Author: Christophe Theron
Date: 22:01:51 05/19/98
Go up one level in this thread
On May 19, 1998 at 21:19:28, Thorsten Czub wrote:
>On May 19, 1998 at 15:01:12, Moritz Berger wrote:
>>
>>One addendum: I also wrote that Genius 5 / DOS (32 MB HT) on P233MMX
>>didn't find any of these moves within 3.5 hours (still wanting to
>>capture on a8 ...).
>
>I guess almost any program with or even without any king-safety
>evaluation WOULD find this move (b6 or g4) in 20 minutes. Depending on
>how fast and state of the art the search is. So what do you measure ?
>King safety ? Search depth ?
That's the point.
The position is interesting because a knowledged program could "sniff"
the solution way before a deep searcher.
>If genius is not finding it, it would only show that genius search tree
>works asymetrically. Take back one move and it should easily find b6/g4
>- or ?
Why don't you try and tell us if it works? Facts, please!
BTW here are the results on this position for Chess Tiger 11.4 (32Mb
hash):
On K5-100MHz:
Finds b6 in 11 min 23 sec (ply 9).
Then finds g4 in 24 min 04 sec (ply 9).
Tiger is exactly twice as fast on a K6-200 (Thorsten can witness), so
here are the results on K6-200 (close to PII-233 I suppose):
Would find b6 in 5 min 42 sec (ply 9).
Would find g4 in 12 min 02 sec.
Christophe
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.