Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Something to grind on.

Author: Uri Blass

Date: 04:21:51 04/21/05

Go up one level in this thread


On April 21, 2005 at 03:57:14, Drexel,Michael wrote:

>On April 21, 2005 at 03:28:37, Drexel,Michael wrote:
>
>>On April 19, 2005 at 21:05:52, Dan Andersson wrote:
>>
>>>[D]5r2/ppp3k1/5r1p/7P/8/2B5/PPP5/5R1K w - - 0 1
>>> A simple position. Depth is paramount.
>>>
>>>MvH Dan Andersson
>>
>>IMO this is yet another questionable testposition.
>>the idea is good, the realisation bad:
>>
>>- there are many solutions
>>
>>- the engines evaluate the position as better for white anyway.
>>
>>- most engines are able to win this position if you play it out.
>>
>>Michael
>
>this is not exactly the same idea and the position is probably too hard for
>computers but it demonstrates my point.
>
>[D] 5rk1/p1p2r1p/1p3ppP/2p3N1/8/8/PPP3B1/5R1K b - - 0 1
>
>whites last move was 1.Nf3-g5.
>avoid 1...fxg5?? +- (a horrible mistake for humans) and find the correct move
>1...Rd7.
>
>Michael

Thanks for this position.

I do not believe that there are positions that are too hard for computers(I
assume no progress in hardware).
There are only positions that are too hard for the programs of today.

Give me a lot of positions that computers cannot solve and if I continue to work
on Movei then I hope that you will find some day that Movei on the hardware of
today have no problem to solve many of them with improvement in playing strength
thanks to the relevant code.

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.