Have you decided why Vas

Discussion of anything and everything relating to chess playing software and machines.

Moderator: Ras

Dann Corbit
Posts: 12777
Joined: Wed Mar 08, 2006 8:57 pm
Location: Redmond, WA USA

Re: Have you decided why Vas

Post by Dann Corbit »

Dr.Ex wrote:
If you win more games than the opponent then it is clear that you do something smart that the opponents do not do.

Calling engines that do something smart that the opponents do not do in the name stupid does not make sense.

Uri[/quote]

Just the fact that a program has evaluation terms which work on average better than that of the competition does not make a program "smart".
It is still a stupid beancounter.[/quote]

As Christophe Theron has often pointed out, search is also knowledge. If the information about what move to play comes simply from search and not from evaluation (which I do not believe to be the case, but I am only speculating) even so, Rybka plays on average excellent chess.

Only blitz, but this is an eye-opener...
From:
http://www.husvankempen.de/nunn/40_4_Ra ... liste.html
We have this:

Code: Select all

no Program Elo + - Games Score Av.Op. Draws 
1 Rybka 3.0 x64 4CPU 3265 20 20 1200 82.0% 3001 24.6% 
10 Naum 3.1 x64 4CPU 3031 12 12 1900 46.6% 3055 38.4% 
12 Zappa Mexico II x64 4CPU 3025 12 12 1850 45.9% 3054 38.0%
Which simply means that Rybka is over 200 Elo stronger than the next strongest computer chess program. I guess that the pattern will hold for longer time control (time will tell, of course).

There are some very smart chess people working on Rybka (pretty much everyone on the team is an IM or so). I guess that the chess knowledge in Rybka is above the chess knowlege in every other program on earth. Whether the information comes form search or evaluation does not matter. What matters is that the information is correct. Now, I am sure that there are some positions that will simply fool Rybka -- perhaps even make Rybka look silly, despite a very long search. But there will be a much larger percentage of positions which will fool other programs and make them look even more commical.

IMO-YMMV.
Top
Dann Corbit
Posts: 12777
Joined: Wed Mar 08, 2006 8:57 pm
Location: Redmond, WA USA

Re: Have you decided why Vas

Post by Dann Corbit »

Uri Blass wrote:
Dr.Ex wrote:[d] 2kr3r/1ppqb1pp/p1n1pp1n/3pPb2/BP1P1P2/2P1BN2/P5PP/RN1Q1RK1 b - b3 0 11

How long does it like to play Kb8 in this position? I mean, come on. Isn't it stupid to walk directly into an attack?
Here is some analysis
Rybka on slow hardware admits after enough time that white is better here

Mischa - Kapaun, Wertungspartie, 3m + 0s 2008
2kr3r/1ppqb1pp/p1n1pp1n/3pPb2/BP1P1P2/2P1BN2/P5PP/RN1Q1RK1 b - b3 0 1

Analysis by Rybka 3 1-cpu 32-bit :

11...Nh6-g4
³ (-0.52) Depth: 2 00:00:00
11...Nh6-g4
³ (-0.55) Depth: 3 00:00:00
11...Nh6-g4
³ (-0.57) Depth: 4 00:00:00
11...Kc8-b8 12.h2-h3
³ (-0.38) Depth: 5 00:00:00
11...Kc8-b8 12.Qd1-d2 Rh8-f8
³ (-0.50) Depth: 6 00:00:00 15kN
11...Kc8-b8 12.Qd1-d2 Bf5-e4 13.Nb1-a3
³ (-0.49) Depth: 7 00:00:01 22kN
11...Kc8-b8 12.Nb1-a3 Nh6-g4 13.Qd1-d2 Bf5-e4
³ (-0.34) Depth: 8 00:00:01 36kN
11...Kc8-b8 12.Nb1-a3 Nh6-g4 13.Qd1-d2 a6-a5 14.Na3-c2 Bf5xc2
³ (-0.36) Depth: 9 00:00:04 93kN
11...Kc8-b8 12.Nb1-a3 Rh8-f8 13.b4-b5 a6xb5 14.Na3xb5 Nh6-g4
³ (-0.36) Depth: 10 00:00:08 176kN
11...Kc8-b8 12.Nb1-a3 Nh6-g4 13.Qd1-d2 a6-a5 14.Na3-c2 Bf5xc2 15.Ba4xc2 a5xb4 16.Ra1-b1
³ (-0.35) Depth: 11 00:00:12 273kN
11...Kc8-b8 12.Nb1-a3 Nh6-g4 13.Be3-c1 Bf5-e4 14.Na3-c2 h7-h5 15.h2-h3 Ng4-h6 16.Nc2-e3
³ (-0.37) Depth: 12 00:00:24 522kN
11...Kc8-b8 12.b4-b5 a6xb5 13.Ba4xb5
= (-0.19) Depth: 13 00:00:55 1113kN
11...Bf5-e4 12.Nb1-d2 Be4-d3 13.Rf1-e1 Nh6-f5 14.Ra1-c1 Kc8-b8 15.Be3-f2 h7-h5 16.Nd2-b3 Bd3-e4 17.Nb3-a5
= (-0.07) Depth: 14 00:07:18 8748kN
11...Bf5-e4 12.Nb1-d2 Be4-d3 13.Rf1-e1 Kc8-b8 14.Nd2-b3 Bd3-e4 15.Nb3-a5 b7-b5
= (-0.04) Depth: 15 00:08:09 9819kN
11...Bf5-e4 12.Nb1-d2 Be4-d3 13.Rf1-e1 Nh6-f5 14.Be3-f2 Kc8-b8 15.Nd2-b3 Bd3-e4 16.Nb3-d2
= (0.00) Depth: 16 00:10:40 12849kN
11...Nh6-g4 12.Be3-c1 Qd7-e8 13.h2-h3 Ng4-h6 14.Nb1-d2 Bf5-d3 15.Rf1-e1 Rh8-g8 16.Re1-e3 Bd3-g6 17.e5xf6 Be7xf6 18.Qd1-e1 Rd8-d6 19.Bc1-a3 b7-b5 20.Ba4-b3
= (0.14) Depth: 17 00:33:13 36883kN

(so k, 13.08.2008)
Here are my results with Rybka 3 (4 Cores @3GHz -- there are machines that are 4x faster than mine):

Code: Select all

Analysis from C:\tmp\wib.epd   
8/12/2008 9:54:59 PM Level: 1200 Seconds
Analyzing engine: Rybka 3

1) 0 1                  
    Avoid move: 
    Best move (Rybka 3): Qd7-e8
    Not found in: 20:00
      2	00:00	       2.482	149.504	+0.52	Nh6g4
      3	00:00	       2.994	180.344	+0.55	Nh6g4
      4	00:00	       3.626	218.413	+0.57	Nh6g4
      5	00:00	       5.497	175.904	+0.52	Nh6g4 Qd1d2
      6	00:00	       8.832	188.416	+0.47	Nh6g4 Qd1d2 Rh8f8
      7	00:00	      18.635	173.474	+0.36	Nh6g4 Qd1d2 f6xe5 Nf3xe5 Ng4xe5 f4xe5 Rh8f8
      7	00:00	      34.743	174.396	+0.39	Rh8f8 h2h3 Bf5e4 Qd1e2
      7	00:00	      57.166	162.605	+0.43	Kc8b8 Nb1a3 Nh6g4 Qd1d2 h7h5
      8	00:00	      65.750	172.194	+0.43	Kc8b8 Nb1a3 Nh6g4 Qd1d2 h7h5
      9	00:00	     107.460	163.748	+0.39	Kc8b8 Nb1a3 Nh6g4 Qd1d2 h7h5 h2h3 Ng4xe3
     10	00:01	     146.535	165.620	+0.39	Kc8b8 Nb1a3 Nh6g4 Qd1d2 h7h5 h2h3 Ng4xe3
     11	00:02	     277.472	170.138	+0.34	Kc8b8 Nb1a3 Nh6g4 Qd1d2 h7h5 h2h3 Ng4xe3 Qd2xe3 Rh8f8 Na3c2
     12	00:05	     948.745	182.581	+0.31	Kc8b8 Nb1a3 Nh6g4 Be3c1 h7h5 h2h3 Ng4h6 Na3c2 Rh8f8 Nc2e3
     13	00:11	   1.897.822	174.449	+0.24	Kc8b8 h2h3 Rh8g8 b4b5 a6xb5 Ba4xb5 Nh6f7 Qd1a4 g7g5 Bb5xc6 Qd7xc6 Qa4xc6 b7xc6 e5xf6 Be7xf6 Nf3e5 Nf7xe5
     14	00:16	   2.774.285	169.867	+0.23	Kc8b8 h2h3 Rh8g8 b4b5 a6xb5 Ba4xb5 Nh6f7 Qd1a4 g7g5 Bb5xc6 Qd7xc6 Qa4xc6 b7xc6 e5xf6 Be7xf6 f4xg5 Nf7xg5 Nf3xg5 Bf6xg5 Be3xg5 Rg8xg5 Kg1h2 Bf5d3 Rf1f3 Bd3e4
     15	01:40	  16.407.493	168.226	-0.16	Kc8b8 b4b5 a6xb5 Ba4xb5 Nh6g4 Be3c1 h7h5 h2h3 Ng4h6 Qd1a4 Rd8f8 Bc1a3 Nh6g8 Bb5xc6 Qd7xc6 Qa4xc6 b7xc6 Ba3c5 Be7xc5 d4xc5 Ng8e7 Nb1d2
     15	01:47	  17.610.027	167.524	+0.04	Nh6g4 Be3c1 Qd7e8 h2h3 Ng4h6 Nb1d2 Rh8f8 b4b5 a6xb5 Ba4xb5
     16	02:12	  21.242.394	164.816	 0.00	Nh6g4 Be3c1 Qd7e8 h2h3 Ng4h6 Nb1d2 Bf5d3 Rf1e1 Qe8d7 Nd2b3 Bd3e4 Nb3d2 Be4d3 Nd2b3 Bd3e4 Nb3d2 Be4d3 Nd2b3 Bd3e4 Nb3d2 Be4d3 Nd2b3 Bd3e4 Nb3d2 Be4d3 Nd2b3 Bd3e4 Nb3d2 Be4d3 Nd2b3 Bd3e4
     17	03:10	  29.855.281	160.934	 0.00	Nh6g4 Be3c1 Qd7e8 h2h3 Ng4h6 Nb1d2 Bf5d3 Rf1e1 Qe8d7 Nd2b3 Bd3e4 Nb3d2 Be4d3 Nd2b3 Bd3e4 Nb3d2 Be4d3 Nd2b3 Bd3e4 Nb3d2 Be4d3 Nd2b3 Bd3e4 Nb3d2 Be4d3 Nd2b3 Bd3e4 Nb3d2 Be4d3 Nd2b3 Bd3e4
     18	04:41	  43.096.481	156.776	-0.09	Nh6g4 Be3c1 Qd7e8 h2h3 Ng4h6 Nb1d2 Bf5d3 Rf1e1 Rh8g8 Re1e3 Bd3f5 g2g4 b7b5 Ba4b3 f6xe5 f4xe5 Qe8g6 Re3e2
     18	05:25	  49.037.080	154.438	 0.00	Qd7e8 Be3c1 Nh6g4 h2h3 Ng4h6 Nb1d2 Bf5d3 Rf1e1 Rh8g8 Re1e3 Bd3f5 g2g4 b7b5 Ba4b3 f6xe5 f4xe5 Qe8g6 Re3e2
     19	06:18	  57.667.070	156.107	-0.09	Qd7e8 Be3c1 Nh6g4 h2h3 Ng4h6 Nb1d2 Bf5d3 Rf1e1 Rh8g8 Re1e3 Bd3f5 g2g4 b7b5 Ba4b3 f6xe5 f4xe5 Qe8g6 Re3e2
     20	15:01	 139.274.727	158.158	-0.18	Qd7e8 Be3c1 Nh6g4
   8/12/2008 10:15:05 PM, Time for this analysis: 00:19:59, Rated time: 20:00

0 of 1 matching moves
8/12/2008 10:15:06 PM, Total time: 12:20:06 AM
Rated time: 20:00 = 1200 Seconds
It seems clear that white is better. It takes Rybka 1:40 to figure it out and 1:47 to find a better move.
On the other hand, white is not exactly stomping black into the ground. I guess that Rybka will win outright 90% of its games using a top level machine against a fide 2500 GM if playing at G/90 starting from this position.

I have the chess expertize of a box of hammers, but here is what I see looking at the position:
Development is at a similar level, but white has a space advantage. Nothing to get excited about. Actually, for black to have this position against white is really good for black. The field is nearly level. Usually, with experts on both sides, white will average 1/2 pawn better or so through the opening.
Top
Marc MP

Re: Have you decided why Vas

Post by Marc MP »

Dann Corbit wrote:
Uri Blass wrote:
Dr.Ex wrote:[d] 2kr3r/1ppqb1pp/p1n1pp1n/3pPb2/BP1P1P2/2P1BN2/P5PP/RN1Q1RK1 b - b3 0 11

How long does it like to play Kb8 in this position? I mean, come on. Isn't it stupid to walk directly into an attack?
Here is some analysis
Rybka on slow hardware admits after enough time that white is better here

Mischa - Kapaun, Wertungspartie, 3m + 0s 2008
2kr3r/1ppqb1pp/p1n1pp1n/3pPb2/BP1P1P2/2P1BN2/P5PP/RN1Q1RK1 b - b3 0 1

Analysis by Rybka 3 1-cpu 32-bit :

11...Nh6-g4
³ (-0.52) Depth: 2 00:00:00
11...Nh6-g4
³ (-0.55) Depth: 3 00:00:00
11...Nh6-g4
³ (-0.57) Depth: 4 00:00:00
11...Kc8-b8 12.h2-h3
³ (-0.38) Depth: 5 00:00:00
11...Kc8-b8 12.Qd1-d2 Rh8-f8
³ (-0.50) Depth: 6 00:00:00 15kN
11...Kc8-b8 12.Qd1-d2 Bf5-e4 13.Nb1-a3
³ (-0.49) Depth: 7 00:00:01 22kN
11...Kc8-b8 12.Nb1-a3 Nh6-g4 13.Qd1-d2 Bf5-e4
³ (-0.34) Depth: 8 00:00:01 36kN
11...Kc8-b8 12.Nb1-a3 Nh6-g4 13.Qd1-d2 a6-a5 14.Na3-c2 Bf5xc2
³ (-0.36) Depth: 9 00:00:04 93kN
11...Kc8-b8 12.Nb1-a3 Rh8-f8 13.b4-b5 a6xb5 14.Na3xb5 Nh6-g4
³ (-0.36) Depth: 10 00:00:08 176kN
11...Kc8-b8 12.Nb1-a3 Nh6-g4 13.Qd1-d2 a6-a5 14.Na3-c2 Bf5xc2 15.Ba4xc2 a5xb4 16.Ra1-b1
³ (-0.35) Depth: 11 00:00:12 273kN
11...Kc8-b8 12.Nb1-a3 Nh6-g4 13.Be3-c1 Bf5-e4 14.Na3-c2 h7-h5 15.h2-h3 Ng4-h6 16.Nc2-e3
³ (-0.37) Depth: 12 00:00:24 522kN
11...Kc8-b8 12.b4-b5 a6xb5 13.Ba4xb5
= (-0.19) Depth: 13 00:00:55 1113kN
11...Bf5-e4 12.Nb1-d2 Be4-d3 13.Rf1-e1 Nh6-f5 14.Ra1-c1 Kc8-b8 15.Be3-f2 h7-h5 16.Nd2-b3 Bd3-e4 17.Nb3-a5
= (-0.07) Depth: 14 00:07:18 8748kN
11...Bf5-e4 12.Nb1-d2 Be4-d3 13.Rf1-e1 Kc8-b8 14.Nd2-b3 Bd3-e4 15.Nb3-a5 b7-b5
= (-0.04) Depth: 15 00:08:09 9819kN
11...Bf5-e4 12.Nb1-d2 Be4-d3 13.Rf1-e1 Nh6-f5 14.Be3-f2 Kc8-b8 15.Nd2-b3 Bd3-e4 16.Nb3-d2
= (0.00) Depth: 16 00:10:40 12849kN
11...Nh6-g4 12.Be3-c1 Qd7-e8 13.h2-h3 Ng4-h6 14.Nb1-d2 Bf5-d3 15.Rf1-e1 Rh8-g8 16.Re1-e3 Bd3-g6 17.e5xf6 Be7xf6 18.Qd1-e1 Rd8-d6 19.Bc1-a3 b7-b5 20.Ba4-b3
= (0.14) Depth: 17 00:33:13 36883kN

(so k, 13.08.2008)
Here are my results with Rybka 3 (4 Cores @3GHz -- there are machines that are 4x faster than mine):

Code: Select all

Analysis from C:\tmp\wib.epd   
8/12/2008 9:54:59 PM Level: 1200 Seconds
Analyzing engine: Rybka 3

1) 0 1                  
    Avoid move: 
    Best move (Rybka 3): Qd7-e8
    Not found in: 20:00
      2	00:00	       2.482	149.504	+0.52	Nh6g4
      3	00:00	       2.994	180.344	+0.55	Nh6g4
      4	00:00	       3.626	218.413	+0.57	Nh6g4
      5	00:00	       5.497	175.904	+0.52	Nh6g4 Qd1d2
      6	00:00	       8.832	188.416	+0.47	Nh6g4 Qd1d2 Rh8f8
      7	00:00	      18.635	173.474	+0.36	Nh6g4 Qd1d2 f6xe5 Nf3xe5 Ng4xe5 f4xe5 Rh8f8
      7	00:00	      34.743	174.396	+0.39	Rh8f8 h2h3 Bf5e4 Qd1e2
      7	00:00	      57.166	162.605	+0.43	Kc8b8 Nb1a3 Nh6g4 Qd1d2 h7h5
      8	00:00	      65.750	172.194	+0.43	Kc8b8 Nb1a3 Nh6g4 Qd1d2 h7h5
      9	00:00	     107.460	163.748	+0.39	Kc8b8 Nb1a3 Nh6g4 Qd1d2 h7h5 h2h3 Ng4xe3
     10	00:01	     146.535	165.620	+0.39	Kc8b8 Nb1a3 Nh6g4 Qd1d2 h7h5 h2h3 Ng4xe3
     11	00:02	     277.472	170.138	+0.34	Kc8b8 Nb1a3 Nh6g4 Qd1d2 h7h5 h2h3 Ng4xe3 Qd2xe3 Rh8f8 Na3c2
     12	00:05	     948.745	182.581	+0.31	Kc8b8 Nb1a3 Nh6g4 Be3c1 h7h5 h2h3 Ng4h6 Na3c2 Rh8f8 Nc2e3
     13	00:11	   1.897.822	174.449	+0.24	Kc8b8 h2h3 Rh8g8 b4b5 a6xb5 Ba4xb5 Nh6f7 Qd1a4 g7g5 Bb5xc6 Qd7xc6 Qa4xc6 b7xc6 e5xf6 Be7xf6 Nf3e5 Nf7xe5
     14	00:16	   2.774.285	169.867	+0.23	Kc8b8 h2h3 Rh8g8 b4b5 a6xb5 Ba4xb5 Nh6f7 Qd1a4 g7g5 Bb5xc6 Qd7xc6 Qa4xc6 b7xc6 e5xf6 Be7xf6 f4xg5 Nf7xg5 Nf3xg5 Bf6xg5 Be3xg5 Rg8xg5 Kg1h2 Bf5d3 Rf1f3 Bd3e4
     15	01:40	  16.407.493	168.226	-0.16	Kc8b8 b4b5 a6xb5 Ba4xb5 Nh6g4 Be3c1 h7h5 h2h3 Ng4h6 Qd1a4 Rd8f8 Bc1a3 Nh6g8 Bb5xc6 Qd7xc6 Qa4xc6 b7xc6 Ba3c5 Be7xc5 d4xc5 Ng8e7 Nb1d2
     15	01:47	  17.610.027	167.524	+0.04	Nh6g4 Be3c1 Qd7e8 h2h3 Ng4h6 Nb1d2 Rh8f8 b4b5 a6xb5 Ba4xb5
     16	02:12	  21.242.394	164.816	 0.00	Nh6g4 Be3c1 Qd7e8 h2h3 Ng4h6 Nb1d2 Bf5d3 Rf1e1 Qe8d7 Nd2b3 Bd3e4 Nb3d2 Be4d3 Nd2b3 Bd3e4 Nb3d2 Be4d3 Nd2b3 Bd3e4 Nb3d2 Be4d3 Nd2b3 Bd3e4 Nb3d2 Be4d3 Nd2b3 Bd3e4 Nb3d2 Be4d3 Nd2b3 Bd3e4
     17	03:10	  29.855.281	160.934	 0.00	Nh6g4 Be3c1 Qd7e8 h2h3 Ng4h6 Nb1d2 Bf5d3 Rf1e1 Qe8d7 Nd2b3 Bd3e4 Nb3d2 Be4d3 Nd2b3 Bd3e4 Nb3d2 Be4d3 Nd2b3 Bd3e4 Nb3d2 Be4d3 Nd2b3 Bd3e4 Nb3d2 Be4d3 Nd2b3 Bd3e4 Nb3d2 Be4d3 Nd2b3 Bd3e4
     18	04:41	  43.096.481	156.776	-0.09	Nh6g4 Be3c1 Qd7e8 h2h3 Ng4h6 Nb1d2 Bf5d3 Rf1e1 Rh8g8 Re1e3 Bd3f5 g2g4 b7b5 Ba4b3 f6xe5 f4xe5 Qe8g6 Re3e2
     18	05:25	  49.037.080	154.438	 0.00	Qd7e8 Be3c1 Nh6g4 h2h3 Ng4h6 Nb1d2 Bf5d3 Rf1e1 Rh8g8 Re1e3 Bd3f5 g2g4 b7b5 Ba4b3 f6xe5 f4xe5 Qe8g6 Re3e2
     19	06:18	  57.667.070	156.107	-0.09	Qd7e8 Be3c1 Nh6g4 h2h3 Ng4h6 Nb1d2 Bf5d3 Rf1e1 Rh8g8 Re1e3 Bd3f5 g2g4 b7b5 Ba4b3 f6xe5 f4xe5 Qe8g6 Re3e2
     20	15:01	 139.274.727	158.158	-0.18	Qd7e8 Be3c1 Nh6g4
   8/12/2008 10:15:05 PM, Time for this analysis: 00:19:59, Rated time: 20:00

0 of 1 matching moves
8/12/2008 10:15:06 PM, Total time: 12:20:06 AM
Rated time: 20:00 = 1200 Seconds
It seems clear that white is better. It takes Rybka 1:40 to figure it out and 1:47 to find a better move.
On the other hand, white is not exactly stomping black into the ground. I guess that Rybka will win outright 90% of its games using a top level machine against a fide 2500 GM if playing at G/90 starting from this position.

I have the chess expertize of a box of hammers, but here is what I see looking at the position:
Development is at a similar level, but white has a space advantage. Nothing to get excited about. Actually, for black to have this position against white is really good for black. The field is nearly level. Usually, with experts on both sides, white will average 1/2 pawn better or so through the opening.
Hi Dan,

Out of curiousity would you have cpu time to check if Rybka 3 still want to castle long here : [d]r3k2r/1ppqb1pp/p1n1pp1n/3pPb2/B2P1P2/2P1BN2/PP4PP/RN1Q1RK1 b kq - 3 1

From my chess player experience it seems to me like a strategical mistake, but hard to prove on this forum (or maybe hard to prove at all!). If Rybka 3.0 would change its mind after some lomg pondering then... see my post here: http://64.68.157.89/forum/viewtopic.php ... 28&t=22900
Top
Dann Corbit
Posts: 12777
Joined: Wed Mar 08, 2006 8:57 pm
Location: Redmond, WA USA

Re: Have you decided why Vas

Post by Dann Corbit »

Marc MP wrote:
Dann Corbit wrote:
Uri Blass wrote:
Dr.Ex wrote:[d] 2kr3r/1ppqb1pp/p1n1pp1n/3pPb2/BP1P1P2/2P1BN2/P5PP/RN1Q1RK1 b - b3 0 11

How long does it like to play Kb8 in this position? I mean, come on. Isn't it stupid to walk directly into an attack?
Here is some analysis
Rybka on slow hardware admits after enough time that white is better here

Mischa - Kapaun, Wertungspartie, 3m + 0s 2008
2kr3r/1ppqb1pp/p1n1pp1n/3pPb2/BP1P1P2/2P1BN2/P5PP/RN1Q1RK1 b - b3 0 1

Analysis by Rybka 3 1-cpu 32-bit :

11...Nh6-g4
³ (-0.52) Depth: 2 00:00:00
11...Nh6-g4
³ (-0.55) Depth: 3 00:00:00
11...Nh6-g4
³ (-0.57) Depth: 4 00:00:00
11...Kc8-b8 12.h2-h3
³ (-0.38) Depth: 5 00:00:00
11...Kc8-b8 12.Qd1-d2 Rh8-f8
³ (-0.50) Depth: 6 00:00:00 15kN
11...Kc8-b8 12.Qd1-d2 Bf5-e4 13.Nb1-a3
³ (-0.49) Depth: 7 00:00:01 22kN
11...Kc8-b8 12.Nb1-a3 Nh6-g4 13.Qd1-d2 Bf5-e4
³ (-0.34) Depth: 8 00:00:01 36kN
11...Kc8-b8 12.Nb1-a3 Nh6-g4 13.Qd1-d2 a6-a5 14.Na3-c2 Bf5xc2
³ (-0.36) Depth: 9 00:00:04 93kN
11...Kc8-b8 12.Nb1-a3 Rh8-f8 13.b4-b5 a6xb5 14.Na3xb5 Nh6-g4
³ (-0.36) Depth: 10 00:00:08 176kN
11...Kc8-b8 12.Nb1-a3 Nh6-g4 13.Qd1-d2 a6-a5 14.Na3-c2 Bf5xc2 15.Ba4xc2 a5xb4 16.Ra1-b1
³ (-0.35) Depth: 11 00:00:12 273kN
11...Kc8-b8 12.Nb1-a3 Nh6-g4 13.Be3-c1 Bf5-e4 14.Na3-c2 h7-h5 15.h2-h3 Ng4-h6 16.Nc2-e3
³ (-0.37) Depth: 12 00:00:24 522kN
11...Kc8-b8 12.b4-b5 a6xb5 13.Ba4xb5
= (-0.19) Depth: 13 00:00:55 1113kN
11...Bf5-e4 12.Nb1-d2 Be4-d3 13.Rf1-e1 Nh6-f5 14.Ra1-c1 Kc8-b8 15.Be3-f2 h7-h5 16.Nd2-b3 Bd3-e4 17.Nb3-a5
= (-0.07) Depth: 14 00:07:18 8748kN
11...Bf5-e4 12.Nb1-d2 Be4-d3 13.Rf1-e1 Kc8-b8 14.Nd2-b3 Bd3-e4 15.Nb3-a5 b7-b5
= (-0.04) Depth: 15 00:08:09 9819kN
11...Bf5-e4 12.Nb1-d2 Be4-d3 13.Rf1-e1 Nh6-f5 14.Be3-f2 Kc8-b8 15.Nd2-b3 Bd3-e4 16.Nb3-d2
= (0.00) Depth: 16 00:10:40 12849kN
11...Nh6-g4 12.Be3-c1 Qd7-e8 13.h2-h3 Ng4-h6 14.Nb1-d2 Bf5-d3 15.Rf1-e1 Rh8-g8 16.Re1-e3 Bd3-g6 17.e5xf6 Be7xf6 18.Qd1-e1 Rd8-d6 19.Bc1-a3 b7-b5 20.Ba4-b3
= (0.14) Depth: 17 00:33:13 36883kN

(so k, 13.08.2008)
Here are my results with Rybka 3 (4 Cores @3GHz -- there are machines that are 4x faster than mine):

Code: Select all

Analysis from C:\tmp\wib.epd   
8/12/2008 9:54:59 PM Level: 1200 Seconds
Analyzing engine: Rybka 3

1) 0 1                  
    Avoid move: 
    Best move (Rybka 3): Qd7-e8
    Not found in: 20:00
      2	00:00	       2.482	149.504	+0.52	Nh6g4
      3	00:00	       2.994	180.344	+0.55	Nh6g4
      4	00:00	       3.626	218.413	+0.57	Nh6g4
      5	00:00	       5.497	175.904	+0.52	Nh6g4 Qd1d2
      6	00:00	       8.832	188.416	+0.47	Nh6g4 Qd1d2 Rh8f8
      7	00:00	      18.635	173.474	+0.36	Nh6g4 Qd1d2 f6xe5 Nf3xe5 Ng4xe5 f4xe5 Rh8f8
      7	00:00	      34.743	174.396	+0.39	Rh8f8 h2h3 Bf5e4 Qd1e2
      7	00:00	      57.166	162.605	+0.43	Kc8b8 Nb1a3 Nh6g4 Qd1d2 h7h5
      8	00:00	      65.750	172.194	+0.43	Kc8b8 Nb1a3 Nh6g4 Qd1d2 h7h5
      9	00:00	     107.460	163.748	+0.39	Kc8b8 Nb1a3 Nh6g4 Qd1d2 h7h5 h2h3 Ng4xe3
     10	00:01	     146.535	165.620	+0.39	Kc8b8 Nb1a3 Nh6g4 Qd1d2 h7h5 h2h3 Ng4xe3
     11	00:02	     277.472	170.138	+0.34	Kc8b8 Nb1a3 Nh6g4 Qd1d2 h7h5 h2h3 Ng4xe3 Qd2xe3 Rh8f8 Na3c2
     12	00:05	     948.745	182.581	+0.31	Kc8b8 Nb1a3 Nh6g4 Be3c1 h7h5 h2h3 Ng4h6 Na3c2 Rh8f8 Nc2e3
     13	00:11	   1.897.822	174.449	+0.24	Kc8b8 h2h3 Rh8g8 b4b5 a6xb5 Ba4xb5 Nh6f7 Qd1a4 g7g5 Bb5xc6 Qd7xc6 Qa4xc6 b7xc6 e5xf6 Be7xf6 Nf3e5 Nf7xe5
     14	00:16	   2.774.285	169.867	+0.23	Kc8b8 h2h3 Rh8g8 b4b5 a6xb5 Ba4xb5 Nh6f7 Qd1a4 g7g5 Bb5xc6 Qd7xc6 Qa4xc6 b7xc6 e5xf6 Be7xf6 f4xg5 Nf7xg5 Nf3xg5 Bf6xg5 Be3xg5 Rg8xg5 Kg1h2 Bf5d3 Rf1f3 Bd3e4
     15	01:40	  16.407.493	168.226	-0.16	Kc8b8 b4b5 a6xb5 Ba4xb5 Nh6g4 Be3c1 h7h5 h2h3 Ng4h6 Qd1a4 Rd8f8 Bc1a3 Nh6g8 Bb5xc6 Qd7xc6 Qa4xc6 b7xc6 Ba3c5 Be7xc5 d4xc5 Ng8e7 Nb1d2
     15	01:47	  17.610.027	167.524	+0.04	Nh6g4 Be3c1 Qd7e8 h2h3 Ng4h6 Nb1d2 Rh8f8 b4b5 a6xb5 Ba4xb5
     16	02:12	  21.242.394	164.816	 0.00	Nh6g4 Be3c1 Qd7e8 h2h3 Ng4h6 Nb1d2 Bf5d3 Rf1e1 Qe8d7 Nd2b3 Bd3e4 Nb3d2 Be4d3 Nd2b3 Bd3e4 Nb3d2 Be4d3 Nd2b3 Bd3e4 Nb3d2 Be4d3 Nd2b3 Bd3e4 Nb3d2 Be4d3 Nd2b3 Bd3e4 Nb3d2 Be4d3 Nd2b3 Bd3e4
     17	03:10	  29.855.281	160.934	 0.00	Nh6g4 Be3c1 Qd7e8 h2h3 Ng4h6 Nb1d2 Bf5d3 Rf1e1 Qe8d7 Nd2b3 Bd3e4 Nb3d2 Be4d3 Nd2b3 Bd3e4 Nb3d2 Be4d3 Nd2b3 Bd3e4 Nb3d2 Be4d3 Nd2b3 Bd3e4 Nb3d2 Be4d3 Nd2b3 Bd3e4 Nb3d2 Be4d3 Nd2b3 Bd3e4
     18	04:41	  43.096.481	156.776	-0.09	Nh6g4 Be3c1 Qd7e8 h2h3 Ng4h6 Nb1d2 Bf5d3 Rf1e1 Rh8g8 Re1e3 Bd3f5 g2g4 b7b5 Ba4b3 f6xe5 f4xe5 Qe8g6 Re3e2
     18	05:25	  49.037.080	154.438	 0.00	Qd7e8 Be3c1 Nh6g4 h2h3 Ng4h6 Nb1d2 Bf5d3 Rf1e1 Rh8g8 Re1e3 Bd3f5 g2g4 b7b5 Ba4b3 f6xe5 f4xe5 Qe8g6 Re3e2
     19	06:18	  57.667.070	156.107	-0.09	Qd7e8 Be3c1 Nh6g4 h2h3 Ng4h6 Nb1d2 Bf5d3 Rf1e1 Rh8g8 Re1e3 Bd3f5 g2g4 b7b5 Ba4b3 f6xe5 f4xe5 Qe8g6 Re3e2
     20	15:01	 139.274.727	158.158	-0.18	Qd7e8 Be3c1 Nh6g4
   8/12/2008 10:15:05 PM, Time for this analysis: 00:19:59, Rated time: 20:00

0 of 1 matching moves
8/12/2008 10:15:06 PM, Total time: 12:20:06 AM
Rated time: 20:00 = 1200 Seconds
It seems clear that white is better. It takes Rybka 1:40 to figure it out and 1:47 to find a better move.
On the other hand, white is not exactly stomping black into the ground. I guess that Rybka will win outright 90% of its games using a top level machine against a fide 2500 GM if playing at G/90 starting from this position.

I have the chess expertize of a box of hammers, but here is what I see looking at the position:
Development is at a similar level, but white has a space advantage. Nothing to get excited about. Actually, for black to have this position against white is really good for black. The field is nearly level. Usually, with experts on both sides, white will average 1/2 pawn better or so through the opening.
Hi Dan,

Out of curiousity would you have cpu time to check if Rybka 3 still want to castle long here : [d]r3k2r/1ppqb1pp/p1n1pp1n/3pPb2/B2P1P2/2P1BN2/PP4PP/RN1Q1RK1 b kq - 3 1

From my chess player experience it seems to me like a strategical mistake, but hard to prove on this forum (or maybe hard to prove at all!). If Rybka 3.0 would change its mind after some lomg pondering then... see my post here: http://64.68.157.89/forum/viewtopic.php ... 28&t=22900
I am quite sure that you are right about this position. Rybka is analyzing it right now, and the eval is taking a slow and inexorable drop into oblivion. It has the clear pattern of "this is a very bad positional move" that I see all the time... A move initially looks like it offers something, but on each new ply the eval stays about the same or drops. When the grinding has stopped I will post the result here.

Meanwhile, I am going to post Dann Corbit's first {probably also last and horribly flawed} law of chess programming:

1. Never, never, never castle opposite sides if you are a computer program unless one of the follwing is true:
A. Your same-side castle shelter is in tatters.
OR
B. You are planning your own pawn storm.

I make this rule because pawn storms are so utterly glacial in chess programming terms that it is very unlikely to see the problem before it is too late.
Top
Dann Corbit
Posts: 12777
Joined: Wed Mar 08, 2006 8:57 pm
Location: Redmond, WA USA

Re: Have you decided why Vas

Post by Dann Corbit »

Dann Corbit wrote:
Marc MP wrote:
Dann Corbit wrote:
Uri Blass wrote:
Dr.Ex wrote:[d] 2kr3r/1ppqb1pp/p1n1pp1n/3pPb2/BP1P1P2/2P1BN2/P5PP/RN1Q1RK1 b - b3 0 11

How long does it like to play Kb8 in this position? I mean, come on. Isn't it stupid to walk directly into an attack?
Here is some analysis
Rybka on slow hardware admits after enough time that white is better here

Mischa - Kapaun, Wertungspartie, 3m + 0s 2008
2kr3r/1ppqb1pp/p1n1pp1n/3pPb2/BP1P1P2/2P1BN2/P5PP/RN1Q1RK1 b - b3 0 1

Analysis by Rybka 3 1-cpu 32-bit :

11...Nh6-g4
³ (-0.52) Depth: 2 00:00:00
11...Nh6-g4
³ (-0.55) Depth: 3 00:00:00
11...Nh6-g4
³ (-0.57) Depth: 4 00:00:00
11...Kc8-b8 12.h2-h3
³ (-0.38) Depth: 5 00:00:00
11...Kc8-b8 12.Qd1-d2 Rh8-f8
³ (-0.50) Depth: 6 00:00:00 15kN
11...Kc8-b8 12.Qd1-d2 Bf5-e4 13.Nb1-a3
³ (-0.49) Depth: 7 00:00:01 22kN
11...Kc8-b8 12.Nb1-a3 Nh6-g4 13.Qd1-d2 Bf5-e4
³ (-0.34) Depth: 8 00:00:01 36kN
11...Kc8-b8 12.Nb1-a3 Nh6-g4 13.Qd1-d2 a6-a5 14.Na3-c2 Bf5xc2
³ (-0.36) Depth: 9 00:00:04 93kN
11...Kc8-b8 12.Nb1-a3 Rh8-f8 13.b4-b5 a6xb5 14.Na3xb5 Nh6-g4
³ (-0.36) Depth: 10 00:00:08 176kN
11...Kc8-b8 12.Nb1-a3 Nh6-g4 13.Qd1-d2 a6-a5 14.Na3-c2 Bf5xc2 15.Ba4xc2 a5xb4 16.Ra1-b1
³ (-0.35) Depth: 11 00:00:12 273kN
11...Kc8-b8 12.Nb1-a3 Nh6-g4 13.Be3-c1 Bf5-e4 14.Na3-c2 h7-h5 15.h2-h3 Ng4-h6 16.Nc2-e3
³ (-0.37) Depth: 12 00:00:24 522kN
11...Kc8-b8 12.b4-b5 a6xb5 13.Ba4xb5
= (-0.19) Depth: 13 00:00:55 1113kN
11...Bf5-e4 12.Nb1-d2 Be4-d3 13.Rf1-e1 Nh6-f5 14.Ra1-c1 Kc8-b8 15.Be3-f2 h7-h5 16.Nd2-b3 Bd3-e4 17.Nb3-a5
= (-0.07) Depth: 14 00:07:18 8748kN
11...Bf5-e4 12.Nb1-d2 Be4-d3 13.Rf1-e1 Kc8-b8 14.Nd2-b3 Bd3-e4 15.Nb3-a5 b7-b5
= (-0.04) Depth: 15 00:08:09 9819kN
11...Bf5-e4 12.Nb1-d2 Be4-d3 13.Rf1-e1 Nh6-f5 14.Be3-f2 Kc8-b8 15.Nd2-b3 Bd3-e4 16.Nb3-d2
= (0.00) Depth: 16 00:10:40 12849kN
11...Nh6-g4 12.Be3-c1 Qd7-e8 13.h2-h3 Ng4-h6 14.Nb1-d2 Bf5-d3 15.Rf1-e1 Rh8-g8 16.Re1-e3 Bd3-g6 17.e5xf6 Be7xf6 18.Qd1-e1 Rd8-d6 19.Bc1-a3 b7-b5 20.Ba4-b3
= (0.14) Depth: 17 00:33:13 36883kN

(so k, 13.08.2008)
Here are my results with Rybka 3 (4 Cores @3GHz -- there are machines that are 4x faster than mine):

Code: Select all

Analysis from C:\tmp\wib.epd   
8/12/2008 9:54:59 PM Level: 1200 Seconds
Analyzing engine: Rybka 3

1) 0 1                  
    Avoid move: 
    Best move (Rybka 3): Qd7-e8
    Not found in: 20:00
      2	00:00	       2.482	149.504	+0.52	Nh6g4
      3	00:00	       2.994	180.344	+0.55	Nh6g4
      4	00:00	       3.626	218.413	+0.57	Nh6g4
      5	00:00	       5.497	175.904	+0.52	Nh6g4 Qd1d2
      6	00:00	       8.832	188.416	+0.47	Nh6g4 Qd1d2 Rh8f8
      7	00:00	      18.635	173.474	+0.36	Nh6g4 Qd1d2 f6xe5 Nf3xe5 Ng4xe5 f4xe5 Rh8f8
      7	00:00	      34.743	174.396	+0.39	Rh8f8 h2h3 Bf5e4 Qd1e2
      7	00:00	      57.166	162.605	+0.43	Kc8b8 Nb1a3 Nh6g4 Qd1d2 h7h5
      8	00:00	      65.750	172.194	+0.43	Kc8b8 Nb1a3 Nh6g4 Qd1d2 h7h5
      9	00:00	     107.460	163.748	+0.39	Kc8b8 Nb1a3 Nh6g4 Qd1d2 h7h5 h2h3 Ng4xe3
     10	00:01	     146.535	165.620	+0.39	Kc8b8 Nb1a3 Nh6g4 Qd1d2 h7h5 h2h3 Ng4xe3
     11	00:02	     277.472	170.138	+0.34	Kc8b8 Nb1a3 Nh6g4 Qd1d2 h7h5 h2h3 Ng4xe3 Qd2xe3 Rh8f8 Na3c2
     12	00:05	     948.745	182.581	+0.31	Kc8b8 Nb1a3 Nh6g4 Be3c1 h7h5 h2h3 Ng4h6 Na3c2 Rh8f8 Nc2e3
     13	00:11	   1.897.822	174.449	+0.24	Kc8b8 h2h3 Rh8g8 b4b5 a6xb5 Ba4xb5 Nh6f7 Qd1a4 g7g5 Bb5xc6 Qd7xc6 Qa4xc6 b7xc6 e5xf6 Be7xf6 Nf3e5 Nf7xe5
     14	00:16	   2.774.285	169.867	+0.23	Kc8b8 h2h3 Rh8g8 b4b5 a6xb5 Ba4xb5 Nh6f7 Qd1a4 g7g5 Bb5xc6 Qd7xc6 Qa4xc6 b7xc6 e5xf6 Be7xf6 f4xg5 Nf7xg5 Nf3xg5 Bf6xg5 Be3xg5 Rg8xg5 Kg1h2 Bf5d3 Rf1f3 Bd3e4
     15	01:40	  16.407.493	168.226	-0.16	Kc8b8 b4b5 a6xb5 Ba4xb5 Nh6g4 Be3c1 h7h5 h2h3 Ng4h6 Qd1a4 Rd8f8 Bc1a3 Nh6g8 Bb5xc6 Qd7xc6 Qa4xc6 b7xc6 Ba3c5 Be7xc5 d4xc5 Ng8e7 Nb1d2
     15	01:47	  17.610.027	167.524	+0.04	Nh6g4 Be3c1 Qd7e8 h2h3 Ng4h6 Nb1d2 Rh8f8 b4b5 a6xb5 Ba4xb5
     16	02:12	  21.242.394	164.816	 0.00	Nh6g4 Be3c1 Qd7e8 h2h3 Ng4h6 Nb1d2 Bf5d3 Rf1e1 Qe8d7 Nd2b3 Bd3e4 Nb3d2 Be4d3 Nd2b3 Bd3e4 Nb3d2 Be4d3 Nd2b3 Bd3e4 Nb3d2 Be4d3 Nd2b3 Bd3e4 Nb3d2 Be4d3 Nd2b3 Bd3e4 Nb3d2 Be4d3 Nd2b3 Bd3e4
     17	03:10	  29.855.281	160.934	 0.00	Nh6g4 Be3c1 Qd7e8 h2h3 Ng4h6 Nb1d2 Bf5d3 Rf1e1 Qe8d7 Nd2b3 Bd3e4 Nb3d2 Be4d3 Nd2b3 Bd3e4 Nb3d2 Be4d3 Nd2b3 Bd3e4 Nb3d2 Be4d3 Nd2b3 Bd3e4 Nb3d2 Be4d3 Nd2b3 Bd3e4 Nb3d2 Be4d3 Nd2b3 Bd3e4
     18	04:41	  43.096.481	156.776	-0.09	Nh6g4 Be3c1 Qd7e8 h2h3 Ng4h6 Nb1d2 Bf5d3 Rf1e1 Rh8g8 Re1e3 Bd3f5 g2g4 b7b5 Ba4b3 f6xe5 f4xe5 Qe8g6 Re3e2
     18	05:25	  49.037.080	154.438	 0.00	Qd7e8 Be3c1 Nh6g4 h2h3 Ng4h6 Nb1d2 Bf5d3 Rf1e1 Rh8g8 Re1e3 Bd3f5 g2g4 b7b5 Ba4b3 f6xe5 f4xe5 Qe8g6 Re3e2
     19	06:18	  57.667.070	156.107	-0.09	Qd7e8 Be3c1 Nh6g4 h2h3 Ng4h6 Nb1d2 Bf5d3 Rf1e1 Rh8g8 Re1e3 Bd3f5 g2g4 b7b5 Ba4b3 f6xe5 f4xe5 Qe8g6 Re3e2
     20	15:01	 139.274.727	158.158	-0.18	Qd7e8 Be3c1 Nh6g4
   8/12/2008 10:15:05 PM, Time for this analysis: 00:19:59, Rated time: 20:00

0 of 1 matching moves
8/12/2008 10:15:06 PM, Total time: 12:20:06 AM
Rated time: 20:00 = 1200 Seconds
It seems clear that white is better. It takes Rybka 1:40 to figure it out and 1:47 to find a better move.
On the other hand, white is not exactly stomping black into the ground. I guess that Rybka will win outright 90% of its games using a top level machine against a fide 2500 GM if playing at G/90 starting from this position.

I have the chess expertize of a box of hammers, but here is what I see looking at the position:
Development is at a similar level, but white has a space advantage. Nothing to get excited about. Actually, for black to have this position against white is really good for black. The field is nearly level. Usually, with experts on both sides, white will average 1/2 pawn better or so through the opening.
Hi Dan,

Out of curiousity would you have cpu time to check if Rybka 3 still want to castle long here : [d]r3k2r/1ppqb1pp/p1n1pp1n/3pPb2/B2P1P2/2P1BN2/PP4PP/RN1Q1RK1 b kq - 3 1

From my chess player experience it seems to me like a strategical mistake, but hard to prove on this forum (or maybe hard to prove at all!). If Rybka 3.0 would change its mind after some lomg pondering then... see my post here: http://64.68.157.89/forum/viewtopic.php ... 28&t=22900
I am quite sure that you are right about this position. Rybka is analyzing it right now, and the eval is taking a slow and inexorable drop into oblivion. It has the clear pattern of "this is a very bad positional move" that I see all the time... A move initially looks like it offers something, but on each new ply the eval stays about the same or drops. When the grinding has stopped I will post the result here.

Meanwhile, I am going to post Dann Corbit's first {probably also last and horribly flawed} law of chess programming:

1. Never, never, never castle opposite sides if you are a computer program unless one of the follwing is true:
A. Your same-side castle shelter is in tatters.
OR
B. You are planning your own pawn storm.

I make this rule because pawn storms are so utterly glacial in chess programming terms that it is very unlikely to see the problem before it is too late.
Thankfully, Rybka finally saw the error of her ways. As is clear from a casual perusal of the evaluation over time, there was nowhere to go but down.

Code: Select all


Analysis from C:\tmp\long-castle-or-not.epd   
8/12/2008 11:05:59 PM Level: 7200 Seconds
Analyzing engine: Rybka 3

1) 3 1                  
    Avoid move: 
    Best move (Rybka 3): Nh6-g4
    Not found in: 2:00:00
      2	00:00	       1.336	41.456	+0.27	00
      3	00:00	       2.658	82.478	+0.33	00
      4	00:00	       3.441	55.056	+0.27	00
      4	00:00	       5.374	85.984	+0.28	Nh6g4
      5	00:00	       6.182	80.131	+0.27	Nh6g4 Be3c1
      6	00:00	      10.657	114.871	+0.27	Nh6g4 Be3c1 00
      7	00:00	      14.219	115.557	+0.27	Nh6g4 Be3c1 00 h2h3
      7	00:00	      59.326	168.749	+0.40	000 Nb1a3 Bf5e4 Qd1d2
      8	00:00	      95.012	182.880	+0.31	000 Nb1a3 Nh6g4 Qd1d2 Rh8f8 Na3c2
      9	00:00	     110.095	184.815	+0.31	000 Nb1a3 Nh6g4 Qd1d2 Rh8f8 Na3c2
     10	00:01	     164.402	185.814	+0.35	000 Nb1a3 Nh6g4 Qd1d2 Be7xa3 b2xa3 h7h5 h2h3
     11	00:01	     305.396	188.958	+0.38	000 Nb1a3 Be7xa3 b2xa3 Rh8f8 Qd1e2 Kc8b8 Ra1d1 Nh6g4
     12	00:02	     445.831	191.177	+0.30	000 Nb1a3 Be7xa3 b2xa3 Kc8b8 Qd1e2 Nh6g4 h2h3 Ng4xe3
     13	00:05	     861.384	185.306	+0.30	000 Nb1a3 Be7xa3 b2xa3 Kc8b8 Qd1e2 Nh6g4 h2h3 Ng4xe3 Qe2xe3 h7h5 Ra1d1 Rh8f8
     14	00:07	   1.367.852	181.717	+0.29	000 Nb1a3 Be7xa3 b2xa3 Nh6g4 Be3c1 h7h5 h2h3 Ng4h6 Nf3h4 Bf5e4 Bc1e3 Nh6f5
     15	00:16	   2.798.305	180.592	+0.28	000 Nb1a3 Be7xa3 b2xa3 Nh6g4 Be3c1 h7h5 Ba4c2 Bf5xc2 Qd1xc2 Ng4h6 Ra1b1 Nh6f5 Qc2b2 Nc6a5
     16	00:29	   5.013.067	173.179	+0.26	000 Nb1a3 Be7xa3 b2xa3 Nh6g4 Qd1d2 h7h5 h2h3 Ng4xe3 Qd2xe3 Kc8b8 Ra1d1 Bf5g6 Rd1d2 Bg6e4 Rd2b2
     17	01:04	  10.639.295	169.545	+0.26	000 h2h3 Kc8b8 Nb1d2 Rh8f8 Rf1e1 Nh6f7 b2b4 Rf8g8 Ba4c2 g7g5 e5xf6 Be7xf6 Bc2xf5 g5xf4 Be3xf4 e6xf5 Nd2b3
     18	02:20	  22.535.611	164.322	+0.15	000 h2h3 Kc8b8 b2b4 Bf5e4 Nb1d2 Be4d3 Rf1e1 Nh6f5 Be3f2 Rh8g8 g2g4 Nf5h6 Re1e3
     19	05:17	  50.537.566	162.940	+0.07	000 h2h3 Kc8b8 b2b4 Bf5g6 Nb1d2 Nh6f5 Be3f2 h7h5 Ba4c2 f6xe5 f4xe5 Nc6a7 a2a4 Qd7c6 Ra1c1 Bg6h7 Nd2b3 Qc6xa4 Nb3c5
     20	17:27	 163.126.235	159.459	 0.00	000 h2h3 Kc8b8 b2b4 Bf5e4 Nb1d2 Be4d3 Rf1e1 Nh6f5 Be3f2 Rh8f8 Nd2b3 Bd3e4 Nb3d2
     20	34:40	 336.204.380	165.468	+0.04	Nh6g4 Be3c1 h7h5 Nf3h4 Bf5e4 Ba4c2 000 Bc2xe4 d5xe4 Nh4g6 Rh8h6 Ng6xe7+ Nc6xe7 Qd1e2 Ne7f5 Qe2xe4 Rh6g6 a2a4 Nf5h4 Nb1d2 f6f5 Qe4e2
   8/12/2008 11:41:14 PM, Time for this analysis: 00:35:03, Rated time: 2:00:00

0 of 1 matching moves
8/12/2008 11:41:15 PM, Total time: 12:35:15 AM
Rated time: 2:00:00 = 7200 Seconds
Top
Marc MP

Re: Have you decided why Vas

Post by Marc MP »

Dann Corbit wrote:
Dann Corbit wrote:
Marc MP wrote:
Dann Corbit wrote:
Uri Blass wrote:
Dr.Ex wrote:[d] 2kr3r/1ppqb1pp/p1n1pp1n/3pPb2/BP1P1P2/2P1BN2/P5PP/RN1Q1RK1 b - b3 0 11

How long does it like to play Kb8 in this position? I mean, come on. Isn't it stupid to walk directly into an attack?
Here is some analysis
Rybka on slow hardware admits after enough time that white is better here

Mischa - Kapaun, Wertungspartie, 3m + 0s 2008
2kr3r/1ppqb1pp/p1n1pp1n/3pPb2/BP1P1P2/2P1BN2/P5PP/RN1Q1RK1 b - b3 0 1

Analysis by Rybka 3 1-cpu 32-bit :

11...Nh6-g4
³ (-0.52) Depth: 2 00:00:00
11...Nh6-g4
³ (-0.55) Depth: 3 00:00:00
11...Nh6-g4
³ (-0.57) Depth: 4 00:00:00
11...Kc8-b8 12.h2-h3
³ (-0.38) Depth: 5 00:00:00
11...Kc8-b8 12.Qd1-d2 Rh8-f8
³ (-0.50) Depth: 6 00:00:00 15kN
11...Kc8-b8 12.Qd1-d2 Bf5-e4 13.Nb1-a3
³ (-0.49) Depth: 7 00:00:01 22kN
11...Kc8-b8 12.Nb1-a3 Nh6-g4 13.Qd1-d2 Bf5-e4
³ (-0.34) Depth: 8 00:00:01 36kN
11...Kc8-b8 12.Nb1-a3 Nh6-g4 13.Qd1-d2 a6-a5 14.Na3-c2 Bf5xc2
³ (-0.36) Depth: 9 00:00:04 93kN
11...Kc8-b8 12.Nb1-a3 Rh8-f8 13.b4-b5 a6xb5 14.Na3xb5 Nh6-g4
³ (-0.36) Depth: 10 00:00:08 176kN
11...Kc8-b8 12.Nb1-a3 Nh6-g4 13.Qd1-d2 a6-a5 14.Na3-c2 Bf5xc2 15.Ba4xc2 a5xb4 16.Ra1-b1
³ (-0.35) Depth: 11 00:00:12 273kN
11...Kc8-b8 12.Nb1-a3 Nh6-g4 13.Be3-c1 Bf5-e4 14.Na3-c2 h7-h5 15.h2-h3 Ng4-h6 16.Nc2-e3
³ (-0.37) Depth: 12 00:00:24 522kN
11...Kc8-b8 12.b4-b5 a6xb5 13.Ba4xb5
= (-0.19) Depth: 13 00:00:55 1113kN
11...Bf5-e4 12.Nb1-d2 Be4-d3 13.Rf1-e1 Nh6-f5 14.Ra1-c1 Kc8-b8 15.Be3-f2 h7-h5 16.Nd2-b3 Bd3-e4 17.Nb3-a5
= (-0.07) Depth: 14 00:07:18 8748kN
11...Bf5-e4 12.Nb1-d2 Be4-d3 13.Rf1-e1 Kc8-b8 14.Nd2-b3 Bd3-e4 15.Nb3-a5 b7-b5
= (-0.04) Depth: 15 00:08:09 9819kN
11...Bf5-e4 12.Nb1-d2 Be4-d3 13.Rf1-e1 Nh6-f5 14.Be3-f2 Kc8-b8 15.Nd2-b3 Bd3-e4 16.Nb3-d2
= (0.00) Depth: 16 00:10:40 12849kN
11...Nh6-g4 12.Be3-c1 Qd7-e8 13.h2-h3 Ng4-h6 14.Nb1-d2 Bf5-d3 15.Rf1-e1 Rh8-g8 16.Re1-e3 Bd3-g6 17.e5xf6 Be7xf6 18.Qd1-e1 Rd8-d6 19.Bc1-a3 b7-b5 20.Ba4-b3
= (0.14) Depth: 17 00:33:13 36883kN

(so k, 13.08.2008)
Here are my results with Rybka 3 (4 Cores @3GHz -- there are machines that are 4x faster than mine):

Code: Select all

Analysis from C:\tmp\wib.epd   
8/12/2008 9:54:59 PM Level: 1200 Seconds
Analyzing engine: Rybka 3

1) 0 1                  
    Avoid move: 
    Best move (Rybka 3): Qd7-e8
    Not found in: 20:00
      2	00:00	       2.482	149.504	+0.52	Nh6g4
      3	00:00	       2.994	180.344	+0.55	Nh6g4
      4	00:00	       3.626	218.413	+0.57	Nh6g4
      5	00:00	       5.497	175.904	+0.52	Nh6g4 Qd1d2
      6	00:00	       8.832	188.416	+0.47	Nh6g4 Qd1d2 Rh8f8
      7	00:00	      18.635	173.474	+0.36	Nh6g4 Qd1d2 f6xe5 Nf3xe5 Ng4xe5 f4xe5 Rh8f8
      7	00:00	      34.743	174.396	+0.39	Rh8f8 h2h3 Bf5e4 Qd1e2
      7	00:00	      57.166	162.605	+0.43	Kc8b8 Nb1a3 Nh6g4 Qd1d2 h7h5
      8	00:00	      65.750	172.194	+0.43	Kc8b8 Nb1a3 Nh6g4 Qd1d2 h7h5
      9	00:00	     107.460	163.748	+0.39	Kc8b8 Nb1a3 Nh6g4 Qd1d2 h7h5 h2h3 Ng4xe3
     10	00:01	     146.535	165.620	+0.39	Kc8b8 Nb1a3 Nh6g4 Qd1d2 h7h5 h2h3 Ng4xe3
     11	00:02	     277.472	170.138	+0.34	Kc8b8 Nb1a3 Nh6g4 Qd1d2 h7h5 h2h3 Ng4xe3 Qd2xe3 Rh8f8 Na3c2
     12	00:05	     948.745	182.581	+0.31	Kc8b8 Nb1a3 Nh6g4 Be3c1 h7h5 h2h3 Ng4h6 Na3c2 Rh8f8 Nc2e3
     13	00:11	   1.897.822	174.449	+0.24	Kc8b8 h2h3 Rh8g8 b4b5 a6xb5 Ba4xb5 Nh6f7 Qd1a4 g7g5 Bb5xc6 Qd7xc6 Qa4xc6 b7xc6 e5xf6 Be7xf6 Nf3e5 Nf7xe5
     14	00:16	   2.774.285	169.867	+0.23	Kc8b8 h2h3 Rh8g8 b4b5 a6xb5 Ba4xb5 Nh6f7 Qd1a4 g7g5 Bb5xc6 Qd7xc6 Qa4xc6 b7xc6 e5xf6 Be7xf6 f4xg5 Nf7xg5 Nf3xg5 Bf6xg5 Be3xg5 Rg8xg5 Kg1h2 Bf5d3 Rf1f3 Bd3e4
     15	01:40	  16.407.493	168.226	-0.16	Kc8b8 b4b5 a6xb5 Ba4xb5 Nh6g4 Be3c1 h7h5 h2h3 Ng4h6 Qd1a4 Rd8f8 Bc1a3 Nh6g8 Bb5xc6 Qd7xc6 Qa4xc6 b7xc6 Ba3c5 Be7xc5 d4xc5 Ng8e7 Nb1d2
     15	01:47	  17.610.027	167.524	+0.04	Nh6g4 Be3c1 Qd7e8 h2h3 Ng4h6 Nb1d2 Rh8f8 b4b5 a6xb5 Ba4xb5
     16	02:12	  21.242.394	164.816	 0.00	Nh6g4 Be3c1 Qd7e8 h2h3 Ng4h6 Nb1d2 Bf5d3 Rf1e1 Qe8d7 Nd2b3 Bd3e4 Nb3d2 Be4d3 Nd2b3 Bd3e4 Nb3d2 Be4d3 Nd2b3 Bd3e4 Nb3d2 Be4d3 Nd2b3 Bd3e4 Nb3d2 Be4d3 Nd2b3 Bd3e4 Nb3d2 Be4d3 Nd2b3 Bd3e4
     17	03:10	  29.855.281	160.934	 0.00	Nh6g4 Be3c1 Qd7e8 h2h3 Ng4h6 Nb1d2 Bf5d3 Rf1e1 Qe8d7 Nd2b3 Bd3e4 Nb3d2 Be4d3 Nd2b3 Bd3e4 Nb3d2 Be4d3 Nd2b3 Bd3e4 Nb3d2 Be4d3 Nd2b3 Bd3e4 Nb3d2 Be4d3 Nd2b3 Bd3e4 Nb3d2 Be4d3 Nd2b3 Bd3e4
     18	04:41	  43.096.481	156.776	-0.09	Nh6g4 Be3c1 Qd7e8 h2h3 Ng4h6 Nb1d2 Bf5d3 Rf1e1 Rh8g8 Re1e3 Bd3f5 g2g4 b7b5 Ba4b3 f6xe5 f4xe5 Qe8g6 Re3e2
     18	05:25	  49.037.080	154.438	 0.00	Qd7e8 Be3c1 Nh6g4 h2h3 Ng4h6 Nb1d2 Bf5d3 Rf1e1 Rh8g8 Re1e3 Bd3f5 g2g4 b7b5 Ba4b3 f6xe5 f4xe5 Qe8g6 Re3e2
     19	06:18	  57.667.070	156.107	-0.09	Qd7e8 Be3c1 Nh6g4 h2h3 Ng4h6 Nb1d2 Bf5d3 Rf1e1 Rh8g8 Re1e3 Bd3f5 g2g4 b7b5 Ba4b3 f6xe5 f4xe5 Qe8g6 Re3e2
     20	15:01	 139.274.727	158.158	-0.18	Qd7e8 Be3c1 Nh6g4
   8/12/2008 10:15:05 PM, Time for this analysis: 00:19:59, Rated time: 20:00

0 of 1 matching moves
8/12/2008 10:15:06 PM, Total time: 12:20:06 AM
Rated time: 20:00 = 1200 Seconds
It seems clear that white is better. It takes Rybka 1:40 to figure it out and 1:47 to find a better move.
On the other hand, white is not exactly stomping black into the ground. I guess that Rybka will win outright 90% of its games using a top level machine against a fide 2500 GM if playing at G/90 starting from this position.

I have the chess expertize of a box of hammers, but here is what I see looking at the position:
Development is at a similar level, but white has a space advantage. Nothing to get excited about. Actually, for black to have this position against white is really good for black. The field is nearly level. Usually, with experts on both sides, white will average 1/2 pawn better or so through the opening.
Hi Dan,

Out of curiousity would you have cpu time to check if Rybka 3 still want to castle long here : [d]r3k2r/1ppqb1pp/p1n1pp1n/3pPb2/B2P1P2/2P1BN2/PP4PP/RN1Q1RK1 b kq - 3 1

From my chess player experience it seems to me like a strategical mistake, but hard to prove on this forum (or maybe hard to prove at all!). If Rybka 3.0 would change its mind after some lomg pondering then... see my post here: http://64.68.157.89/forum/viewtopic.php ... 28&t=22900
I am quite sure that you are right about this position. Rybka is analyzing it right now, and the eval is taking a slow and inexorable drop into oblivion. It has the clear pattern of "this is a very bad positional move" that I see all the time... A move initially looks like it offers something, but on each new ply the eval stays about the same or drops. When the grinding has stopped I will post the result here.

Meanwhile, I am going to post Dann Corbit's first {probably also last and horribly flawed} law of chess programming:

1. Never, never, never castle opposite sides if you are a computer program unless one of the follwing is true:
A. Your same-side castle shelter is in tatters.
OR
B. You are planning your own pawn storm.

I make this rule because pawn storms are so utterly glacial in chess programming terms that it is very unlikely to see the problem before it is too late.
Thankfully, Rybka finally saw the error of her ways. As is clear from a casual perusal of the evaluation over time, there was nowhere to go but down.

Code: Select all


Analysis from C:\tmp\long-castle-or-not.epd   
8/12/2008 11:05:59 PM Level: 7200 Seconds
Analyzing engine: Rybka 3

1) 3 1                  
    Avoid move: 
    Best move (Rybka 3): Nh6-g4
    Not found in: 2:00:00
      2	00:00	       1.336	41.456	+0.27	00
      3	00:00	       2.658	82.478	+0.33	00
      4	00:00	       3.441	55.056	+0.27	00
      4	00:00	       5.374	85.984	+0.28	Nh6g4
      5	00:00	       6.182	80.131	+0.27	Nh6g4 Be3c1
      6	00:00	      10.657	114.871	+0.27	Nh6g4 Be3c1 00
      7	00:00	      14.219	115.557	+0.27	Nh6g4 Be3c1 00 h2h3
      7	00:00	      59.326	168.749	+0.40	000 Nb1a3 Bf5e4 Qd1d2
      8	00:00	      95.012	182.880	+0.31	000 Nb1a3 Nh6g4 Qd1d2 Rh8f8 Na3c2
      9	00:00	     110.095	184.815	+0.31	000 Nb1a3 Nh6g4 Qd1d2 Rh8f8 Na3c2
     10	00:01	     164.402	185.814	+0.35	000 Nb1a3 Nh6g4 Qd1d2 Be7xa3 b2xa3 h7h5 h2h3
     11	00:01	     305.396	188.958	+0.38	000 Nb1a3 Be7xa3 b2xa3 Rh8f8 Qd1e2 Kc8b8 Ra1d1 Nh6g4
     12	00:02	     445.831	191.177	+0.30	000 Nb1a3 Be7xa3 b2xa3 Kc8b8 Qd1e2 Nh6g4 h2h3 Ng4xe3
     13	00:05	     861.384	185.306	+0.30	000 Nb1a3 Be7xa3 b2xa3 Kc8b8 Qd1e2 Nh6g4 h2h3 Ng4xe3 Qe2xe3 h7h5 Ra1d1 Rh8f8
     14	00:07	   1.367.852	181.717	+0.29	000 Nb1a3 Be7xa3 b2xa3 Nh6g4 Be3c1 h7h5 h2h3 Ng4h6 Nf3h4 Bf5e4 Bc1e3 Nh6f5
     15	00:16	   2.798.305	180.592	+0.28	000 Nb1a3 Be7xa3 b2xa3 Nh6g4 Be3c1 h7h5 Ba4c2 Bf5xc2 Qd1xc2 Ng4h6 Ra1b1 Nh6f5 Qc2b2 Nc6a5
     16	00:29	   5.013.067	173.179	+0.26	000 Nb1a3 Be7xa3 b2xa3 Nh6g4 Qd1d2 h7h5 h2h3 Ng4xe3 Qd2xe3 Kc8b8 Ra1d1 Bf5g6 Rd1d2 Bg6e4 Rd2b2
     17	01:04	  10.639.295	169.545	+0.26	000 h2h3 Kc8b8 Nb1d2 Rh8f8 Rf1e1 Nh6f7 b2b4 Rf8g8 Ba4c2 g7g5 e5xf6 Be7xf6 Bc2xf5 g5xf4 Be3xf4 e6xf5 Nd2b3
     18	02:20	  22.535.611	164.322	+0.15	000 h2h3 Kc8b8 b2b4 Bf5e4 Nb1d2 Be4d3 Rf1e1 Nh6f5 Be3f2 Rh8g8 g2g4 Nf5h6 Re1e3
     19	05:17	  50.537.566	162.940	+0.07	000 h2h3 Kc8b8 b2b4 Bf5g6 Nb1d2 Nh6f5 Be3f2 h7h5 Ba4c2 f6xe5 f4xe5 Nc6a7 a2a4 Qd7c6 Ra1c1 Bg6h7 Nd2b3 Qc6xa4 Nb3c5
     20	17:27	 163.126.235	159.459	 0.00	000 h2h3 Kc8b8 b2b4 Bf5e4 Nb1d2 Be4d3 Rf1e1 Nh6f5 Be3f2 Rh8f8 Nd2b3 Bd3e4 Nb3d2
     20	34:40	 336.204.380	165.468	+0.04	Nh6g4 Be3c1 h7h5 Nf3h4 Bf5e4 Ba4c2 000 Bc2xe4 d5xe4 Nh4g6 Rh8h6 Ng6xe7+ Nc6xe7 Qd1e2 Ne7f5 Qe2xe4 Rh6g6 a2a4 Nf5h4 Nb1d2 f6f5 Qe4e2
   8/12/2008 11:41:14 PM, Time for this analysis: 00:35:03, Rated time: 2:00:00

0 of 1 matching moves
8/12/2008 11:41:15 PM, Total time: 12:35:15 AM
Rated time: 2:00:00 = 7200 Seconds
Thank you for investigating Dan,

I was curious about if Rybka search would be sufficient to change its move here (even if it likes castling long in the first place). Apparently it does. That is what I wanted to know.

Have a good day,
Top
PauloSoare
Posts: 1335
Joined: Thu Mar 09, 2006 5:30 am
Location: Cabo Frio, Brasil

Re: Have you decided why Vas

Post by PauloSoare »

I do not understand the reason for this discussion on Kn/s. It would be better if people
try to understand the reason for Rybka 3 be much stronger than Rybka 2.3.2a.
It is a very big difference and never seen before, from one version to another,
without quoting Rybka 2.3.2a is already very strong.

I think some people are not understanding that this is a historical moment,
and discussing things without importance.
I compare Rybka 3 to the phenomenon of swimming,
Michael Phelps.

Paulo Soares
Top
gerold
Posts: 10121
Joined: Thu Mar 09, 2006 12:57 am
Location: van buren,missouri

Re: Have you decided why Vas

Post by gerold »

PauloSoare wrote:I do not understand the reason for this discussion on Kn/s. It would be better if people
try to understand the reason for Rybka 3 be much stronger than Rybka 2.3.2a.
It is a very big difference and never seen before, from one version to another,
without quoting Rybka 2.3.2a is already very strong.

I think some people are not understanding that this is a historical moment,
and discussing things without importance.
I compare Rybka 3 to the phenomenon of swimming,
Michael Phelps.

Paulo Soares
I agree Paulo. Lets see who can come close to Rybka.
Great program.

Best to you,

Gerold.
Top
Dann Corbit
Posts: 12777
Joined: Wed Mar 08, 2006 8:57 pm
Location: Redmond, WA USA

Re: Have you decided why Vas

Post by Dann Corbit »

Marc MP wrote:
Dann Corbit wrote:
Dann Corbit wrote:
Marc MP wrote:
Dann Corbit wrote:
Uri Blass wrote:
Dr.Ex wrote:[d] 2kr3r/1ppqb1pp/p1n1pp1n/3pPb2/BP1P1P2/2P1BN2/P5PP/RN1Q1RK1 b - b3 0 11

How long does it like to play Kb8 in this position? I mean, come on. Isn't it stupid to walk directly into an attack?
Here is some analysis
Rybka on slow hardware admits after enough time that white is better here

Mischa - Kapaun, Wertungspartie, 3m + 0s 2008
2kr3r/1ppqb1pp/p1n1pp1n/3pPb2/BP1P1P2/2P1BN2/P5PP/RN1Q1RK1 b - b3 0 1

Analysis by Rybka 3 1-cpu 32-bit :

11...Nh6-g4
³ (-0.52) Depth: 2 00:00:00
11...Nh6-g4
³ (-0.55) Depth: 3 00:00:00
11...Nh6-g4
³ (-0.57) Depth: 4 00:00:00
11...Kc8-b8 12.h2-h3
³ (-0.38) Depth: 5 00:00:00
11...Kc8-b8 12.Qd1-d2 Rh8-f8
³ (-0.50) Depth: 6 00:00:00 15kN
11...Kc8-b8 12.Qd1-d2 Bf5-e4 13.Nb1-a3
³ (-0.49) Depth: 7 00:00:01 22kN
11...Kc8-b8 12.Nb1-a3 Nh6-g4 13.Qd1-d2 Bf5-e4
³ (-0.34) Depth: 8 00:00:01 36kN
11...Kc8-b8 12.Nb1-a3 Nh6-g4 13.Qd1-d2 a6-a5 14.Na3-c2 Bf5xc2
³ (-0.36) Depth: 9 00:00:04 93kN
11...Kc8-b8 12.Nb1-a3 Rh8-f8 13.b4-b5 a6xb5 14.Na3xb5 Nh6-g4
³ (-0.36) Depth: 10 00:00:08 176kN
11...Kc8-b8 12.Nb1-a3 Nh6-g4 13.Qd1-d2 a6-a5 14.Na3-c2 Bf5xc2 15.Ba4xc2 a5xb4 16.Ra1-b1
³ (-0.35) Depth: 11 00:00:12 273kN
11...Kc8-b8 12.Nb1-a3 Nh6-g4 13.Be3-c1 Bf5-e4 14.Na3-c2 h7-h5 15.h2-h3 Ng4-h6 16.Nc2-e3
³ (-0.37) Depth: 12 00:00:24 522kN
11...Kc8-b8 12.b4-b5 a6xb5 13.Ba4xb5
= (-0.19) Depth: 13 00:00:55 1113kN
11...Bf5-e4 12.Nb1-d2 Be4-d3 13.Rf1-e1 Nh6-f5 14.Ra1-c1 Kc8-b8 15.Be3-f2 h7-h5 16.Nd2-b3 Bd3-e4 17.Nb3-a5
= (-0.07) Depth: 14 00:07:18 8748kN
11...Bf5-e4 12.Nb1-d2 Be4-d3 13.Rf1-e1 Kc8-b8 14.Nd2-b3 Bd3-e4 15.Nb3-a5 b7-b5
= (-0.04) Depth: 15 00:08:09 9819kN
11...Bf5-e4 12.Nb1-d2 Be4-d3 13.Rf1-e1 Nh6-f5 14.Be3-f2 Kc8-b8 15.Nd2-b3 Bd3-e4 16.Nb3-d2
= (0.00) Depth: 16 00:10:40 12849kN
11...Nh6-g4 12.Be3-c1 Qd7-e8 13.h2-h3 Ng4-h6 14.Nb1-d2 Bf5-d3 15.Rf1-e1 Rh8-g8 16.Re1-e3 Bd3-g6 17.e5xf6 Be7xf6 18.Qd1-e1 Rd8-d6 19.Bc1-a3 b7-b5 20.Ba4-b3
= (0.14) Depth: 17 00:33:13 36883kN

(so k, 13.08.2008)
Here are my results with Rybka 3 (4 Cores @3GHz -- there are machines that are 4x faster than mine):

Code: Select all

Analysis from C:\tmp\wib.epd   
8/12/2008 9:54:59 PM Level: 1200 Seconds
Analyzing engine: Rybka 3

1) 0 1                  
    Avoid move: 
    Best move (Rybka 3): Qd7-e8
    Not found in: 20:00
      2	00:00	       2.482	149.504	+0.52	Nh6g4
      3	00:00	       2.994	180.344	+0.55	Nh6g4
      4	00:00	       3.626	218.413	+0.57	Nh6g4
      5	00:00	       5.497	175.904	+0.52	Nh6g4 Qd1d2
      6	00:00	       8.832	188.416	+0.47	Nh6g4 Qd1d2 Rh8f8
      7	00:00	      18.635	173.474	+0.36	Nh6g4 Qd1d2 f6xe5 Nf3xe5 Ng4xe5 f4xe5 Rh8f8
      7	00:00	      34.743	174.396	+0.39	Rh8f8 h2h3 Bf5e4 Qd1e2
      7	00:00	      57.166	162.605	+0.43	Kc8b8 Nb1a3 Nh6g4 Qd1d2 h7h5
      8	00:00	      65.750	172.194	+0.43	Kc8b8 Nb1a3 Nh6g4 Qd1d2 h7h5
      9	00:00	     107.460	163.748	+0.39	Kc8b8 Nb1a3 Nh6g4 Qd1d2 h7h5 h2h3 Ng4xe3
     10	00:01	     146.535	165.620	+0.39	Kc8b8 Nb1a3 Nh6g4 Qd1d2 h7h5 h2h3 Ng4xe3
     11	00:02	     277.472	170.138	+0.34	Kc8b8 Nb1a3 Nh6g4 Qd1d2 h7h5 h2h3 Ng4xe3 Qd2xe3 Rh8f8 Na3c2
     12	00:05	     948.745	182.581	+0.31	Kc8b8 Nb1a3 Nh6g4 Be3c1 h7h5 h2h3 Ng4h6 Na3c2 Rh8f8 Nc2e3
     13	00:11	   1.897.822	174.449	+0.24	Kc8b8 h2h3 Rh8g8 b4b5 a6xb5 Ba4xb5 Nh6f7 Qd1a4 g7g5 Bb5xc6 Qd7xc6 Qa4xc6 b7xc6 e5xf6 Be7xf6 Nf3e5 Nf7xe5
     14	00:16	   2.774.285	169.867	+0.23	Kc8b8 h2h3 Rh8g8 b4b5 a6xb5 Ba4xb5 Nh6f7 Qd1a4 g7g5 Bb5xc6 Qd7xc6 Qa4xc6 b7xc6 e5xf6 Be7xf6 f4xg5 Nf7xg5 Nf3xg5 Bf6xg5 Be3xg5 Rg8xg5 Kg1h2 Bf5d3 Rf1f3 Bd3e4
     15	01:40	  16.407.493	168.226	-0.16	Kc8b8 b4b5 a6xb5 Ba4xb5 Nh6g4 Be3c1 h7h5 h2h3 Ng4h6 Qd1a4 Rd8f8 Bc1a3 Nh6g8 Bb5xc6 Qd7xc6 Qa4xc6 b7xc6 Ba3c5 Be7xc5 d4xc5 Ng8e7 Nb1d2
     15	01:47	  17.610.027	167.524	+0.04	Nh6g4 Be3c1 Qd7e8 h2h3 Ng4h6 Nb1d2 Rh8f8 b4b5 a6xb5 Ba4xb5
     16	02:12	  21.242.394	164.816	 0.00	Nh6g4 Be3c1 Qd7e8 h2h3 Ng4h6 Nb1d2 Bf5d3 Rf1e1 Qe8d7 Nd2b3 Bd3e4 Nb3d2 Be4d3 Nd2b3 Bd3e4 Nb3d2 Be4d3 Nd2b3 Bd3e4 Nb3d2 Be4d3 Nd2b3 Bd3e4 Nb3d2 Be4d3 Nd2b3 Bd3e4 Nb3d2 Be4d3 Nd2b3 Bd3e4
     17	03:10	  29.855.281	160.934	 0.00	Nh6g4 Be3c1 Qd7e8 h2h3 Ng4h6 Nb1d2 Bf5d3 Rf1e1 Qe8d7 Nd2b3 Bd3e4 Nb3d2 Be4d3 Nd2b3 Bd3e4 Nb3d2 Be4d3 Nd2b3 Bd3e4 Nb3d2 Be4d3 Nd2b3 Bd3e4 Nb3d2 Be4d3 Nd2b3 Bd3e4 Nb3d2 Be4d3 Nd2b3 Bd3e4
     18	04:41	  43.096.481	156.776	-0.09	Nh6g4 Be3c1 Qd7e8 h2h3 Ng4h6 Nb1d2 Bf5d3 Rf1e1 Rh8g8 Re1e3 Bd3f5 g2g4 b7b5 Ba4b3 f6xe5 f4xe5 Qe8g6 Re3e2
     18	05:25	  49.037.080	154.438	 0.00	Qd7e8 Be3c1 Nh6g4 h2h3 Ng4h6 Nb1d2 Bf5d3 Rf1e1 Rh8g8 Re1e3 Bd3f5 g2g4 b7b5 Ba4b3 f6xe5 f4xe5 Qe8g6 Re3e2
     19	06:18	  57.667.070	156.107	-0.09	Qd7e8 Be3c1 Nh6g4 h2h3 Ng4h6 Nb1d2 Bf5d3 Rf1e1 Rh8g8 Re1e3 Bd3f5 g2g4 b7b5 Ba4b3 f6xe5 f4xe5 Qe8g6 Re3e2
     20	15:01	 139.274.727	158.158	-0.18	Qd7e8 Be3c1 Nh6g4
   8/12/2008 10:15:05 PM, Time for this analysis: 00:19:59, Rated time: 20:00

0 of 1 matching moves
8/12/2008 10:15:06 PM, Total time: 12:20:06 AM
Rated time: 20:00 = 1200 Seconds
It seems clear that white is better. It takes Rybka 1:40 to figure it out and 1:47 to find a better move.
On the other hand, white is not exactly stomping black into the ground. I guess that Rybka will win outright 90% of its games using a top level machine against a fide 2500 GM if playing at G/90 starting from this position.

I have the chess expertize of a box of hammers, but here is what I see looking at the position:
Development is at a similar level, but white has a space advantage. Nothing to get excited about. Actually, for black to have this position against white is really good for black. The field is nearly level. Usually, with experts on both sides, white will average 1/2 pawn better or so through the opening.
Hi Dan,

Out of curiousity would you have cpu time to check if Rybka 3 still want to castle long here : [d]r3k2r/1ppqb1pp/p1n1pp1n/3pPb2/B2P1P2/2P1BN2/PP4PP/RN1Q1RK1 b kq - 3 1

From my chess player experience it seems to me like a strategical mistake, but hard to prove on this forum (or maybe hard to prove at all!). If Rybka 3.0 would change its mind after some lomg pondering then... see my post here: http://64.68.157.89/forum/viewtopic.php ... 28&t=22900
I am quite sure that you are right about this position. Rybka is analyzing it right now, and the eval is taking a slow and inexorable drop into oblivion. It has the clear pattern of "this is a very bad positional move" that I see all the time... A move initially looks like it offers something, but on each new ply the eval stays about the same or drops. When the grinding has stopped I will post the result here.

Meanwhile, I am going to post Dann Corbit's first {probably also last and horribly flawed} law of chess programming:

1. Never, never, never castle opposite sides if you are a computer program unless one of the follwing is true:
A. Your same-side castle shelter is in tatters.
OR
B. You are planning your own pawn storm.

I make this rule because pawn storms are so utterly glacial in chess programming terms that it is very unlikely to see the problem before it is too late.
Thankfully, Rybka finally saw the error of her ways. As is clear from a casual perusal of the evaluation over time, there was nowhere to go but down.

Code: Select all


Analysis from C:\tmp\long-castle-or-not.epd   
8/12/2008 11:05:59 PM Level: 7200 Seconds
Analyzing engine: Rybka 3

1) 3 1                  
    Avoid move: 
    Best move (Rybka 3): Nh6-g4
    Not found in: 2:00:00
      2	00:00	       1.336	41.456	+0.27	00
      3	00:00	       2.658	82.478	+0.33	00
      4	00:00	       3.441	55.056	+0.27	00
      4	00:00	       5.374	85.984	+0.28	Nh6g4
      5	00:00	       6.182	80.131	+0.27	Nh6g4 Be3c1
      6	00:00	      10.657	114.871	+0.27	Nh6g4 Be3c1 00
      7	00:00	      14.219	115.557	+0.27	Nh6g4 Be3c1 00 h2h3
      7	00:00	      59.326	168.749	+0.40	000 Nb1a3 Bf5e4 Qd1d2
      8	00:00	      95.012	182.880	+0.31	000 Nb1a3 Nh6g4 Qd1d2 Rh8f8 Na3c2
      9	00:00	     110.095	184.815	+0.31	000 Nb1a3 Nh6g4 Qd1d2 Rh8f8 Na3c2
     10	00:01	     164.402	185.814	+0.35	000 Nb1a3 Nh6g4 Qd1d2 Be7xa3 b2xa3 h7h5 h2h3
     11	00:01	     305.396	188.958	+0.38	000 Nb1a3 Be7xa3 b2xa3 Rh8f8 Qd1e2 Kc8b8 Ra1d1 Nh6g4
     12	00:02	     445.831	191.177	+0.30	000 Nb1a3 Be7xa3 b2xa3 Kc8b8 Qd1e2 Nh6g4 h2h3 Ng4xe3
     13	00:05	     861.384	185.306	+0.30	000 Nb1a3 Be7xa3 b2xa3 Kc8b8 Qd1e2 Nh6g4 h2h3 Ng4xe3 Qe2xe3 h7h5 Ra1d1 Rh8f8
     14	00:07	   1.367.852	181.717	+0.29	000 Nb1a3 Be7xa3 b2xa3 Nh6g4 Be3c1 h7h5 h2h3 Ng4h6 Nf3h4 Bf5e4 Bc1e3 Nh6f5
     15	00:16	   2.798.305	180.592	+0.28	000 Nb1a3 Be7xa3 b2xa3 Nh6g4 Be3c1 h7h5 Ba4c2 Bf5xc2 Qd1xc2 Ng4h6 Ra1b1 Nh6f5 Qc2b2 Nc6a5
     16	00:29	   5.013.067	173.179	+0.26	000 Nb1a3 Be7xa3 b2xa3 Nh6g4 Qd1d2 h7h5 h2h3 Ng4xe3 Qd2xe3 Kc8b8 Ra1d1 Bf5g6 Rd1d2 Bg6e4 Rd2b2
     17	01:04	  10.639.295	169.545	+0.26	000 h2h3 Kc8b8 Nb1d2 Rh8f8 Rf1e1 Nh6f7 b2b4 Rf8g8 Ba4c2 g7g5 e5xf6 Be7xf6 Bc2xf5 g5xf4 Be3xf4 e6xf5 Nd2b3
     18	02:20	  22.535.611	164.322	+0.15	000 h2h3 Kc8b8 b2b4 Bf5e4 Nb1d2 Be4d3 Rf1e1 Nh6f5 Be3f2 Rh8g8 g2g4 Nf5h6 Re1e3
     19	05:17	  50.537.566	162.940	+0.07	000 h2h3 Kc8b8 b2b4 Bf5g6 Nb1d2 Nh6f5 Be3f2 h7h5 Ba4c2 f6xe5 f4xe5 Nc6a7 a2a4 Qd7c6 Ra1c1 Bg6h7 Nd2b3 Qc6xa4 Nb3c5
     20	17:27	 163.126.235	159.459	 0.00	000 h2h3 Kc8b8 b2b4 Bf5e4 Nb1d2 Be4d3 Rf1e1 Nh6f5 Be3f2 Rh8f8 Nd2b3 Bd3e4 Nb3d2
     20	34:40	 336.204.380	165.468	+0.04	Nh6g4 Be3c1 h7h5 Nf3h4 Bf5e4 Ba4c2 000 Bc2xe4 d5xe4 Nh4g6 Rh8h6 Ng6xe7+ Nc6xe7 Qd1e2 Ne7f5 Qe2xe4 Rh6g6 a2a4 Nf5h4 Nb1d2 f6f5 Qe4e2
   8/12/2008 11:41:14 PM, Time for this analysis: 00:35:03, Rated time: 2:00:00

0 of 1 matching moves
8/12/2008 11:41:15 PM, Total time: 12:35:15 AM
Rated time: 2:00:00 = 7200 Seconds
Thank you for investigating Dan,

I was curious about if Rybka search would be sufficient to change its move here (even if it likes castling long in the first place). Apparently it does. That is what I wanted to know.

Have a good day,
Here's a longer look:

Code: Select all

    Searching move: 
    Best move (Rybka 3 Dynamic): Nh6-g4
    Not found in: 3:20:00
      2	00:00	         697	713.728	+0.34	00
      3	00:00	       1.081	1.106.944	+0.33	00
      4	00:00	       2.028	2.076.672	+0.34	00
      4	00:00	       2.621	167.744	+0.36	Nh6g4
      5	00:00	       3.099	198.336	+0.34	Nh6g4 Be3c1
      6	00:00	       5.752	184.064	+0.34	Nh6g4 Be3c1 00
      6	00:00	       9.983	212.970	+0.40	00 h2h3 Bf5e4
      7	00:00	      14.179	154.460	+0.40	00 h2h3 Bf5e4 g2g4
      8	00:00	      29.854	162.609	+0.40	00 h2h3 b7b5 Ba4b3 Bf5e4 Nb1d2
      9	00:00	      45.672	165.844	+0.40	00 h2h3 b7b5 Ba4b3 Bf5e4 Nb1d2
     10	00:00	     110.339	157.363	+0.32	00 Be3f2 Bf5e4 e5xf6 Be7xf6 Ba4c2 Nh6g4 Nb1d2
     10	00:01	     177.153	159.126	+0.33	Nh6g4 Be3c1 000 h2h3 Ng4h6 b2b4 Kc8b8
     10	00:02	     329.160	172.762	+0.37	000 Nb1a3 Kc8b8 Na3c2 Rh8f8 Qd1d2 Bf5e4
     11	00:02	     481.727	177.569	+0.29	000 Nb1a3 Kc8b8 Na3c2 Nh6g4 Qd1d2 Rh8f8 b2b4 h7h5
     11	00:03	     581.136	178.222	+0.34	Nh6g4 Be3c1 000 h2h3 Ng4h6 b2b4 Kc8b8 Bc1e3 Bf5e4
     12	00:04	     694.022	175.173	+0.34	Nh6g4 Be3c1 000 h2h3 Ng4h6 b2b4 Kc8b8 b4b5 a6xb5
     13	00:06	   1.078.249	166.509	+0.24	Nh6g4 Be3c1 000 Nf3h4 Ng4h6 Nh4xf5 Nh6xf5 Ba4c2 Kc8b8 b2b4
     13	00:10	   1.882.202	177.752	+0.29	000 Nb1a3 Nh6g4 Qd1d2 Be7xa3 b2xa3 Bf5e4 h2h3 Ng4h6 Nf3h4 Kc8b8
     14	00:14	   2.354.397	173.633	+0.27	000 Nb1a3 Nh6g4 Qd1d2 Be7xa3 b2xa3 Bf5e4 h2h3 Ng4h6 Nf3h4 Kc8b8 Ra1d1
     15	00:17	   2.912.636	172.082	+0.27	000 Nb1a3 Nh6g4 Qd1d2 Be7xa3 b2xa3 Bf5e4 h2h3 Ng4h6 Nf3h4 Kc8b8 Ra1d1 Rh8e8
     16	00:29	   4.948.648	171.770	+0.27	000 Nb1a3 Nh6g4 Qd1d2 Be7xa3 b2xa3 Bf5e4 h2h3 Ng4h6 Nf3h4 Kc8b8 Ra1d1 Rh8e8 Qd2b2
     17	01:02	  10.759.545	175.859	+0.27	000 Nb1a3 Be7xa3 b2xa3 Kc8b8 Nf3h4 Bf5e4 Qd1h5 Rh8f8 Ra1d1 Rd8c8 Qh5e2 Rf8g8 Rd1e1 Rc8d8
     18	03:21	  33.593.742	170.885	+0.14	000 h2h3 Kc8b8 Nb1d2 Rh8g8 b2b4 Bf5d3 Rf1e1 Nh6f5 Be3f2
     19	06:08	  59.788.812	166.287	+0.14	000 h2h3 Kc8b8 Nb1d2 Rh8g8 b2b4 Bf5d3 Rf1e1 Nh6f5 Be3f2 Rd8f8 g2g4 Nf5h6 Nd2b3 Bd3e4 Nb3a5 b7b5 Ba4c2
     20	13:51	 131.779.772	162.223	+0.04	000 h2h3 Kc8b8 b2b4 Rd8g8 b4b5 a6xb5 Ba4xb5 Nh6f7 Qd1a4 g7g5 e5xf6 Be7xf6 Nf3e5 Bf6xe5 f4xe5 g5g4 Bb5xc6 Qd7xc6 Qa4xc6 b7xc6 h3h4 g4g3 Nb1d2 Rg8g4 Rf1f4 Rh8g8 Ra1f1
     20	34:14	 319.552.769	159.290	+0.13	Nh6g4 Be3c1 h7h5 b2b4 00 h2h3 Ng4h6 e5xf6 Be7xf6 Nf3e5 Qd7e8 Ba4xc6 b7xc6 Nb1d2 Bf6xe5 f4xe5 Qe8g6 Kg1h2 Bf5c2 Rf1xf8+ Ra8xf8 Qd1e1 Nh6f5 Nd2f3 Bc2e4 Nf3g5 Be4d3 Ng5f3 Bd3c4
     21	37:14	 358.272.092	164.204	+0.13	Nh6g4 Be3c1 h7h5 b2b4 00 h2h3 Ng4h6 e5xf6 Be7xf6 Nf3e5 Qd7e8 Ba4xc6 b7xc6 Nb1d2 Bf6xe5 f4xe5 Qe8g6 Kg1h2 Bf5c2 Rf1xf8+ Ra8xf8 Qd1e1 Nh6f5 Nd2f3 Bc2e4 Nf3g5 Be4d3 Ng5f3 Bd3c4
     22	49:44	 476.876.605	163.631	+0.10	Nh6g4 Be3c1 h7h5 Ba4c2 Ng4h6 b2b4 00 Nb1d2 f6xe5 f4xe5 a6a5 b4b5 Bf5xc2 Qd1xc2 Nc6d8 Nd2b3 Nh6f5 a2a4 Nd8f7 c3c4 c7c6 c4xd5 Qd7xd5 b5xc6 b7xc6 Ra1b1 g7g5
     23	1:32:29	 905.035.630	166.995	+0.08	Nh6g4 Be3c1 h7h5 Ba4c2 Ng4h6 b2b4 00 Nb1d2 f6xe5 f4xe5 a6a5 b4b5 Bf5xc2 Qd1xc2 Nc6d8 Nd2b3 Nh6f5 a2a4 Nd8f7 c3c4 c7c6 b5b6
     24	2:12:41	1.306.049.349	167.995	+0.08	Nh6g4 Be3c1 h7h5 Ba4c2 Ng4h6 b2b4 00 Nb1d2 f6xe5 f4xe5 a6a5 b4b5 Bf5xc2 Qd1xc2 Nc6d8 Nd2b3 Nh6f5 a2a4 Nd8f7 c3c4 c7c6 b5b6
   8/13/2008 7:17:41 AM, Time for this analysis: 03:20:00, Rated time: 6:40:00
Top
swami
Posts: 6659
Joined: Thu Mar 09, 2006 4:21 am

Re: Academic role model wouldnt oppose Vas

Post by swami »

Ok, This thread has been moved after a heavy clean up. (Mods have so far decided to remove Clone insinuations/ few personal attacks)

1. Please keep this thread free from Clone allegations.

2. Read the sticky thread posted by Chris Whittington:

http://64.68.157.89/forum/viewtopic.php?t=23021

If there are still clone insinuations in this thread, please report it to the moderators via complaint function. Be warned that refusing to comply with rules will result in deletion of posts.

Thanks for your understanding.

Best Regards.
Top

Return to “Computer Chess Club: General Topics”