Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Anand too deep today for the rest of the world, even Hydra

Author: Sandro Necchi

Date: 23:38:02 10/01/05

Go up one level in this thread


On October 01, 2005 at 18:33:27, Eelco de Groot wrote:

>On October 01, 2005 at 05:45:27, Sandro Necchi wrote:
>
>>On October 01, 2005 at 04:31:46, Sandro Necchi wrote:
>>
>>>Hi,
>>>
>>>well, Shredder 9 Columbus' egg 9c will alternate Be3 and Qd2 as white after a
>>>while (if you have a fast computer), but I am not sure this setting is better
>>>thant Columbus' egg 9b as I have not tested yet.
>>>
>>>Maybe someone can.
>>>
>>>Columbus' egg 9c
>>>
>>>by Sandro Necchi
>>>
>>>Queen = 99
>>>100
>>>Rook = 97
>>>100
>>>Bishop = 97
>>>100
>>>Knight = 99
>>>100
>>>Pawn = 100
>>>100
>>>Style = Intelligent
>>>Combinations = all
>>>Forward pruning = 3
>>>Selectivity = -1
>>>Draw score = -15
>>                               King safety = 135
>>>Center control = 104
>>>Pawn structure = 100
>>>102
>>>Passed pawn = 102
>>>100
>>>Bishop pair = 101
>>>100
>>>Time usage = 100
>>>Knowledge or speed = 100
>>>Threads = 1
>>>
>>>Have fun
>>>
>>>Sandro
>>
>>It will show Qd2 as best move if you select the option show me the best 3 moves
>>after some time.
>>
>>I guess in normal mode it will take much longer to find that move as it is
>>cutting that line too early.
>>
>>Sandro
>
> Hello Sandro!
>
>Thanks for your Shredder Columbus'Egg settings. I occasionally have let the
>first Columbus' Egg analyze some positions in my Shredder 9.02 but not often
>enough to be able to say whether it is an improvement. That could only be done
>with many games and I still have 150 Pro Deo Vulcan versions that also need
>testing :)

Well, the best version of the Columbus' egg is the 9b which I am using now until
the first versions of Shredder 10 will arrive...

>
>I could play testgames of Pro Deo Vulcan against Shredder in the Shredder
>interface but I do not entirely trust Pro Deo's communications in the interface
>as UCI engine. So I stick with ERT in ChessPartner for engine-engine games.
>
>In tuning Pro Deo settings I am always a bit wary of trying very small changes
>like 99, 101 etc or even uneven numbers. Not sure if that is really justified,
>the only exception being Knight Mobility = 88 that is still there as other
>numbers seem worse and things like My Queen = 96, Extensions (remaining) = 3

Well, with Shredder there are only some parameters one can change, so the
changes are limited.

>
>What I like trying out in Shredder Classic interface is among other things the
>facilities for making books. I even started with my own opening book,
>newbook1.bkt, so far it only knows 7 moves in the French. But all of them
>manually entered! I also would like to add to some comments. At this rate I can
>do no more than one opening in a year maybe...

I agree making/changing books is so simple in Shredder classic GUI that I use
only this one as it is difficult to accept systems with are not so easy to use
when there is one so good and simple!

>
>
>More about Anand's novelty: Pro Deo Vulcan {9k17pc} has just also found 23.Qd2!
>At 18 ply with a score of +0.52 after about 5½ hours, with Shredder partly
>sharing the CPU at the end.

Well, Colubus' egg 9b prefer Qd2 too, but the evaluation is about equal and I
think it is correct.

>
>Pro Deo Vulcan {9k17pc} Athlon 2009 MHz
>
> 1.e4 e5 2.Nf3 Nc6 3.Bb5 a6 4.Ba4 Nf6 5.O-O Be7 6.Re1 b5 7.Bb3 d6 8.c3
> O-O 9.h3 Bb7 10.d4 Re8 11.Nbd2 Bf8 12.a4 h6 13.Bc2 exd4 14.cxd4 Nb4
> 15.Bb1 c5 16.d5 Nd7 17.Ra3 c4 18.axb5 axb5 19.Nd4 Qb6 20.Nf5 Ne5 21.Rg3
> g6 22.Nf3 Ned3
>
> [D]r3rbk1/1b3p2/1q1p2pp/1p1P1N2/1np1P3/3n1NRP/1P3PP1/1BBQR1K1 w - -
>
> 00:00:00   1.00  -0.96  23.Bxd3 Nxd3
> 00:00:00   1.06  0.33  23.Be3
> 00:00:00   2.00  -0.07  23.Be3 Qc7
> 00:00:00   2.26  0.17  23.Bxd3 Nxd3 24.Be3
> 00:00:00   3.00  -0.51  23.Bxd3 Nxd3 24.Be3 Qc7 25.Bxh6
> 00:00:00   3.01  -0.20  23.Be3 Qc7 24.Nxh6 Bxh6
> 00:00:00   4.00  -0.20  23.Be3 Qc7 24.Nxh6 Bxh6 25.Bxh6
> 00:00:00   4.02  -0.20  23.Re2
> 00:00:00   4.02  0.15  23.Re2 Kh7 24.Be3 Qc7
> 00:00:00   5.00  -0.08  23.Re2 Nxc1 24.Qxc1 Ra1 25.Nxh6 Bxh6 26.Qxh6
> 00:00:00   6.00  -0.30  23.Re2 Nxc1 24.Qxc1 Ra1 25.N3d4 Nd3 26.Qd1
> 00:00:00   6.01  -0.19  23.Be3 Qa5 24.Nxh6 Bxh6 25.Bxh6
> 00:00:00   7.00  0.06  23.Be3 Qd8 24.Bxh6 Nxe1 25.Nxe1 Qf6 26.Qd2
>                        Ra1 27.Qxb4
> 00:00:00   8.00  0.06  23.Be3 Qd8 24.Bxh6 Nxe1 25.Nxe1 Qf6 26.Qd2
>                        Ra1 27.Qxb4
> 00:00:01   9.00  -0.15  23.Be3 Qd8 24.Bd4 Nxe1 25.Nxe1 Ra1 26.Qh5
>                         Kh7
> 00:00:03  10.00  0.06  23.Be3 Qd8 24.Bxh6 Nxe1 25.Nxe1 Qf6 26.Qd2
>                        Ra1 27.Qxb4 Rxb1
> 00:00:08  11.00  0.02  23.Be3 Qc7 24.Nxh6 Bxh6 25.Bxh6 Nxe1 26.Nxe1
>                        Qe7 27.Bd2 Na6 28.Qh5 Nc5
> 00:00:19  12.00  0.07  23.Be3 Qd8 24.Bxh6 Nxe1 25.Qxe1 Ra1 26.Qxb4
>                        Rxb1 27.Kh2 Bc8 28.Bxf8 Rxf8
> 00:00:53  13.00  0.05  23.Be3 Qc7 24.Nxh6 Bxh6 25.Bxh6 Nxe1 26.Nxe1
>                        Qe7 27.Bd2 Na6 28.Qh5 Qe5 29.Qxe5 Rxe5 30.Ra3
> 00:01:54  14.00  0.23  23.Be3 Qc7 24.Bxh6 Nxe1 25.Qxe1
> 00:06:53  15.00  0.21  23.Be3 Qc7 24.Bxh6 Bc8 25.Bxf8 Kxf8 26.Qd2
>                        gxf5 27.Qh6
> 00:19:55  16.00  0.16  23.Be3 Qd8 24.Bxh6 Nxe1 25.Qxe1 Nxd5
> 00:37:00  17.00  0.23  23.Be3 Qd8 24.Bxh6 Nxe1 25.Qxe1 Ra1 26.Qxb4
>                        Rxb1 27.Kh2 Rxe4
> 01:39:21  18.00  0.24  23.Be3 Qd8 24.Bxh6 Bc8 25.Bg5 Qb6 26.Be3 Qa5
>                        27.N3h4 Bxf5 28.exf5 Bg7 29.fxg6 Nxe1 30.Qxe1
> 03:24:34  18.04  0.24  23.Rf1 (This was a false Fail High)
> 04:36:48  18.05  0.24  23.Qd2
> 05:31:16  18.05  0.51  23.Qd2 Nxe1 24.Nxe1 Nxd5 25.Nxh6 Bxh6
>
>I think the continuation 23.Qd2 Nxe1 24.Nxe1 Nxd5 25.Nxh6 Bxh6  and then
>26.Qxh6 Ra1 is equal to the transposition that Fruit found but probably Pro Deo
>Vulcan did not see 26..Ra1 yet. After 26..Ra1 Shredder thinks it is a draw, no
>more than +0.06 for 27.Txg6 or 27.e5.

Yes, 24...Nd5 is surely stronger than Adams move and should be a draw.

After 23.Be3 Qd8 24.Bxh6 the move Qf6!
>from Yan vs Lin 1988 seems indeed strong and refuting 23.Be3

Well, this variation would give black some advantage improving the below
variation with 27...Nd5 -0.96 19/44 (in black favor; Columbus' egg 9b) expecting
the following moves 28. Rg4 Nf6 29. Rh4 Nh5 30. b3 f6 31. Bd2 g5 32. Rg4 Nhf4
33. Bc3 Rb1 34. Qb1 Ne2+ 35. Ne2 Be4 36. Rg5 Kf7
>
>[D]r3rbk1/1b3p2/3p1qpB/1p1P1N2/1np1P3/3n1NRP/1P3PP1/1B1QR1K1 w - -
>
>Shredder 9.02 in background:
>19/45  41:28  -0.34    25.Te2 Ta1 26.Lg5 Dh8 27.P5d4 Pxd5
>                       28.exd5 Txe2 29.Pxe2 Dxb2 30.Pd2 Lxd5
>                       31.Te3 f6 32.Lf4 g5 33.Tg3 (555.037.663) 223
>
>You can add this to Shredder 10 book!

Yes, but up to 27...Nd5!=+
>
> Eelco

Sandro



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.