Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: One average mate position...

Author: leonid

Date: 17:55:37 01/23/01

Go up one level in this thread


On January 23, 2001 at 20:07:05, Paul wrote:

>On January 23, 2001 at 19:35:20, leonid wrote:
>
>>On January 23, 2001 at 17:27:20, Paul wrote:
>>
>>>On January 23, 2001 at 17:25:04, leonid wrote:
>>>
>>>>On January 23, 2001 at 17:10:39, leonid wrote:
>>>>
>>>>>On January 23, 2001 at 16:09:05, Paul wrote:
>>>>>
>>>>>>On January 23, 2001 at 15:48:55, leonid wrote:
>>>>>>
>>>>>>>Went to see exact time on 80386. Cyrix 33 Mhz. Time 0.22 sec.
>>>>>>
>>>>>>Hmmm ... not bad! And 'frute force'? Is much too deep probably.
>>>>>>
>>>>>>Paul
>>>>>
>>>>>You make me laugh making me remember that funny mistake.
>>>>>
>>>>>Brute force should take for sure much longer time. 12 moves is impressively deep
>>>>>position.
>>>>>
>>>>>Leonid.
>>>>
>>>>Brute force. AMD 400 Mhz. 3 min. 27 sec.
>>>>
>>>>Leonid.
>>>
>>>Yes!!!!! I knew it :)
>>>
>>>Paul
>>
>>And how long it take on your program by brute force, if you are ready to try.
>>Please indicate CPU and hash to give me some general idea.
>>
>>Leonid.
>
>Ok, I've not let it run it completely through till the mate ... but it will
>take approximately 25 minutes on my trusty old PII 400MHz. That's a pure
>alpha beta search to 24 ply, no extensions, no nullmove, no mate-find
>speeding tricks, no nothing :) With just nullmove switched on, that drops
>to 4'30" until WM12 announcement. So these extensions are really worth a
>whole lot! Coz with extensions on it only takes <3700 nodes, no time at all.
>
>Paul

Hi, Paul! I just printed your positions and tried one. Probably you know what.
The most strange one with one line of white pawns. I like this kind of fun.
Simply could not resist this temptation.

If you was able to find above position with usual chess program, your code is
more performant that mine. All search for mate my program do with special mate
solver. I doubt very much that my chess program will succeed in looking through
12 moves in 24 minutes. Now it actually can't see  beyond 6 moves limit. I
expect to extend my solver and chess program horizon when they will be totally
rewritten.

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.