Author: Uri Blass
Date: 05:56:39 11/01/05
Go up one level in this thread
On November 01, 2005 at 08:48:31, Uri Blass wrote: >On November 01, 2005 at 08:16:47, Ryan B. wrote: > >>On November 01, 2005 at 07:15:58, K. Burcham wrote: >> >>>On November 01, 2005 at 04:59:32, Ryan B. wrote: >>> >>>>On November 01, 2005 at 04:38:38, Jonas Cohonas wrote: >>>> >>>>I beat Fruit 2.2 more often than I beat Prodeo. In the human world of chess it >>>>takes a lot more than a few mobility factors and placing pieces in the center to >>>>have a positional style. Fruit is a very good engine but it is better at >>>>beating computers than humans in my experiance. >>>> >>>>Ryan >>> >>>Another human claiming he can beat one of our top programs, without posted the >>>games. Doesnt the charter say that if you claim to beat the top programs, you >>>must post some of the games? I think the charter says it is ok not to post games >>>where human beat Diep or Movei. >>> >>>kburcham >> >>Wow at least give me some time to respond before calling for moderation. I just >>said I have beaten Fruit, I did not claim to be better than it or that I am some >>sort of computer killer. Anywhay here is a quick king safty win. > > >I expected to find long time when fruit believes that it is better because of >not understanding king safety but I did not find it. > >Even at move 16 >Fruit can find Rxc3 with a clear advantage for black > >Fruit 2.2 - Ryan, Computer chess game 2005 >[D]2rq1rk1/1b1nbpp1/pn1pp2p/1p5P/4P1P1/PNN1BP2/1PP2Q2/1K1R1B1R b - - 0 1 > >Analysis by Fruit 2.2.1: > >16...Nc4 > ² (0.53) Depth: 1/6 00:00:00 >16...Nc4 17.Bxc4 Rxc4 > ² (0.43) Depth: 2/8 00:00:00 >16...Nc4 17.Bxc4 Rxc4 18.Qg3 > = (0.15) Depth: 3/11 00:00:00 >16...Nc4 17.Bxc4 Rxc4 18.Bf4 Qc7 > ² (0.34) Depth: 4/15 00:00:00 >16...Nc4 17.Bxc4 bxc4 18.Nd4 Ne5 19.Bf4 > = (0.13) Depth: 5/15 00:00:00 >16...Nc4 17.Bxc4 Rxc4 18.Bd4 Re8 19.Nd2 Rc7 > = (0.17) Depth: 6/16 00:00:00 29kN >16...Nc4 17.Bxc4 Rxc4 18.Qg3 Qc7 19.Rd4 Ne5 20.Rxc4 Qxc4 > = (0.09) Depth: 7/16 00:00:00 70kN >16...Nc4 17.Bxc4 Rxc4 18.Bd4 Re8 19.Nd2 Rc7 20.f4 Nc5 > = (0.10) Depth: 8/22 00:00:00 176kN >16...Nc4 17.Bxc4 Rxc4 18.Rhg1 Bh4 19.Qg2 Qc7 20.Qh2 Be7 21.g5 hxg5 22.Bxg5 > = (0.03) Depth: 9/28 00:00:01 554kN >16...Nc4 17.Bxc4 Rxc4 18.Rhg1 Re8 19.Rd4 d5 20.exd5 Bh4 21.Qh2 exd5 > = (0.00) Depth: 10/28 00:00:02 1103kN >16...Nc4 17.Bxc4 Rxc4 18.Rhg1 Re8 19.Qg3 d5 20.exd5 exd5 21.Nxd5 Bxa3 22.Rge1 >Ne5 > = (0.03) Depth: 11/35 00:00:04 2441kN >16...Nc4 17.Bxc4 Rxc4 18.Rhg1 Re8 19.Bf4 d5 20.Qa7 Qa8 21.Qxa8 Bxa8 22.Na5 Rc5 >23.exd5 Bxd5 24.Nxd5 Rxd5 25.Rxd5 exd5 > = (0.09) Depth: 12/35 00:00:07 4319kN >16...d5 17.exd5 Nxd5 18.Nxd5 Bxd5 19.Bd3 Qc7 20.Rhg1 Bxb3 21.cxb3 Bc5 22.Be4 >Bxe3 23.Qxe3 Nc5 24.Ka2 Nxe4 25.Qxe4 > = (-0.24) Depth: 12/35 00:00:44 26407kN >16...d5 17.Bd2 Qc7 18.Bd3 Ne5 19.Na5 Nxd3 20.cxd3 b4 21.axb4 Bxb4 22.Nxb7 Qxb7 >23.Rc1 e5 24.Be3 > ³ (-0.34) Depth: 13/44 00:01:56 71132kN >16...d5 17.Bd2 Qc7 18.Na5 Ba8 19.Bd3 Nc5 20.Be3 Nxd3 21.Rxd3 Nd7 22.Nb3 Ne5 >23.Rdd1 dxe4 24.Nxe4 Bxe4 25.fxe4 Nxg4 > ³ (-0.44) Depth: 14/44 00:03:30 128347kN >16...d5 17.Bd2 Qc7 18.g5 hxg5 19.h6 gxh6 20.Rxh6 Kg7 21.Rh3 Rh8 22.Rg3 Rh5 >23.exd5 Bxd5 24.Nd4 Nc4 25.Nxd5 Nxd2+ 26.Qxd2 exd5 > ³ (-0.44) Depth: 15/44 00:05:28 198331kN >16...d5 17.Bd2 Qc7 18.g5 hxg5 19.h6 gxh6 20.Rxh6 Kg7 21.Rh3 Rh8 22.Rg3 Rh5 23.f4 >dxe4 24.Be2 Rhh8 25.fxg5 > ³ (-0.31) Depth: 16/48 00:08:31 307331kN >16...Rxc3 17.bxc3 Na4 18.Bd2 Qc7 19.Bd3 Ne5 20.Ka2 Nxc3+ 21.Bxc3 Qxc3 22.Qd4 Qc7 >23.Qe3 d5 24.Qf4 Bd6 25.Qe3 Qc3 > µ (-0.71) Depth: 16/50 00:12:32 462913kN > >(, 01.11.2005) > >Uri I can add that it is very good that you beat Fruit but it seems that you simply did it by a tactical shot. Rxc3 is a tactical shot that fruit simply missed and I think that a lot of programs need a long time for it. I tried prodeo that you consider as better and I do not see that it finds Rxc3 easily Maybe it can find it faster and I did not check it but it seems to be clearly a search problem. Uri 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.