Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Strength question

Author: Uri Blass

Date: 14:17:13 01/06/03

Go up one level in this thread


On January 06, 2003 at 15:43:18, Steve Maughan wrote:

>Uri,
>
>>strong in tactics?
>>
>>I do not agree.
>
>Hmmmmmmmmmm, I think you're being argumenatative for the sake of it.  Little
>Goliath is not perfect but if there is one thing is is good at it's tactics.

No

I believe that even shredder6 is better than goliath in tactics.

test positions are misleading.

Tactics is not only sacrifices.
I remember a game from the ssdf games when  Goliath was simply outsearched by
shredder based on my analysis.

It chose a move that is losing material in the middle game and my shredder5.32
could see the tactics earlier.
>
>All of the following positions are solved quickly by LG and I would certainly
>say that this *does* indicate tactical strength.  If test positions don't games
>certainly don't e.g. Shredder 4 is weak at tactics but nevertheless strong
>
>Times for a 1.5 GHz P4 - I'd be interested in Movei's times.  For it to be "not
>quite as good" I'd expect it to solve all of these within 30 secs each

I suggest that you download movei

Public Movei can solve the first 2 in less than 30 seconds.
The second one is easy for movei.


>
>[d]3r1b2/p4bkp/1p1P1p2/r3p1p1/2q1N3/2N2P2/1P1R2PP/2QR3K w - - 0 1
>
>LG 2 secs (Nxf6)

Public Movei p850 needs 26.86 seconds to fail high

9 -57 2686 3451622 e4f6
9 -28 2845 3660501 e4f6
9 31 4723 6022109 e4f6 f7g6 b2b4 c4b4 c3e4 d8a8 c1c7 g7h8 d6d7 g6e4 f3e4
9 31 4756 6065598 e4f6 f7g6 b2b4 c4b4 c3e4 d8a8 c1c7 g7h8 d6d7 g6e4 f3e4


>
>[d]r1b2r2/3pNpkp/3pn1p1/2pN3P/2PnP3/q3QP2/4BKP1/1R5R w - - 0 1
>
>LG 3 secs (Qh6)

Public movei on 850 mhz.

7 57 511 702228 e3h6
7 86 544 740950 e3h6
7 9979 1395 1739365 e3h6 g7h6 h5g6 h6g5 h1h5 g5h5 f3f4 a3f3 e2f3 d4f3 b1h1 f3h4
d5f6 h5h6 h1h4 h6g7 f6e8 f8e8 h4h7 g7f6
7 9979 1401 1742953 e3h6 g7h6 h5g6 h6g5 h1h5 g5h5 f3f4 a3f3 e2f3 d4f3 b1h1 f3h4
d5f6 h5h6 h1h4 h6g7 f6e8 f8e8 h4h7 g7f6

>
>[d]r5r1/8/bRP1pk1p/3pNp2/5P2/7P/PR4P1/6K1 w - - 0 1
>
>LG 8 secs (Rxa6)

This one is more difficult and movei needs some minutes but I believe that this
problem is going to be solved in the future after I improve my extensions.

Note that I do not claim that movei is better than crafty in tactics inspite of
the fact that it is often better than it in test positions.

I use test positions to test my program to help me to test search changes but I
do not claim to be able to compare the tactical strength of movei with the
tactical strength of other programs based on tactical tests because they are
misleading and the main problem is that test positions include no tactics with
no sacrifices and in chess there are often cases when you win material with no
sacrifice and it is important to see it.

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.