Author: Robert Hyatt
Date: 10:30:09 11/29/01
Go up one level in this thread
On November 28, 2001 at 17:01:48, Dann Corbit wrote: >Subject: Re: Has anyone ever formed a board position with more legal moves than >this? > >Posted by Dann Corbit (Profile) on November 28, 2001 at 16:58:25: > >In Reply to: Re: Has anyone ever formed a board position with more legal moves >than this? posted by Dann Corbit on November 28, 2001 at 16:54:41: > > >On November 28, 2001 at 16:54:41, Dann Corbit wrote: > >>Some really simple checkmates. >>Mate in 2: >>[D]n1r1r1b1/1P1P1P1P/1Q6/3NBNK1/R7/4p1p1/3PBPkP/2R5 w - - bm Nf4+; >> >>Mate in 1: >>[D]4Q1qk/1Q4RN/3Q3N/Q4Q1B/2Q4B/4Q3/1Q4R1/3Q3K w - - bm Nf7# Qcxg8# Qexg8# R7g3# >>R7g4# R7g5# R7g6# Rc7# Rd7# Re7# Rf7# Rxg8#; >> >>[D]5k2/2K5/3N1B2/P1NB4/6Q1/4R3/P1PP1P1P/1R6 w - - bm Be7# Bg7# Nd7# Ne6# Qc8# >>Qg7# Qg8# Rb8# Re8#; >> >>[D]3Q4/1Q4Q1/4Q3/2Q4R/Q4Q2/3Q4/1Q4Rp/1K1BBNNk w - - bm Ng3# Qxh2# Rc2# Rd2# Re2# >>Rf2# Rg3# Rg4# Rg6# Rgg5# Rgxh2# Rhxh2#; >> >>[D]q2Q3r/n6R/kpB1N1K1/p1p1Bppp/1PN3P1/1n1pp1b1/P1PPPP1P/r5Rb w - - bm Nc7# >>Qxb6#; >> >>[D]3nkn2/4P3/4N2P/4B3/P1N1B1K1/1R6/5Q2/3R4 w - - bm Qxf8# exd8=Q# exf8=Q#; >> >>How long does your program take? >>The last one gave crafty a cross-eyed hissy fit on my machine. >Maybe it was the last two. > >The strangest part is that it took over five minutes of wall time to complete >this test (with the last two positions getting the lion's share) and yet crafty >only reported a couple seconds time used?! > >E:\crafty\release>crafty1 >EPD Kit revision date: 1996.04.21 >unable to open book file [e:\crafty\release/books.bin]. >hash table memory = 192M bytes. >pawn hash table memory = 80M bytes. >EGTB cache memory = 32M bytes. >draw score set to 0.00 pawns. >choose from book moves randomly (using weights.) >choose from 5 best moves. >book learning enabled >result learning enabled >position learning enabled >threshold set to 9 pawns. >5 piece tablebase files found >19045kb of RAM used for TB indices and decompression tables > >Crafty v18.12 > >White(1): epdpfga \manyt.epd >This command takes two parameters >EG fault: a problem occurred during epdpfga processing >White(1): epdpfga \many.epd many.out >PFGA: EPD record: 1 >middle-game phase > clearing hash tables > time surplus 0.00 time limit 30.00 (3:30) > nss depth time score variation (1) > 1 0.03 Mate 1. Rxg8# > 1-> 0.03 Mate 1. Rxg8# > 2 0.03 Mate 1. Rxg8# > 2-> 0.03 Mate 1. Rxg8# > 3 0.03 Mate 1. Rxg8# > 3-> 0.03 Mate 1. Rxg8# > 4 0.03 Mate 1. Rxg8# > 4-> 0.03 Mate 1. Rxg8# > time=0.03 cpu=0% mat=94 n=5421 fh=100% nps=10k > ext-> chk=48 cap=0 pp=0 1rep=0 mate=0 > predicted=0 nodes=5421 evals=1 > endgame tablebase-> probes done=0 successful=0 >PFGA: EPD record: 2 > time surplus 0.00 time limit 30.00 (3:30) > nss depth time score variation (1) > 1 0.00 Mate 1. Ne6# > 1-> 0.00 Mate 1. Ne6# > 2 0.00 Mate 1. Ne6# > 2-> 0.00 Mate 1. Ne6# > 3 0.00 Mate 1. Ne6# > 3-> 0.01 Mate 1. Ne6# > 4 0.01 Mate 1. Ne6# > 4-> 0.11 Mate 1. Ne6# > time=0.11 cpu=100% mat=37 n=81594 fh=100% nps=741k > ext-> chk=7342 cap=0 pp=0 1rep=985 mate=0 > predicted=0 nodes=81594 evals=10 > endgame tablebase-> probes done=0 successful=0 >PFGA: EPD record: 3 > time surplus 0.00 time limit 30.00 (3:30) > nss depth time score variation (1) > 1 0.02 37.98 1. dxc8=Q > 1 0.04 38.70 1. Nfxe3+ Kxh2 2. dxe8=Q > 1-> 0.05 38.70 1. Nfxe3+ Kxh2 2. dxe8=Q > 2 0.05 39.09 1. Nfxe3+ Kxf2 2. dxe8=Q > 2 0.05 ++ 1. dxc8=Q!! > 2 0.07 41.17 1. dxc8=Q Nxb6 2. fxe8=Q Bxh7 3. Nfxe3+ > 2 0.07 Mat02 1. Nf4+ Kxf2 2. Qxe3# > 2-> 0.09 Mat02 1. Nf4+ Kxf2 2. Qxe3# > 3 0.09 Mat02 1. Nf4+ Kxf2 2. Qxe3# > 3-> 0.14 Mat02 1. Nf4+ Kxf2 2. Qxe3# > 4 0.15 Mat02 1. Nf4+ Kxf2 2. Qxe3# > 4-> 0.45 Mat02 1. Nf4+ Kxf2 2. Qxe3# > time=0.45 cpu=100% mat=20 n=2507183 fh=99% nps=5571k > ext-> chk=12352 cap=279 pp=110 1rep=972 mate=0 > predicted=0 nodes=2507183 evals=9984 > endgame tablebase-> probes done=0 successful=0 >PFGA: EPD record: 4 > time surplus 0.00 time limit 30.00 (3:30) > nss depth time score variation (1) > 1 0.00 Mate 1. Rhxh2# > 1-> 0.00 Mate 1. Rhxh2# > 2 0.00 Mate 1. Rhxh2# > 2-> 0.01 Mate 1. Rhxh2# > 3 0.01 Mate 1. Rhxh2# > 3-> 0.04 Mate 1. Rhxh2# > 4 0.04 Mate 1. Rhxh2# > 4-> 0.51 Mate 1. Rhxh2# > time=0.51 cpu=98% mat=102 n=474587 fh=100% nps=930k > ext-> chk=24279 cap=53 pp=0 1rep=4452 mate=0 > predicted=0 nodes=474587 evals=1 > endgame tablebase-> probes done=0 successful=0 >PFGA: EPD record: 5 > time surplus 0.00 time limit 30.00 (3:30) > nss depth time score variation (1) > 1 3.10 4.32 1. Qxa8 Rxa8 2. Bxa1 exf2 > 1 3.12 Mate 1. Qxb6# > 1-> 3.13 Mate 1. Qxb6# > 2 3.13 Mate 1. Qxb6# > 2-> 3.14 Mate 1. Qxb6# > 3 3.14 Mate 1. Qxb6# > 3-> 3.19 Mate 1. Qxb6# > 4 3.19 Mate 1. Qxb6# > 4-> 3.39 Mate 1. Qxb6# > time=3.39 cpu=100% mat=0 n=108311323 fh=100% nps=31950k > ext-> chk=4476 cap=132 pp=7 1rep=468 mate=0 > predicted=0 nodes=108311323 evals=964783 > endgame tablebase-> probes done=0 successful=0 >PFGA: EPD record: 6 > time surplus 0.00 time limit 30.00 (3:30) > nss depth time score variation (1) > 1 0.00 Mate 1. exd8=Q# > 1-> 0.00 Mate 1. exd8=Q# > 2 0.00 Mate 1. exd8=Q# > 2-> 0.02 Mate 1. exd8=Q# > 3 0.02 Mate 1. exd8=Q# > 3-> 0.26 Mate 1. exd8=Q# > 4 0.26 Mate 1. exd8=Q# > 4-> 2.54 Mate 1. exd8=Q# > time=2.54 cpu=100% mat=28 n=1968998 fh=100% nps=775k > ext-> chk=139901 cap=214 pp=15066 1rep=22502 mate=0 > predicted=0 nodes=1968998 evals=1 > endgame tablebase-> probes done=0 successful=0 >White(1): The problem with such positions is that I use a simple q-search for each root move to order the moves. In these positions, the q-search blows up in a ridiculous fashion with the queens running around ripping stuff right and left. Crafty may not count the time spent in the RootMoves() procedure as part of the search time, since it is normally an instantaneous calculation. But when there are lots and lots of captures, it can take a long time, which is a problem of sorts, but it doesn't affect real games so I don't sweat it...
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.