Author: Jeremiah Penery
Date: 12:09:16 10/20/01
Go up one level in this thread
On October 20, 2001 at 08:57:03, Slater Wold wrote: >On October 20, 2001 at 05:10:13, Simon Finn wrote: > >>On October 18, 2001 at 23:34:45, Slater Wold wrote: >> >>>On October 18, 2001 at 22:55:02, Dann Corbit wrote: >>> >>>>On October 18, 2001 at 21:01:46, Slater Wold wrote: >>>> >>>>>On October 18, 2001 at 09:14:43, José Antônio Fabiano Mendes wrote: >>>>> >>>>>> L.D.Evans-Haik,London 1978 [unusual and difficult position] >>>>>> [D]8/p2P3P/k7/8/8/2q5/6PK/1Q6 b >>>>>> "Surprisingly Black can deliver perpetual check or force stalemate." >> <SNIP> >>>>> Qa2+ 13. Kd6 Qd2+ 14. Kc7 Qc2+ <HT> >>>>> 16 46.88 -- 1. ... Qe5+ >>>>> time=60:01 cpu=201% mat=2 n=1693306751 fh=99% nps=470k >>>>> ext-> chk=1363197943 cap=274706 pp=267447 1rep=117221890 mate=2380 >>>>> >>>>> predicted=0 nodes=1693306751 evals=7323296 >>>>> endgame tablebase-> probes done=48149 successful=4738 >>>>> hashing-> trans/ref=73% pawn=99% used=14% >>>>> SMP-> split=4869 stop=472 data=13/32 cpu=120:40 elap=60:01 >>>>>Black(1): >>>>> >>>>> >>>>>Fails low @ 46 seconds, and can't resolve it in 59 minutes. :( >>>> >>>>This looks like a bug. That position should never fail low, but always fail >>>>high. >> >>Why? The position is ultimately drawn, but that doesn't mean there won't be some >>nasty shocks on the way. >> >>>> >>>>Either that, or crafty may have found some strange, suicidal attack that >>>>succeeds. >>> >>>If I had to guess, I'd say it'll find something. I did it with 10 minutes at >>>first, and it did this same thing. 59 minutes to resolve a fail low _does_ seem >>>a tad long. Perhaps Bob can help here. >> >>I had a long look at this position yesterday. >> >>White will win if he can get his king to a1 - this allows him to >>interpose the queen and eventually kill the checks. >> >>To prevent this, Black has to answer Ka2 with Qa4+/Qa5+ and Kb2 with Qd4+/e5+. >> >>This requirement restricts where Black can check when the White king is on other >>squares too. >> >>I think (but I'm not 100% sure) that answering Ke3 with Qg3+ loses >>because the White king can eventually get to a1. >> >>Crafty may just have discovered this and needs to spend a >>lot of time finding an alternative system of defence that works. >>(Black should arrange things to answer Ke3 with Qc5+, for example.) >> >>Simon >> >>> >>>I can give it more time, if requested. > >That's a very good explaination. However, after about 54 billion nodes, I would >expect it to have exhausted all of these possibilities. :) I think Crafty has some search flaws that cause this kind of behavior. I've seen cases like this (I'm just making up these moves because I don't have an actual position handy): 1) The move Ne5 looks good in the first couple plies, but gets white mated in 2 moves. 2) Crafty sees this very fast, and finds a better root move, Nh2, which prevents the mate. 3) Nh2 fails low in the next iteration, but isn't getting mated. 4) Crafty is searching other moves at that ply to find a "better" root move. When it gets to Ne5, it practically hangs. Even though it has already discovered it was Mated in 2 after Ne5 and it took practically zero time to find it before, this move is searched for a _long_ time at the root, for no apparent reason.
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.