Author: Dieter Buerssner
Date: 10:14:54 09/01/02
Go up one level in this thread
On September 01, 2002 at 03:01:07, Manfred Rosenboom wrote:
Hi Manfred. It is nice, to see you trying Yace in several positions. In this
one, I think you made a small error:
>Yace UCI 134 MB:
>
>Position stored with depth 62
Not the above line. It means, that Yace already has the solution stored
internally. Most probably, you analyzed this position already, before creating
the output you posted.
To avoid this with Yace-UCI, clear hash tables in engine-setup, and everything
remembered from previous analysis will be forgotten. It has also the advantage,
that you will see the full PV. Note, that Yace under UCI will never clear HTs or
other "learned" info by itself - Yace as WB engines would do this, for each new
position.
I get the following with 20M hash on K6-2 475 with about 60% CPU time (in text
mode):
10460 0.175 8.79 9t 1. Kd8 Ka7 2. Kxc7 Ka6 3. b8=Q+ Ka5 4. Kd7 Ka4
5. c7 Ka3 6. Qb5 {661}
11926 0.194 8.79 9. 1. Kd8 Ka7 2. Kxc7 Ka6 3. b8=Q+ Ka5 4. Kd7 Ka4
5. c7 Ka3 6. Qb5 {661}
12541 0.226 9.21 10++ 1. Kd8 Ka7 2. Kxc7 Ka6 3. b8=Q+ Ka5 4. Kd7 Ka4
5. c7 Rg8
15314 0.261 10.26 10++ 1. Kd8 Rxh7 2. gxh7 Ka7
16284 0.295 15.29 10++ 1. Kd8 Rxh7 2. gxh7 Ka7
66873 0.862 Mat08 10t 1. Kd8 Rxh7 2. gxh7 g5 3. h8=Q Bg7 4. Qxg7H
Ka7H 5. Qe5H g4H 6. Qa5+H Kb8H 7. Qxc7+H Ka7H
8. b8=Q#H {2352}
103401 1.155 Mat08 10. 1. Kd8 Rxh7 2. gxh7 g5 3. h8=Q Bg7 4. Qxg7 Ka7
5. Qe5 g4 6. Qa5+ Kb8 7. Qxc7+ Ka7 8. b8=Q#
{2352}
214347 1.615 Mat07 11t 1. Kd8 Rxh7 2. gxh7 Ka7 3. h8=Q Kb6 4. Qh5H
Kxc6H 5. b8=QH g5H 6. Qxg5H Bg7H 7. Qxc7#H
{2002}
259886 1.813 Mat07 11. 1. Kd8 Rxh7 2. gxh7 Ka7 3. h8=Q Kb6 4. Qh5 Kxc6
5. b8=Q g5 6. Qxg5 Bg7 7. Qxc7# {2002}
and score/PV does not change anymore
Regards,
Dieter
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.