Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Crafty 15.18 Vs Fritz 5 Match Game 5

Author: Howard Exner

Date: 17:37:36 08/02/98

Go up one level in this thread


On August 02, 1998 at 12:53:21, Mark Young wrote:

>On August 02, 1998 at 11:43:05, Howard Exner wrote:
>
>>On August 01, 1998 at 21:10:27, Mark Young wrote:
>>
>>>On August 01, 1998 at 10:38:35, Howard Exner wrote:
>>>
>>>>On July 31, 1998 at 15:06:30, Mark Young wrote:
>>>>
>>>>>Results of games 1 thur 5
>>>>>
>>>>>Crafty 0 0 0 0 0
>>>>>
>>>>>Fritz5 1 1 1 1 1
>>>>>
>>>>>This game is a good example of Fritz 5 ability to attack queenside, for anyone
>>>>>who has not seen it before. Capped off with a nice long-term pawn sac to cash in
>>>>>its positional plus.

>>>>At this point of the game instead of Ne1 the move gxf4 suggests itself.
>>>>This way the passed h pawn does not become a reality for black.
>>>>
>>>It does seem there should be a way to draw this, but position is like a Chinese
>>>finger puzzle. There might be some subtle defense that I am just over looking,
>>>but I was unable to save the game for white. I let fritz have a shot at it and
>>>here’s what it came up with.
>>>
>>>67. gxf4 exf4 68. h3 b2 69. Ne1 g4 70. hxg4 hxg4 71. Rb1 gxf3+ 72. Kxf3 Na4 73.
>>>Kg4 Rd7 74. Nf3 Rg7+ 75. Kxf4 Rg2 76. Rh1 Nc3
>>
>>Since this is a solid winning line for black also, it makes the entire sequence
>>of moves starting with 65. ... a3 all that more attractive. In the analysis
>>you give 67. ... exf4 as the proper recapture (not gxf4). This recapture
>>would be a routine matter for humans as they would see, without much
>>calculating,
>>that exf4 gives a kingside pawn majority. Computers must calculate instead
>>of "seeing" that common pattern. I find this an instructive ending.
>>
>> It's interesting how calculating moves and positional knowledge are
>>intertwined. Computers often resolve or mimic positional knowledge
>>by simply going deep in their calculating. We on the other hand rely
>>on positional understanding to assist us in "seeing" the course the game
>>may take.
>>
>>>
>I have come to the same conclusion, but think of it in a different way. I think
>its us human who mimic long term tactical calculations with are unique ability
>to understand positional play.

I agree with that. I intended the interpretation you have given on how we
"see" (not see clearly but intuitively). Sometimes I use the analogy that
knowledge is like a road sign during a country car drive in the fog (or
during a snow storm if you live in those climates). The sign provides the
driver with a clue as to what is coming ahead - we know the curve is
ahead but we can't see it.



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.