Author: Tord Romstad
Date: 10:10:19 08/18/03
Go up one level in this thread
On August 18, 2003 at 12:45:25, Thorsten Czub wrote:
>On August 18, 2003 at 12:07:07, Tord Romstad wrote:
>
>>From black's point of view, I suppose? Slight advantage for
>>white?
>
>slight advantage for white it thought.
>It seems for the moves you need singular extensions :-))
Does Rebel use singular extensions? When I give Gothmog a few minutes
of thinking time per move, it is able to reproduce most of Rebel's move
in this attack (though it probably misses some important ones). I don't
use singular extensions, as they would slow down my program almost to
a halt (it is already too slow).
>>>try it out with YOUR program.
>>
>>OK. Gothmog is no better than Fritz here:
>>
>> 9 -0.31 188.73 26066387 c8d7 f2d4 b4d5 h5f7 g8f7 f5f6 g7g6 g5d5
>> d7e6 d5a5 f7g8 a1e1
>
>
>
>>At least it gives white a slightly larger plus than Fritz.
>
>
>right ! AND more important, it sees the right moves:
>
>20...Bd7 21.Bd4 Nxd5
>
>although 22.Rac1 Qb3
>with 23.Rc3 is of course a better idea !!
>But the programs maybe believe they could take the rook.
Gothmog likes 23. Rc3, too, although it needs some time before it sees
that white is winning:
1 -3.94 0.00 10 d4g7 f8g7 h5f7 g8f7
1 > -3.94 0.00 13 h5f7 g8f7
1 -1.71 0.00 131 h5f7 g8f7
1 -1.60 0.00 515 f5f6
1 -1.41 0.01 732 a4a5
2 -1.66 0.01 1318 a4a5 b7b5
2 -1.63 0.01 1665 h5f7 g8f7 a4a5
2 -1.55 0.02 1996 f5f6 g7g6
2 -1.52 0.03 2446 g5h4 b3d3
3 -1.45 0.05 3662 g5h4 b3d3 a4a5
4 -1.57 0.10 11679 g5h4 e8e7 d4c5 e7e5 h5f7 g8f7 h4h7 f8c5
c1c5 d7a4
4 -1.41 0.16 15009 d4g7 f8g7 f5f6 d5f6 f1f6
5 -1.41 0.76 55400 d4g7 f8g7 f5f6 d5f6 f1f6
6 -1.47 1.44 104157 d4g7 f8g7 f5f6 d5f6 f1f6 d7e6
6 > -1.47 3.60 254724 c1c3 d5c3 d4g7 f8g7
6 -0.97 4.27 303573 c1c3 b3a2 c3g3 f7f6 g5h6 g8h8
7 > -0.47 14.78 1074073 c1c3 b3a2 c3g3 f7f6 d4f6 d5f6 g5f6
7 -0.10 18.52 1350175 c1c3 d5c3 d4g7 h7h6 g7h6 g8h7 h6f8 e8f8
e2c3 d7a4 c3e4 b3b2
8 0.00 28.83 2075998 c1c3 d5c3 d4g7 h7h6 g7h6 g8h7 h5g6 f7g6
g5g6 h7h8 h6f8 e8f8 g6h6 h8g8 h6g6 g8h8
g6h6
9 0.23 54.52 3884591 c1c3 d5c3 d4g7 h7h6 g7h6 g8h7 h5g6 f7g6
f5g6 h7g8 g6g7 f8e7 f1f8 g8h7 g7g8Q b3g8
g5g8 h7h6
10 > 0.73 116.21 8508925 c1c3 d5c3 d4g7 h7h6 g7h6 g8h7 h5g6 f7g6
f5g6 h7g8 g6g7 f8e7 f1f8 e7f8 g7f8Q g8h7
e2c3
10 1.63 347.55 24661179 c1c3 d7a4 c3b3 a4b3 f5f6 g7g6 h5g4 b3c4
d4e5 b8d8
After that, it agrees with all of Rebel's moves until 27. Rg3. Gothmog
prefers 27. Rc7, with a +2 score for white (I am too lazy to check whether
this assessment is correct at the moment). After 27... Bc6, it wants
to play 28. Rxf8, and needs two minutes before it realizes that 28. Rf1
is much stronger.
>a good attack by Rebel !
Yes, a nice game!
>maybe this positions is too complicate for most programs ?!
I noticed that Gothmog was extremely slow in these positions. It
needs more than three minutes to complete a 9-ply search, which is
unusual.
I will check how the MTD(f) version of my program behaves later
tonight.
>How do other programs play ??
I don't have many other programs to test, I'm afraid. I'll have
to leave this task to someone else. :-)
Tord
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.