Author: Christophe Theron
Date: 10:20:18 10/09/99
Go up one level in this thread
On October 09, 1999 at 00:12:56, Dann Corbit wrote:
>r2r2k1/pQBb3p/6p1/3Pb2q/8/4pPP1/PP4BP/R2R3K w - -
>
>Zarkov 4.0 32 bit was stuck on Bxd8 for about 2200 seconds {around 62 million
>nodes}, then made this choice:
>22.Bxe5 Qxe5 23.f4 Qb8 24.Qb3 Qxb3 25.axb3 e2 26.Rd2 Bb5 27.Bf3 Re8 28.Re1
>Rad8 29.Bxe2 Bxe2 30.Rdxe2 Rxe2 31.Rxe2 Rxd5
>with an eval of about +2.00.
>
>About 10 minutes into the analysis, it could see that Bxd8 was bad
>(eval -0.23) but it took a very long time to find a suitable alternative.
>
>Crafty seems to avoid Bxd8 right away. How do other engines do? Perhaps it is
>some sort of a bug.
>
>Also, I only seem to get about 28KNPS for Zarkov 32 bit. Is it a slow searcher,
>or do I have my machine configured improperly? (It's only a 300MHz PII on the
>machine w/ Zarkov 4).
Tiger 12 running on K6-2 300MHz, 16Mb hash avoids the trap in 121.65 seconds and
plays Bxe5.
Please note that in a real game, due to its time management and the fact that
the move Bxd8 has been identified as being not so good in the previous
iteration, Tiger would avoid the trap in a game in 45 minutes (G/45) on this
computer. On a 450MHz computer it would avoid the trap in G/30.
Christophe
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.