Computer Chess Club Archives


Search

Terms

Messages

Subject: Cool Positions, LambChop analysis

Author: Peter McKenzie

Date: 01:41:36 01/31/00

Go up one level in this thread


On January 31, 2000 at 00:42:31, Jeremiah Penery wrote:

>Ferret vs. Junior:
>[D]r2q1rk1/1p3ppp/1n1p4/1PbB4/p2pPn2/2P2N1P/1BQ2PP1/4RRK1 w - -
>
>Ferret played Bxb7 here, but I think it was a mistake.  Better I think is cxd4,
>as it doesn't allow black to penetrate.

This is a tricky looking position, glad I didn't have either side against one of
those tactical monsters!  Here is LambChop's take on it (all analysis on my
P133):

 0    0     0       0 d5b7
 1    2     0      54 d5b7 d4c3 b2c3
 2  -17     0     256 c3d4 b6d5 d4c5
 2  -17     0     464 c3d4 b6d5 d4c5
 3  -17   100    1643 c3d4 b6d5 d4c5
 4   14   100    4959 c3d4 b6d5 d4c5 d5f6
 5   12   200   14459 c3d4 b6d5 d4c5 d5b4 c2c4 d6c5 c4c5
 6    9   500   46277 c3d4 c5b4 d5b7 a8a7 b7c6 b4e1 f1e1
 7    9  1300  126205 c3d4 c5b4 d5b7 a8a7 b7c6 b4e1 f1e1
 8   -4  4500  465815 c3d4 c5b4 d5b7 a8a7 b7c6 b4e1 f1e1 a4a3 b2c1
 9   10  9900 1049760 c3d4 c5b4 d5b7 a8a7 b7c6 b4e1 f1e1 f7f6 d4d5
10   13 41900 4453523 c3d4 c5b4 d5b7 a8a7 b7c6 b4e1 f1e1 a4a3 b2c1 d8f6 c1f4 f6
f4

So it didn't even consider Bxb7.  I tried the position after Bxb7 to see what it
thought:

 0    0     0       0 d4c3
 1   67     0     162 d4d3 c2d2
 1   67     0     198 d4d3 c2d2
 2   12     0     972 d4d3 c2d2 f4e2 e1e2 d3e2 d2e2
 3    2     0    3268 d4d3 c2d2 a8a7 d2f4 a7b7
 4    2   100    9174 d4d3 c2d2 a8a7 d2f4 a7b7
 5   -9   400   34156 d4d3 c2d2 a8a7 d2f4 a7b7 f3d4 c5d4 c3d4
 6    5   900   89382 d4d3 c2d2 a8a7 d2f4 a7b7 f3d4 a4a3 b2c1B c5d4 c3d4 a3a2
 7   28  2800  287762 d4d3 c2d2 a8a7 d2f4 a7b7 b2c1 b6c4 e4e5
 8   23  6600  700787 d4d3 c2d2 a8a7 d2f4 a7b7 f4f5 g7g6 f5g4 b6c4
 9   36 24700 2612524 d4d3 c2d2 a8a7 d2f4 a7b7 f3d2 a4a3 b2a1 d6d5 f4g3 d5e4 e1
e4 a3a2
10   32 100000 10864212 d4d3 c2d2 a8a7 d2f4 a7b7 f3d2 d6d5 e4d5 a4a3 b2a1 g7g5
f4f3 d8d5

The decision between the two moves is definitely a positional one based on eval.
 LambChop understands something about the activity of the bishop on c5, and the
lack of activity of the bishop on b2 which probably leads to it prefering cxd4.
I honestly don't know if cxd4 is really better, but I'm happy enough that Chop
likes it.

>
>Crafty vs. LGG
>[D]r6k/P1rb1npp/5q2/3R1p2/1p1Q3P/4P1P1/5P2/R1B2BK1 w - -
>
>Rxd7 looks good here, but Crafty failed (very) low on it and played Qf4.  I
>tried to analyze a bit after Rxd7 Rxd7 Qxd7 Qxa1 Qxf7 Qxc1 Qxf5, and white still
>seemed to be ahead.  Can Rxd7 win, or is there really a problem with it?

I don't think Rxd7 is very good: after Rxd7 Rxd7 Qxd7 Qxa1 Qxf7 Qxc1 Qxf5 Qa1
black wins the a7 pawn and should hold the draw.  White wins black's outside
passer but then its QB+4P vs QR+2P, all pawns on same side which looks drawish
at best for white.

Here is Chop's analysis:

 0    0     0       0 d4f6
 1  248     0      66 d4f6 g7f6
 2  243     0     383 d4f6 g7f6 c1b2 a8a7 b2f6 h8g8
 3  243     0    1038 d4f6 g7f6 c1b2 a8a7 b2f6 h8g8
 4  244     0    3611 d4f6 g7f6 c1b2 a8a7 b2f6 h8g8 a1d1
 5  246   100   12181 d4f6 g7f6 c1b2 a8a7 b2f6 h8g8 a1d1 b4b3 f1e2
 6  262   400   34011 d4f6 g7f6 d5a5 f7d6 c1b2 d6e4 b2d4
 7  282   900  108324 d4f6 g7f6 d5a5 f7d6 c1b2 h8g7 a1d1 d6e4
 8  272  3200  387017 d4f6 g7f6 d5a5 f7d6 c1b2 d6e4 f1d3 c7c6 b2d4
 9  283  7200  896514 d4f6 g7f6 d5a5 f7d6 c1b2 h8g7 h4h5 d6e4 f1h3 e4c5
10  358 32600 4135614 c1b2
10  368 60500 7412987 c1b2 f6d4 b2d4 b4b3 a1b1 a8a7 d4a7 c7a7 b1b3 g7g6 f1e2 a7c
7

So Rxd7 isn't considered at all, and Qf4 isn't played either.  Bb2 looks strong
though.  Here is analysis after Rxd7:

After Rxd7:
 0    0     0       0 f6d4
 1 -129     0      26 c7d7 d4f6 g7f6
 1 -129     0      87 c7d7 d4f6 g7f6
 2 -129     0     502 c7d7 d4f6 g7f6
 3  -64     0     998 c7d7 d4d7 f6a1 d7f7 a1c1
 4  -64   100    1820 c7d7 d4d7 f6a1 d7f7 a1c1
 5  -64   100    4891 c7d7 d4d7 f6a1 d7f7 a1c1
 6 -231   300   19500 c7d7 d4d7 f6a1 d7f7 a1c1 f7f5
 7 -185   500   47574 c7d7 d4d7 f6a1 d7f7 a1c1 f7f5 b4b3 f5d5
 8 -136  1000  110125 c7d7 d4d7 f6a1 d7f7 a1c1 f7f5 c1a1 f5f4 a1a7 f4b4
 9 -136  1900  242753 c7d7 d4d7 f6a1 d7f7 a1c1 f7f5 c1a1 f5f4 a1a7 f4b4
10 -266  6100  778174 c7d7 d4d7 f6a1 d7f7 a1c1 f7b7 c1c8 b7c8Q a8c8 f1g2 b4b3 a7
a8Q c8a8 g2a8 b3b2
11 -266 11300 1503733 c7d7 d4d7 f6a1 d7f7 a1c1 f7b7 c1c8
12 -155 32200 4465228 c7d7 d4d7 f6a1 d7f7 a1c1 f7f5 c1a1 f5e4 a1a7 f1d3 a7a1 g1h
2 g7g6 e4b4

So score is down to white by 1.5 pawns, but even this evaluation is too
optimistic for white I think.

>
>Nimzo vs. Crafty
>[D]8/8/1pp5/3k1p1p/Pp1p1P2/1P4PP/2P2K2/8 w - -
>
>Nimzo played Kf3 here and ended up losing.  Ke2 might draw here.  Does it?

My gut says that white will loose not matter what, black just seems to be boss.
Here is Chop's output:

 0    0     0       0 a4a5
 1  -24     0      19 h3h4
 1    7     0      20 f2f3
 1    7     0      25 f2f3
 2   29     0     150 f2f3 c6c5
 3    7     0     585 f2f3 c6c5 h3h4
 4   14     0    1625 f2f3 d5d6 h3h4 d6d5
 5   -3   100    5384 f2f3 c6c5 g3g4 h5g4 h3g4 f5g4 f3g4 d5e4 f4f5
 6    4   100   13470 f2f3 c6c5 f3f2 d5e4 f2e2 h5h4
 7    6   200   43071 f2f3 d5e6 g3g4 h5g4 h3g4 f5g4 f3g4 e6f6 g4g3 f6f5 g3f3
 8   10   500  118246 f2e1 d5e4 e1e2 h5h4 g3h4 e4f4 h4h5
 8   10   500  118246 f2e1 d5e4 e1e2 h5h4 g3h4 e4f4 h4h5
 9   15   900  239358 f2f3 d5d6 g3g4 h5g4 h3g4 f5g4 f3g4 c6c5 g4f5 c5c4 b3c4 b4b
3 c4c5 b6c5
 9   15  1200  330698 f2f3 d5d6 g3g4 h5g4 h3g4 f5g4 f3g4 c6c5 g4f5 c5c4 b3c4 b4b
3 c4c5 b6c5
10  -37  3500 1039859 f2f3 c6c5 f3e2 c5c4 b3c4 d5c4 e2d2 d4d3 c2d3 c4b3 a4a5 b6a
5 d2e3
11    4  7700 2233293 f2f3 c6c5 f3e2 c5c4 b3c4 d5c4 e2d2 d4d3 c2d3 c4b3 d3d4 b3c
4 d2c2 c4d4
12  -12 23700 6270266 f2f3 c6c5 f3e2 c5c4 e2d2 c4b3 c2b3 d5e4 d2e2 h5h4 g3h4 e4f
4 h4h5
13   -4 63800 14659241 f2e1 c6c5 e1d1 d5e4 d1e2 h5h4 g3h4 e4f4 e2d3 f4e5 h4h5

So it became less happy with Kf3, and finally switched to Ke1.  I'm sure some
other programs can get much deeper and closer to the truth here.

>
>What do other programs think of all these.

I would be interested in what other programs think too.



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.