Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: watch this:

Author: Uri Blass

Date: 02:48:54 11/28/01

Go up one level in this thread


On November 27, 2001 at 20:04:44, Mike S. wrote:

>On November 27, 2001 at 19:56:32, ALI MIRAFZALI wrote:
>
>>On November 27, 2001 at 19:51:03, Mike S. wrote:
>>
>>>(...)
>
>>>New game - Fritz 7
>>>r1b2rk1/pppp1ppp/2n5/3B4/2Q5/4q2N/PPP4P/1K1R1R2 b - - 0 1
>>>
>>>Analysis by Fritz 7:
>>>
>>>17...Qxh3 18.Bxf7+
>>>  -+  (-2.60)   Depth: 7/25   00:00:00  136kN
>>>  ±  (1.25)   Depth: 12/40   00:02:57  64690kN
>
>>Are you saying that the first version  sticks with 17...Qxh3??? after 5 or
>>10 seconds? Please explain
>
>The lines above are copied & pasted from the engine output of the first Fritz 7
>engine version. Qxh3 is chosen from the beginning (00:00:00) and kept for
>several minutes; it switches to another move after some more minutes. 7a doesn't
>consider Qxh3 as pm.
>
>I cannot explain more. Probably an engine bug which was fixed.
>
>Regards,
>M.Scheidl

I can explain
If you use only material evaluation you can find that after Qxh3 white has a
queen for rook knight and pawn.

You also can find that without Qxh3 black is two pawns up but all the white
pieces are prepared against the black king so you can evaluate the king attack
of white as more than 2 pawns and prefer Qxh3 if you believe that after
sacrificing the queen for a rook and a knight white has not enough material for
attack.

It is simply a question of evaluating positional factors and I guess that the
first Fritz7 was wrong in that case.

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.