Author: Uri Blass
Date: 01:47:33 09/01/02
Go up one level in this thread
On September 01, 2002 at 04:30:44, Alexander Kure wrote: >Hi Steffen, > >Here is Brutus' analysis: > >D]r1b2r2/pp1p1p1k/1qn2Ppp/5p2/3p4/3B1N2/PPPQ1PPP/R3R1K1 w - - 0 1 > >Analysis by Brutus-V099.011: > >1.c3 dxc3 > ² (0.28) Depth: 5/1 00:00:01 >1.c3 dxc3 2.Qxc3 Nb4 > = (0.10) Depth: 7/5 00:00:01 >1.Rab1 > = (0.14) Depth: 7/6 00:00:01 1kN >1.Rab1 Qd8 2.Bc4 Qxf6 > ² (0.28) Depth: 7/9 00:00:01 5kN >1.Rab1 Qd8 2.Rbd1 d5 3.Bb5 > = (0.25) Depth: 8/9 00:00:01 6kN >1.Qf4 > ² (0.29) Depth: 8/9 00:00:01 19kN >1.Qf4 Qxb2 2.Qh4 h5 3.Bxf5 Kg8 > ² (0.52) Depth: 8/12 00:00:01 39kN >1.Qf4 d5 2.Qh4 h5 3.b3 Bd7 > ± (0.72) Depth: 9/12 00:00:02 43kN >1.Qf4 d6 2.b3 Qc5 3.Qh4 h5 4.Qf4 > ² (0.70) Depth: 10/14 00:00:05 167kN >1.Qf4 Nd8 2.Qxd4 Ne6 3.Qxb6 axb6 4.Bc4 Nf4 5.Re7 > ± (0.85) Depth: 11/16 00:00:16 656kN >1.Qf4 > ± (1.17) Depth: 12/20 00:00:32 1385kN >1.Qf4 > ± (1.16) Depth: 12/22 00:00:36 1521kN >1.Qf4 d5 2.Qh4 h5 3.g4 Kg8 4.Qg5 > ± (1.02) Depth: 12/22 00:00:37 1562kN >1.Qf4 > ± (1.34) Depth: 13/25 00:01:28 3863kN >1.Qf4 > ± (1.33) Depth: 13/26 00:01:30 3958kN >1.Qf4 d5 2.Qh4 h5 3.g4 Nb4 4.gxh5 g5 5.Nxg5+ Kh8 6.h6 Qxf6 7.Qh5 > ± (1.12) Depth: 13/26 00:01:32 4067kN >1.Ng5+ > ± (1.16) Depth: 13/28 00:01:48 4777kN >1.Ng5+ > +- (1.58) Depth: 13/31 00:02:51 7311kN >1.Ng5+ hxg5 2.Qxg5 Nb4 3.Re5 d5 4.Bxf5 Bxf5 5.Qh4+ Kg8 6.Rxf5 Nxc2 7.Rh5 Qxf6 >8.Qxf6 > +- (1.98) Depth: 13/31 00:03:42 9146kN >1.Ng5+ > +- (2.30) Depth: 14/31 00:05:23 12948kN >1.Ng5+ > +- (2.72) Depth: 14/37 00:07:03 16468kN >1.Ng5+ > +- (4.00) Depth: 14/37 00:10:53 24922kN >1.Ng5+ > +- (3.99) Depth: 14/37 00:11:15 26417kN >1.Ng5+ hxg5 2.Qxg5 Nb4 3.Re5 Nxd3 > +- (4.06) Depth: 14/37 00:13:03 29645kN >1.Ng5+ hxg5 2.Qxg5 Nb4 3.Re5 Nxd3 4.Qh4+ Kg8 5.Rxf5 Ne5 6.Qh6 Qxf6 7.Rxf6 d6 >8.Rf4 Ng4 9.Qh4 d3 10.cxd3 Ne5 > +- (4.12) Depth: 15/37 00:17:54 41005kN >1.Ng5+ hxg5 2.Qxg5 Nb4 3.Re5 Nxd3 4.Qh4+ Kg8 5.Rxf5 Ne5 6.Qh6 Qxf6 7.Rxf6 d6 >8.Rf4 Ng4 9.Qh4 Bf5 10.h3 Ne5 11.Qh6 > +- (4.11) Depth: 16/37 00:24:56 61392kN I am surprised by the small number of nodes per second for brutus. 61392knodes/1496 seconds is less than 50knodes per second. I thought that brutus was supposed to be a fast searcher because it is not a software but hardware that was designed for chess. > >Please keep in mind that a pawn is 0.64 for Brutus (factor = 25/16). I have no clear value to decide how much is a pawn in my program because it is dependent on the square of the pawn and on positional factors(it is less than 1 if it is doubled pawn and it is more than 1 if it is in the 6th rank or the 7th rank. If we want to know what is the meaning of the numbers it is better to take some epd file of random positions from games when one side has a pawn advantage and to calculate the average of the advantage of the better side. 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.