Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Probably due to an endless and planless loop like this one :-)

Author: Terry McCracken

Date: 17:14:00 11/17/03

Go up one level in this thread


On November 17, 2003 at 20:06:22, Jorge Pichard wrote:

>On November 17, 2003 at 19:57:26, Jorge Pichard wrote:
>
>>On November 17, 2003 at 19:44:19, Terry McCracken wrote:
>>
>>>On November 17, 2003 at 19:00:22, Amir Ban wrote:
>>>
>>>>On November 17, 2003 at 16:18:20, Terry McCracken wrote:
>>>>
>>>>>None of my top commercial programmes find 14..Ne8! followed by 15..f5
>>>>>
>>>>>Junior 8 comes close, but chose Nh5 and no f5, although in analysis mode it does
>>>>>look at Ne8 and f5, strange...?
>>>>>
>>>>>Junior 8 also gets the second best evaluation compared to all other commercials,
>>>>>and basically feels White is winning at move 45 but not at 14., but it does have
>>>>>a good eval for White, about + 0.80
>>>>>
>>>>>Crafty lives up to its name and considers White +- at move 14 and 45 ++!
>>>>>
>>>>>I tested Shredder 7.04, Chess Tiger 15, HIARCS 9, Junior 8, and yes even Fritz
>>>>>8...at least Fritz 8 ver. 8.0.0.23 didn't look at 14...Bd6?!:)
>>>>>
>>>>>Junior scored the best among the commercials but it was Crafty 19.03 that was on
>>>>>the right track consistently.
>>>>>
>>>>>The test was 3 min. for each programme, on my very slow PIII 500, 128MB of RAM
>>>>>using 64MB of hash.
>>>>>
>>>>>My resources were down so the tests were not optimum but still good.
>>>>>
>>>>>Here's Craftys scores at moves 14.. and 45..
>>>>>
>>>>>Hiarcs 9 - McCracken,T
>>>>>r1bq1rk1/1p1nbppp/1Pp2n2/p1Pp4/Q2Pp3/BNN1P2P/P4PP1/R3KB1R b KQ - 0 1
>>>>>
>>>>>Analysis by Crafty 19.03:
>>>>>
>>>>>14...Re8 15.Nxa5
>>>>>  ²  (0.62)   Depth: 1/7   00:00:00
>>>>>14...Re8 15.Nxa5
>>>>>  ²  (0.62)   Depth: 2/7   00:00:00
>>>>>14...Re8 15.Be2 Rf8 16.Nxa5
>>>>>  ±  (0.81)   Depth: 3/7   00:00:00
>>>>>14...Re8 15.Be2 Rf8 16.Nxa5
>>>>>  ±  (0.81)   Depth: 4/8   00:00:00
>>>>>14...Re8 15.Be2 Nf8 16.0-0 Ng6 17.Nxa5
>>>>>  ±  (1.33)   Depth: 5/13   00:00:00
>>>>>14...Nxb6 15.cxb6 Bxa3 16.Qxa3 Qxb6
>>>>>  ±  (1.29)   Depth: 5/13   00:00:00
>>>>>14...Nb8 15.Be2 Bf5 16.0-0 Nbd7 17.Nxa5
>>>>>  ±  (1.20)   Depth: 5/13   00:00:00
>>>>>14...Rb8 15.Qxa5 Re8 16.Be2 Rf8
>>>>>  ±  (1.06)   Depth: 5/13   00:00:01
>>>>>14...Rb8 15.Qxa5 Kh8 16.g4 Kg8 17.Be2
>>>>>  ±  (1.17)   Depth: 6/19   00:00:01  40kN
>>>>>14...Rb8 15.Be2 Ra8 16.0-0 Nb8 17.Bb4 Bf5 18.Bxa5
>>>>>  +-  (1.52)   Depth: 7/19   00:00:01  87kN
>>>>>14...Bd6 15.Be2 Nxb6 16.cxb6 Bxa3 17.Qxa3 Qxb6
>>>>>  +-  (1.46)   Depth: 7/19   00:00:01  134kN
>>>>>14...Ne8 15.Be2 Bh4 16.0-0 Qg5 17.Bb4 Nef6 18.Bxa5
>>>>>  ±  (1.24)   Depth: 7/19   00:00:02  155kN
>>>>>14...Re8 15.Be2 Nf8 16.0-0 Ng6 17.Bb2 Bf5 18.Nxa5
>>>>>  ±  (1.22)   Depth: 7/19   00:00:02  180kN
>>>>>14...h6 15.Be2 Nxb6 16.cxb6 Bxa3 17.Qxa3 Qxb6
>>>>>  ±  (1.21)   Depth: 7/19   00:00:02  202kN
>>>>>14...h6 15.Be2 Nh7 16.0-0 Bh4 17.Nxa5 Qf6 18.Qb4
>>>>>  ±  (1.34)   Depth: 8/19   00:00:03  275kN
>>>>>14...Ne8 15.Be2 f5 16.0-0 f4 17.Nxa5 f3 18.gxf3 exf3
>>>>>  ±  (1.19)   Depth: 8/19   00:00:03  369kN
>>>>>14...Ne8 15.Be2 f5 16.0-0 f4 17.Nxa5 f3 18.gxf3 exf3
>>>>>  ±  (1.19)   Depth: 9/23   00:00:05  540kN
>>>>>14...Ne8 15.Be2 f5 16.g3 Ndf6 17.Nxa5 Qd7 18.0-0 f4 19.h4 fxe3 20.fxe3
>>>>>  ±  (1.23)   Depth: 10/24   00:00:08  935kN
>>>>>14...Ne8 15.Nxa5 Nef6 16.Bc1 Rb8 17.Be2 Nxc5 18.dxc5 Bxc5 19.0-0 Bxb6
>>>>>  ±  (1.18)   Depth: 11/26   00:00:26  3862kN
>>>>>14...Ne8 15.Be2 f5 16.g3 Nef6 17.0-0 f4 18.gxf4 Nb8 19.Kg2 Qd7 20.Rh1
>>>>>  ±  (1.24)   Depth: 12/28   00:01:05  10391kN
>>>>>14...Ne8 15.Be2 f5 16.g3 Nef6 17.0-0 Qe8 18.Nxa5 Qf7 19.Qb4 Qg6 20.Bb2 Qg5
>>>>>  ±  (1.26)   Depth: 13/30   00:01:49  17863kN
>>>>>
>>>>>(McCracken, None 17.11.2003)
>>>>>
>>>>>
>>>>>
>>>>>
>>>>>
>>>>>
>>>>>Hiarcs 9 - McCracken,T
>>>>>r6k/1r1qnppp/NPp2n1b/2Pp4/3Pp3/1RB1P1PP/R1Q2P2/K4B2 b - - 0 1
>>>>>
>>>>>Analysis by Crafty 19.03:
>>>>>
>>>>>45...Ne8
>>>>>  ±  (1.36)   Depth: 1/6   00:00:00
>>>>>45...Nf5
>>>>>  ±  (1.09)   Depth: 1/6   00:00:00
>>>>>45...Nf5 46.Ra5
>>>>>  ±  (1.39)   Depth: 2/6   00:00:00
>>>>>45...Nf5 46.g4 Ne7
>>>>>  ±  (1.37)   Depth: 3/6   00:00:00
>>>>>45...Nf5 46.Nc7 Rxa2+ 47.Kxa2 Kg8
>>>>>  +-  (1.77)   Depth: 4/8   00:00:00
>>>>>45...Qf5 46.g4 Qf3 47.Ra5
>>>>>  +-  (1.58)   Depth: 4/8   00:00:00
>>>>>45...Qf5 46.g4 Qd7 47.Nc7 Rab8
>>>>>  +-  (1.66)   Depth: 5/11   00:00:00
>>>>>45...Qf5 46.Nc7 Rxa2+ 47.Kxa2 Bxe3 48.Ba6 Rxc7 49.fxe3
>>>>>  +-  (1.90)   Depth: 6/14   00:00:00  39kN
>>>>>45...Qf5 46.Nc7 Rc8 47.Ba6 Rcxc7 48.bxc7 Rxc7
>>>>>  +-  (2.51)   Depth: 7/15   00:00:01  157kN
>>>>>45...Nf5 46.Nc7 Rab8 47.Ra7 Qxc7 48.bxc7 Rxa7+ 49.Kb2 Rxb3+ 50.Kxb3 Rxc7
>>>>>  +-  (2.48)   Depth: 7/15   00:00:01  188kN
>>>>>45...Qc8 46.Nc7 Rxa2+ 47.Qxa2 Ne8 48.Nxe8 Qxe8 49.Ba6 Rb8
>>>>>  +-  (2.33)   Depth: 7/15   00:00:02  217kN
>>>>>45...Ne8 46.Nb4 Rbb8 47.Rbb2 Rxa2+ 48.Rxa2 Qb7 49.Ra7
>>>>>  +-  (1.86)   Depth: 7/15   00:00:02  265kN
>>>>>45...Ne8 46.Nb4 Rbb8 47.Rbb2 Kg8 48.Rxa8 Rxa8+ 49.Ra2 Rb8
>>>>>  +-  (2.15)   Depth: 8/22   00:00:04  419kN
>>>>>45...Ne8 46.Nb4 Rbb8 47.Rba3 Rxa3 48.Rxa3 Nf6 49.Ra7 Rb7
>>>>>  +-  (2.32)   Depth: 9/22   00:00:08  1123kN
>>>>>45...Ne8 46.Nb4 Rbb8 47.h4 Rxa2+ 48.Kxa2 Qb7 49.Ba6 Ra8 50.Ra3 Qb8
>>>>>  +-  (2.41)   Depth: 10/23   00:00:16  2605kN
>>>>>45...Ne8 46.Nb4 Rbb8 47.h4 Kg8 48.Ba6 Nf5 49.Kb1 g6 50.Rab2 Bg7
>>>>>  +-  (2.46)   Depth: 11/24   00:00:32  5642kN
>>>>>45...Ne8 46.Nb4 Rbb8 47.h4 Kg8 48.Rba3 Rxa3 49.Rxa3 Qb7 50.Ra7 Ra8 51.Rxa8 Qxa8+
>>>>>  +-  (2.48)   Depth: 12/25   00:01:00  11240kN
>>>>>45...Ne8 46.Nb4 Rbb8 47.h4 Kg8 48.Rxa8 Rxa8+ 49.Ba6 Qc8 50.Ra3 Qd8 51.b7 Rb8
>>>>>52.Kb1
>>>>>  +-  (2.54)   Depth: 13/26   00:01:54  22385kN
>>>>>
>>>>>(McCracken, None 17.11.2003)
>>>>>
>>>>>
>>>>>Congratulations Dr. Robert Hyatt!
>>>>>
>>>>>Also, Amir and Shay for being close...maybe Deep Junior 8 would have done
>>>>>better?...I can't test it.
>>>>>
>>>>>Terry
>>>>
>>>>Thanks, but I don't think it's deserved. I get only +0.30 on my notebook.
>>>>
>>>>If you ask me, +0.30 is better than +0.80 on move 14, because I don't see that
>>>>white is better.
>>>>
>>>>IIRC Crafty has asymetric evaluation and its eval from the white side will be
>>>>different.
>>>>
>>>>Why is it urgent to play f5 at move 15 ? What's wrong with 5 moves later ?
>>>>
>>>>Amir
>>>
>>>IMO White is clearly better, and was better throughout most of the game, by move
>>>14 I feel that Black had to fight for a draw. I doubt 15.f5! would have saved
>>>Black, but it was the best practical chance. Leaving it another 5 moves, gives
>>>White 5 moves to continue to improve and consolidate his position.
>>>
>>>Fritz had no chance in this setup....long term planning is above all programmes,
>>>even Crafty.
>>>
>>>I think Juniors style is top notch, and Rebel for that matter, but until enough
>>>knowledge and speed, and the right algorithms are created and perfected, long
>>>range planning will allude all computers.
>>>
>>>But I believe that day will come, when it all comes together, and in my
>>>lifetime.
>>>
>>>Chess programmes and I are about the same age, so I'm optimistic;-)
>>>But I'm not optimistic about my chances to beat them ever, by then:/
>>>
>>>Terry
>
>If a program doesn't have the correct algorithm to handle closed positions like
>crafty, it will goes into a planless loop, searching and searching without
>finding the correct plan, just like this post if you follow the link underneath
>:-)
>
>http://www.talkchess.com/forums/1/message.html?328315

LOL! I did, you swine! I won't repeat that mistake!;-)



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.