Author: leonid
Date: 04:51:57 04/29/01
Go up one level in this thread
On April 29, 2001 at 00:56:45, Heiner Marxen wrote: >On April 27, 2001 at 21:58:22, leonid wrote: > >>On April 27, 2001 at 19:28:41, leonid wrote: >> >>>On April 27, 2001 at 18:44:32, Mike S. wrote: >>> >>>>On April 27, 2001 at 08:28:44, leonid wrote: >>>> >>>>>(...) >>>>>[D]4nqqQ/1r2qR1Q/2q1NBQN/q3n1PK/kr1qn1RQ/4RBq1/5NpN/3bb2R b - - >>>> >>>>Black has the move? >>> >>>Yes. >>> >>>>Please tell at least how many moves the mate takes, to allow an estimation of >>>>the calculation time required. >>> >>>I really don't know exact minimum amount of moves for this position and expect >>>that somebody will come with response. I had too bad branching factor when I >>>tried to solve it in perfect way. As a result, I saw this position by brute >>>force only 6 moves deep (already 23 min). In 6 moves for sure no mate. And in 10 >>>- mate was found by my selective search. Shortest mate exist without any doubt >>>between 7 and 10 moves. > >Chest on a K7/600 with 350 MB hash does not find a mate in 8 in nearly 3 hours. >Since the effective branching factor is around 20, I give up, here. You have good branching factor here. Mine had (no hash) 36 between 4 and 5 moves and as much as 58 between 5 and 6 moves. Search done by brute force. 4 moves - 1 sec. 5 moves - 35 sec. 6 moves - 33 min. 58 sec. By curiosity I looked into Genius 2 work. It was: 3 moves - 0 sec 4 moves - 33 sec. 5 moves - 39 min 45 sec. It could be that it use small hash. Cheers, Leonid. >For the "no mate in 6" Chest used only 34 seconds. > >Cheers, Heiner > > >>Will just add, maybe this could help, that mate that my solver found started >>with move g3-h4. >> >> >>>Cheers, >>>Leonid. >>> >>>>With white to move, I didn't find a #5, using Mate 2.0 with standard options. >>>> >>>>Regards, >>>>M.Scheidl
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.