Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: position

Author: Sune Fischer

Date: 03:00:47 11/05/02

Go up one level in this thread


On November 05, 2002 at 05:12:25, Uri Blass wrote:

>On November 05, 2002 at 04:28:59, Sune Fischer wrote:
>
>>On November 05, 2002 at 00:39:31, Will Singleton wrote:
>>
>>>White to play, bm is d7.  How long until your engine shows a positive score?
>>>(amd 1.6ghz).
>>>
>>>Amateur  ply 15  score 33  secs 212
>>>Yace     ply 15  score 9   secs 103
>>>Crafty   ply 16  score 40  secs 69
>>>Ruffian  ply 12  score 51  secs 5  (just amazing)
>>>
>>>[D]5kb1/6pp/p2P1p2/1p6/2r5/P1r2N1P/6P1/2R2RK1 w - - 0 36
>>>
>>>(remember, d7 must show a positive score)
>>>
>>>--------------------
>>>
>>
>>Yeah man, Frenzee: d7 in 0.00 secs with a positive score :)
>>
>>2	-614	0	66		1.g3 Rxc1 2.Rxc1 Rxc1+
>>2	-416	0	152		1.d7 Rxc1 2.Rxc1 Rxc1+
>
>no the score is negative.

Yeah, at ply 2, but it reaches ply three and four in 0.00 secs also with
positive score :)
Ply 2 is the first iteration, so it will have a lot of fail highs at ply 2
typically, almost random move ordering of course.

>It seems that Frenzee does not do pawn promotion in the qsearch.

Right, not if they aren't captures, but I'm change this as we speak actually :)

>My qsearch detects 1.d7 Rxc1 2.d8Q so I have not this stupid line.
>
>Movei's buggy version can show positive score for d7 after search only at depth
>13.

Well I get it at ply 3, _three_! :)

>depth=13 +0.37 d6d7 f8e7 c1d1 e7d8 f1e1 g8f7 f3d4 c3c1 d1c1 c4d4 c1c6 f7c4 g1f2
>d4d2 f2g3 d8d7 c6a6
>Nodes: 19739458 NPS: 160353
>Time: 00:02:03.10
>
>depth=14 +0.27 d6d7 f8e7 c1d1 e7d8 f1e1 g8f7 f3d4 c3c1 d1c1 c4d4 c1c6 d8d7 c6a6
>d4a4 e1d1 d7e7 a6b6 a4a3 d1e1 e7d7 b6b5
>Nodes: 34345849 NPS: 161993
>Time: 00:03:32.02
>
>
>>2	-386	0	256		1.Nd2 Rxc1 2.Nxc4 Rxf1+ 3.Kxf1 Bxc4+
>>2	-60	0	365		1.Ra1 Re4 2.axb4
>>2	-49	0	430		1.Rxc3 Rxc3
>>3	25	0	600		1.Rxc3 Rxc3 2.d7 Rxa3
>>3	144	0	1342		1.d7 Rd3 2.Rxc4 bxc4
>>4	145	0	2889		1.d7 Ke7 2.Rxc3 Rxc3
>
>White is a pawn down after d7 Ke7 Rxc3 Rxc3.
>
>I think that it is a mistake to give more than a pawn bonus for a pawn in the
>7th.

May be, maybe not, depends on the position :)
Guess it is a little lucky here, but according to my observations pawns
promoting is quite often the deciding factor in practical games, so no I don't
consider a pawn too much, in particular when backed up by rooks like here.

-S.

>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.