible. It just >not do it in this case for some strange reason. > > >Time Depth Score Positions Moves >0:00 1/3 -7.19 3085 44...Bxd6 45.Rf1 Qb6 >0:00 1/4 -8.30 11867 44...Bxd6 45.e5 fxe5 46.Rf1 Qd4 >0:00 1/5 -8.19 36046 44...Bxd6 45.e5 fxe5 46.Rf1 Qb6 >0:00 1/5 -9.15 69154 44...Qb1 45.Rh5 Bb5+ 46.Rxb5 Qxb5+ > 47.Ke1 Qxb7 48.dxe7 Qxe7 49.Ke2 >0:00 1/6 -9.65 181791 44...Qb1 45.Rg2+ Kf7 46.e5 Bb5+ > 47.Rxb5 Qxb5+ 48.Kf2 Qb2+ 49.Kg1 > Be3+ 50.Kh2 Qxe5+ 51.Kh3 Qxd6 >0:01 1/6 -12.23 265186 44...Bh3 45.Kd3 Qxf2 46.b8=Q Qe3+ > 47.Kc4 Qc1+ 48.Kb3 Qb1+ 49.Kc3 > Qxb8 50.dxe7 >0:01 1/7 -12.81 378623 44...Bh3 45.Rd5 Bf1+ 46.Rxf1 Qe3+ > 47.Kd1 Qc1+ 48.Ke2 Qc2+ 49.Ke1 > Bg3+ 50.Rf2 Bxf2+ 51.Kf1 Bg3 52.Rd2 > Qxd2 >0:03 1/8 -14.92 922369 44...Bh3 45.Kd3 Qxf2 46.b8=Q Bf1+ > 47.Kc3 Qe3+ 48.Kb4 Qd4+ 49.Kb3 > Bc4+ 50.Ka4 Qd1+ 51.Kb4 Qb3+ 52.Kc5 > Be3+ 53.Kc6 Qxb8 >0:10 1/9 -18.39 3412848 44...Bh3 45.Kd3 Qxf2 46.b8=Q Bf1+ > 47.Kc3 Qe3+ 48.Kb4 Qd4+ 49.Kb3 > Bc1 50.Qf8+ Kxf8 51.dxe7+ Kxe7 > 52.Ra7+ Ke6 53.Re7+ Kxe7 >0:22 1/10 -Mate12 8237828 44...Bh3 45.Kd3 Qxf2 46.b8=Q Bf1+ > 47.Kc3 Qe3+ 48.Kb4 Qd4+ 49.Kb3 > Bc1 50.Qf8+ Kxf8 51.dxe7+ Kxe7 > 52.Ra7+ Ke6 53.Re7+ Kxe7 54.Kc2 > Qb2+ 55.Kd1 Qd2# >0:35 1/11 -Mate12 13571310 44...Bh3 45.Kd3 Qxf2 46.b8=Q Bf1+ > 47.Kc3 Qe3+ 48.Kb4 Qd4+ 49.Kb3 > Bc1 50.Qf8+ Kxf8 51.dxe7+ Kxe7 > 52.Ra7+ Ke6 53.Re7+ Kxe7 54.Kc2 > Qb2+ 55.Kd1 Qd2# >1:25 1/12 -Mate12 33922295 44...Bh3 45.Kd3 Qxf2 46.b8=Q Bf1+ > 47.Kc3 Qe3+ 48.Kb4 Qd4+ 49.Kb3 > Bc1 50.Qf8+ Kxf8 51.dxe7+ Kxe7 > 52.Ra7+ Ke6 53.Re7+ Kxe7 54.Kc2 > Qb2+ 55.Kd1 Qd2# >3:50 1/13 -Mate12 94598192 44...Bh3 45.Kd3 Qxf2 46.b8=Q Bf1+ > 47.Kc3 Qe3+ 48.Kb4 Qd4+ 49.Ka3 > Bc1+ 50.Ka2 Qd2+ 51.Ka1 Qxa5+ 52.Kb1 > Bd3+ 53.Kxc1 Qc3+ 54.Kd1 Qc2+ 55.Ke1 > Qe2# >9:12 1/14 -Mate12 233772540 44...Bh3 45.Kd3 Qxf2 46.b8=Q Bf1+ > 47.Kc3 Qe3+ 48.Kb4 Qd4+ 49.Ka3 > Bc1+ 50.Ka2 Qd2+ 51.Ka1 Qxa5+ 52.Kb1 > Bd3+ 53.Kxc1 Qc3+ 54.Kd1 Qc2+ 55.Ke1 > Qe2# >23:41 1/15 -Mate12 618831458 44...Bh3 45.Kd3 Qxf2 46.b8=Q Bf1+ > 47.Kc3 Qe3+ 48.Kb4 Qd4+ 49.Ka3 > Bc1+ 50.Ka2 Qd2+ 51.Ka1 Qxa5+ 52.Kb1 > Bd3+ 53.Kxc1 Qc3+ 54.Kd1 Qc2+ 55.Ke1 > Qe2# >59:30 2/16 -Mate12 1587576671 44...Bh3 45.Kd3 Qxf2 46.b8=Q Bf1+ > 47.Kc3 Qe3+ 48.Kb4 Qd4+ 49.Ka3 > Bc1+ 50.Ka2 Qd2+ 51.Ka1 Qxa5+ 52.Kb1 > Bd3+ 53.Kxc1 Qc3+ 54.Kd1 Qc2+ 55.Ke1 > Qe2# > >Then forcing moves to the point where your line for Mate in 11 differs... > >[D]1Q6/4p1kp/3P1p2/R7/1K1qPb2/5P2/8/5b2 w - - 0 5 > >CM9K now has a bizarre result. It sees the mate in 11 line with 49.Kb3 in less >than a second then returns to the longer mate (mate in 7 at this point) and >stays with it. So then, forcing 49.Kb3 enables CM9K to pick up on the mate in 6 >with no difficulty: Why do you force Kb3 when CM9k already saw mate in 6 against it based on your analysis. I did not check with chest but I doubt if there is mate in 6 after Ka3 Based on experience chess programs can show wrong mate or wrong pv and Ruffian does not see mate in 6 after Ka3 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.