Author: Uri Blass
Date: 11:16:29 09/18/01
Go up one level in this thread
On September 18, 2001 at 13:05:08, Gian-Carlo Pascutto wrote: >Like many others here, I use a standard testset >for testing improvements and doing regression >testing to my engine. > >So far WAC has been adequate, but as you all >know it has two annoying problems, namely that >over 2/3 of the problems are solved by 2 ply >searches and hence useless, and the level of >most of the others isn't very hard either with >nowadays CPU's. > >ECM could have been better, but it's very big >(over 800 positions) and contains a lot of garbage. > >Based on some parts of ECM I got from Dann's >ftp and a day with Goliath Light, Hiarcs and >my common sense I filtered 200 positions of >varying difficulty which have, as far as I could >tell and verify, correct solutions. >The suite is meant for relatively quick testing >of improvements and/or regressions. The recommended >time is 10 seconds per position. Running all >positions takes a little over half an hour. It >is what I currently use instead of WAC. > >I would appreciate feedback about the positions, >especially if you think there is an incorrect >or dual soltion. Performance data of other amateur >or professional programs is appreciated too. >My latest Sjeng gets 80/200 at 10s on an Athlon 1000. > <snipped> >2r2bk1/4qp2/3n2p1/2R1p1Np/2p1N3/r6P/1Q3PP1/3R2K1 w - - bm Rxc8; id "ECM.973"; I am not sure about it but the following line may be an alternative solution: 1.Rxd6 Rxc5 2.Qxa3 Rd5(Rc6 3.Qf3 Rxd6 with the same position as the main line) 3.Qf3 Rxd6 4.Nxd6 e4 5.Ngxe4 f5 6.Nxf5 gxf5 7.Qxf5 Short search suggests that black has no good alternative and that black is 2 pawns down at the end of the main line. Here is Deep fritz's opinion for the position after 1.Rxd6 New position 2r2bk1/4qp2/3R2p1/2R1p1Np/2p1N3/r6P/1Q3PP1/6K1 b - - 0 1 Analysis by Deep Fritz: 1...Qxd6 2.Rxc8 +- (1.41) Depth: 1/8 00:00:00 1...Rxc5! ± (1.34) Depth: 1/8 00:00:00 1...Rxc5! 2.Qxa3 ± (1.16) Depth: 1/10 00:00:00 1...Rxc5 2.Qxa3 ± (1.16) Depth: 2/8 00:00:00 1...Rxc5 2.Qxa3 Qxg5 3.Nxg5 ± (1.12) Depth: 3/12 00:00:00 1kN 1...Rb3! ± (1.09) Depth: 3/16 00:00:00 2kN 1...Rb3 2.Qxb3 cxb3 3.Rxc8 ± (1.28) Depth: 4/17 00:00:00 5kN 1...Rxc5! ± (1.25) Depth: 4/17 00:00:00 6kN 1...Rxc5! 2.Nf6+ Kg7 3.Qxa3 Qxf6 4.Rxf6 ± (1.03) Depth: 4/17 00:00:00 7kN 1...Rxc5! ± (0.72) Depth: 5/15 00:00:00 22kN 1...Rxc5! 2.Qxa3 Rd5 3.Qf3 = (0.00) Depth: 5/18 00:00:00 33kN 1...Rxc5 2.Qxa3 Rd5 3.Qf3 Rxd6 4.Nxd6 Qxd6 5.Qxf7+ Kh8 = (0.00) Depth: 6/19 00:00:00 46kN 1...Rxc5-- ² (0.47) Depth: 7/15 00:00:00 65kN 1...Rxc5-- 2.Qxa3 Rd5 3.Qf3 Rb5 4.Nf6+ Kg7 5.Nge4 Rb1+ 6.Kh2 +- (1.66) Depth: 7/22 00:00:00 103kN 1...Rb3! +- (1.62) Depth: 7/22 00:00:00 151kN 1...Rb3! 2.Qxb3 cxb3 3.Rxc8 Kg7 4.Kf1 b2 ± (0.97) Depth: 7/24 00:00:00 220kN 1...Rb3 2.Qxb3 cxb3 3.Rxc8 Kg7 4.Kf1 b2 5.Rd1 ± (0.87) Depth: 8/23 00:00:00 406kN 1...Rb3 2.Qxb3 cxb3 3.Rxc8 b2 4.Rb8 Qa7 5.Rdd8 Qa1+ 6.Kh2 ± (0.91) Depth: 9/24 00:00:01 883kN 1...Rb3-- ± (1.22) Depth: 10/24 00:00:01 1191kN 1...Rb3-- 2.Qxb3 cxb3 3.Rxc8 Kg7 4.Rb6 Qa3 5.Rf6 Qa1+ 6.Kh2 b2 7.Rxf7+ +- (1.75) Depth: 10/26 00:00:02 1720kN 1...Rxc5! +- (1.72) Depth: 10/26 00:00:04 3287kN 1...Rxc5 2.Qxa3 Rd5 3.Qf3 Rxd6 4.Nxd6 e4 5.Qxf7+ Qxf7 6.Ngxf7 c3 7.Kf1 +- (1.47) Depth: 11/26 00:00:06 4982kN 1...Rxc5! ± (1.16) Depth: 12/30 00:00:18 13494kN 1...Rxc5! 2.Qxa3 Rd5 3.Qf3 Rxd6 4.Nxd6 e4 5.Qf4 f6 6.Ngxe4 f5 ± (1.06) Depth: 12/30 00:00:20 15375kN 1...Rxc5 2.Qxa3 Rd5 3.Qf3 Rxd6 4.Nxd6 e4 5.Ndxe4 f5 6.Qe2 fxe4 7.Qxc4+ ± (0.97) Depth: 13/31 00:00:43 32361kN 1...Rxc5 2.Qxa3 Rd5 3.Qf3 Rxd6 4.Nxd6 e4 5.Ndxe4 f5 6.Qe2 fxe4 7.Qxc4+ ± (1.00) Depth: 14/34 00:01:39 74383kN 1...Rxc5 2.Qxa3 Rd5 3.Qf3 Rxd6 4.Nxd6 e4 5.Ndxe4 f5 6.Qe2 ± (1.00) Depth: 15/35 00:05:49 257310kN 1...Rxc5-- ± (1.31) Depth: 16/34 00:07:55 350309kN 1...Rxc5-- ± (1.31) Depth: 16/40 00:14:40 648326kN (Blass, Tel-aviv 18.09.2001) 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.