Author: Uri Blass
Date: 00:33:27 03/05/01
Go up one level in this thread
On March 05, 2001 at 03:27:05, Uri Blass wrote:
>On March 04, 2001 at 21:49:58, Steven J. Brann wrote:
>
>>On March 04, 2001 at 12:49:05, Heiner Marxen wrote:
>>
>>>On March 04, 2001 at 08:24:38, Steven J. Brann wrote:
>>>
>>>>[D]2k1rbr1/1pp5/p3P1qp/1Q4p1/3P1p2/6N1/5PPP/R1B1R1K1 w - - 0 1
>>>>
>>>>This game was played by correspondence in 1875. Mrs. Berry announced mate in 18
>>>>from this position. Can your computer do the same? I doubt it. CM8000 found a
>>>>winning line in 18 moves (different than Mrs. Berry's), but wouldn't announce
>>>>mate until 6 moves into the solution. I'll post Ms. Berry's play and CM8000's
>>>>line after you have a go at it.
>>>
>>>This one appears to be out of reach for Chest. On my K7/600 (350 MB hash)
>>>it needs already twenty minutes for "no mate in 8". The effective branching
>>>factor is around 8. Provided, there is not a much shorter mate: No chance!
>>>
>>>Now I'm waiting for the solution, to study it carefully!
>>>
>>>Heiner
>>
>>Here's the original mate and the CM8000 equivalent. Enjoy
>
>I tried to go forward and backward with yace under chessbase to check this
>solution:
>>
>>{Mrs. Berry announces Mate in 18 from this position} 1. Rxa6 bxa6 2.
>>Qxa6+ Kb8 3. Qb5+ Kc8 4. Qd7+ Kb8 5. Bxf4 gxf4 6. Qb5+ Kc8 7. Ra1 Ba3
>
>Yace is calculating 7...Qb1+ after learning that 7...Ba3 is a mate against
>black.
I can add that the proof of Mrs. Berry is not convincing.
There was no chess playing programs at that time so she should give a full tree
before saying mate in 18.
Uri
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.