Author: Tim Foden
Date: 02:18:09 04/26/01
Go up one level in this thread
On April 25, 2001 at 14:51:16, leonid wrote: >On April 25, 2001 at 12:15:15, Tim Foden wrote: > >>On April 25, 2001 at 11:48:17, leonid wrote: >> >>>Hello! >>> >>>If you want to know how much others understand what you are been talking about, >>>then ask your computer to solve this problem. >>> >>>[D]3k4/r2b2r1/3b4/1Q1Q1Q2/1qnqnq2/1N1B1N2/QpQBQpQ1/q1RKR1q1 b - - >>> >>>Please indicate your result. >> >>Green Light Chess v.214-pr8, 48MB hash, 920MHz Duron: >> >>>anal >> Game stage: Opening >> Current eval: -15.84 >> Ply Time Score Nodes Principal Variation >> 1 0.02 -5.283 6114 Rxg2=R+> >> 2 0.03 ++ 6914 fxe1=R+ (a=-5.78 b=-4.78 e=-4.78) >> 2 0.17 1.234 66233 fxe1=R+ Nxe1 bxc1=R+ Qxc1 Qxa2 Qxg1 Bxb5 Qgxd4 Qxf5 >> 2 0.20 1.234 80579 fxe1=R+ Nxe1 bxc1=R+ Qxc1 Qxa2 Qxg1 Bxb5 Qgxd4 Qxf5 >> 3 0.71 -- 344742 fxe1=R+ (a=0.73 b=1.73 e=0.73) >> 3 0.82 ++ 375845 fxe1=R+ (a=-321.00 b=0.73 e=0.73) >> 3 1.24 1.234 650671 fxe1=R+ Nxe1 bxc1=Q+ Qxc1 Qxa2 Qxg1 Bxb5 Qgxd4 Qxf5 >> 3 1.24 1.234 650830 fxe1=R+ Nxe1 bxc1=Q+ Qxc1 Qxa2 Qxg1 Bxb5 Qgxd4 Qxf5 >> 4 4.82 -- 2497050 fxe1=R+ (a=0.73 b=1.73 e=0.73) >> 4 5.60 0.311 2958369 fxe1=R+ Nxe1 bxc1=R+ Nxc1 Bxf5 Qxd6+ Qfxd6 Qxb4 >> Rxa2 Qxd6+ Qxd6 Qxg1 >> 4 7.36 0.311 3946113 fxe1=R+ Nxe1 bxc1=R+ Nxc1 Bxf5 Qxd6+ Qfxd6 Qxb4 >> Rxa2 Qxd6+ Qxd6 Qxg1 >> 5 13.81 ++ 7298057 fxe1=R+ (a=-0.19 b=0.81 e=0.81) >> 5 18.97 1.429 10341k fxe1=R+ Nxe1 Bxf5 Bxf4 bxc1=Q+ Nxc1 Qbxe1+ Qxe1 >> Bg4+ Qce2 Bxe2+ Qaxe2 Rxg2 Qxg1 Rxg1+ >> 5 33.96 1.429 19412k fxe1=R+ Nxe1 Bxf5 Bxf4 bxc1=Q+ Nxc1 Qbxe1+ Qxe1 >> Bg4+ Qce2 Bxe2+ Qaxe2 Rxg2 Qxg1 Rxg1+ >> 6 1:32 ++ 47913k fxe1=R+ (a=0.93 b=1.93 e=1.93) >> 6 1:52 6.936 60816k fxe1=R+ Nxe1 Qfxd2+ Qcxd2 Qxd2+ Qxd2 bxc1=R+ Nxc1 >> Nc3+ Qxc3 Ne3+ Ke2 Rxg2+ Qxg2 Rxa2+ Kf3 Qxg2+ >> Nxg2 Nxf5 Qxa1 Rxa1 >> 6 3:39 Mate09 125187k Qfxd2+ Qcxd2 Qxd2+ Qxd2 Nc3+ Qxc3 Ne3+ Kd2 Qxc3+ >> Ke2 fxe1=R+ Nxe1 Qcxe1+ Rxe1 Qaxe1+ Kf3 Qxg2# >> <ht> >> 6 3:45 Mate09 129508k Qfxd2+ Qcxd2 Qxd2+ Qxd2 Nc3+ Qxc3 Ne3+ Kd2 Qxc3+ >> Ke2 fxe1=R+ Nxe1 Qcxe1+ Rxe1 Qaxe1+ Kf3 Qxg2# >> <ht> >> 7 4:47 Mate09 190284k Qfxd2+ Qcxd2 Qxd2+ Qxd2 Nc3+ Qxc3 Ne3+ Kd2 Qxc3+ >> Ke2 fxe1=R+ Nxe1 Qcxe1+ Rxe1 Qaxe1+ Kf3 Qxg2# >> 7 13:11 Mate09 673532k Qfxd2+ Qcxd2 Qxd2+ Qxd2 Nc3+ Qxc3 Ne3+ Kd2 Qxc3+ >> Ke2 fxe1=R+ Nxe1 Qcxe1+ Rxe1 Qaxe1+ Kf3 Qxg2# >> 8 13:11 0/96 Qfxd2+ >exit >> local: t=13:18 nps=852316.3 n=680342699 (f=98631584 q=581711115) >> total: t=13:18 nps=852316.3 n=680342699 draws=31431 >> trans: probes=58874345 hits=9318764 (15.83%) draft=3940488 (6.69%) >> tcuts: exact=208 (0.00%) upper=2226487 (3.78%) lower=1257502 (2.14%) >> tstor: exact=519 (0.00%) upper=35637627 (78.27%) lower=9896310 (21.73%) >> ext: check=77088620 recap=3252308 ppush=0 1rep=2075102 thrt=0 >> q-moves: gen=221211184 tested=38828955 made/un=25177634 max-dep=6 >> max eval diff: part-1=3.005 part-2=3.780 >>> >> >>Cheers, Tim. > >Hi, Tim! > >Can you say me in simple English, what was your average NPS for this position? I >can see as much as 852k, that is really quick search. I remember only one >position, that had 117 initial legal moves, and that went as far as 1200 per >second. Yes, the NPS is 852k. You can confirm this yourself since it searched 680342699 nodes in 13:18 (798 seconds), so NPS = 680342699 / 791 = 852K I wouldn't read too much into NPS values though, as I can easily make changes that make it less, but the engine is still as strong. Simply adding some kind of Futility Pruning to the search does this (as the search does proportionally fewer quiescence nodes, which are very fast). > >Actual position on my solver was 251k for selective and 46k for brute force. > >Salut, >Leonid.
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.