position and surprising opinion of stockfish

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

Moderator: Ras

Uri Blass
Posts: 11220
Joined: Thu Mar 09, 2006 12:37 am
Location: Tel-Aviv Israel

position and surprising opinion of stockfish

Post by Uri Blass »

[d]r1bq1rk1/ppp2pp1/2np1n1p/2b1p3/2B1P3/2NP1N1P/PPP2PP1/R1BQ1RK1 w - - 2 8

This is a position from game of 2 children.

White played
8.Qe1
I told her after the game that I think that it is better to develop pieces
by 8.Be3


Stockfish prefers 8.Na4 that I also consider to be not bad but
I am surprised that stockfish find 8.Qe1 as better than 8.Be3

Do other programs have the same opinion?
User avatar
AdminX
Posts: 6396
Joined: Mon Mar 13, 2006 2:34 pm
Location: Acworth, GA

Re: position and surprising opinion of stockfish

Post by AdminX »

Komodo agrees that Na4 is best here. However it does not show Be3 or Qe1 in it's 1st six lines.

Code: Select all

Analysis by Komodo 9.02 64-bit:

1. =  (0.13): 8.Na4 Bb6 9.a3 Re8 10.c3 d5 11.exd5 Nxd5 12.Re1 Bf5 13.b4 Qd6 14.Qb3 Be6 15.Qc2 Bf5 16.Qa2 Be6 17.b5 Na5 18.Rxe5 Nxc4 19.dxc4 Ne7 20.Nxb6 axb6 21.Re1 Ng6 
2. =  (0.05): 8.a3 Nd4 9.b4 Nxf3+ 10.Qxf3 Bd4 11.Bd2 Be6 12.Rad1 Bxc3 13.Bxc3 Bxc4 14.dxc4 Qe8 15.Rfe1 Qc6 16.Qd3 a6 17.c5 Rfe8 18.cxd6 Rad8 19.f3 Rxd6 20.Qe3 Qa4 21.Rxd6 cxd6 22.Qb6 Qxc2 
3. =  (0.02): 8.a4 Be6 9.Bxe6 fxe6 10.Nb1 a5 11.Nbd2 Nh5 12.Nb3 Bb6 13.Kh1 Qe8 14.Bd2 Qg6 15.c3 Nf6 16.Qe2 d5 17.Rae1 dxe4 18.dxe4 Rad8 19.Be3 Bxe3 20.Qxe3 
4. =  (0.00): 8.Ne2 Na5 9.Bb3 Nxb3 10.axb3 Re8 11.Bd2 a6 12.Ng3 Bd7 13.Re1 Nh7 14.Be3 Bxe3 15.Rxe3 Qf6 16.Qe1 Nf8 17.Qb4 b6 18.Qc4 Ra7 19.Rxa6 Rxa6 20.Qxa6 Ne6 
5. =  (0.00): 8.Bb3 a5 9.a4 Re8 10.Re1 Bd7 11.Ne2 Nh5 12.Nc3 Nf6 
6. =  (0.00): 8.Nd5 Nxd5 9.Bxd5 Qf6 10.c3 a5 11.Qe2 a4 12.Be3 Bxe3 13.fxe3 Ne7 14.Bc4 a3 15.b4 Qg6 16.Qf2 Bxh3 17.Nh4 Qg5 18.Nf3 Qg6 

"Good decisions come from experience, and experience comes from bad decisions."
__________________________________________________________________
Ted Summers
Ferdy
Posts: 4856
Joined: Sun Aug 10, 2008 3:15 pm
Location: Philippines

Re: position and surprising opinion of stockfish

Post by Ferdy »

Uri Blass wrote:[d]r1bq1rk1/ppp2pp1/2np1n1p/2b1p3/2B1P3/2NP1N1P/PPP2PP1/R1BQ1RK1 w - - 2 8

This is a position from game of 2 children.

White played
8.Qe1
I told her after the game that I think that it is better to develop pieces
by 8.Be3


Stockfish prefers 8.Na4 that I also consider to be not bad but
I am surprised that stockfish find 8.Qe1 as better than 8.Be3

Do other programs have the same opinion?

Code: Select all

Intel(R) Core(TM) i7-2600K CPU @ 3.40GHz
Cores: Physical = 4, Logical = 8
Physical Memory: Total = 12 GB, Available = 8 GB

id name Deuterium v2015.1.35.189
Hash: 256 MB, Threads: 1, MultiPV: 5, AllocTime: 10.0m

FEN: r1bq1rk1/ppp2pp1/2np1n1p/2b1p3/2B1P3/2NP1N1P/PPP2PP1/R1BQ1RK1 w - - 2 8

  +0.08/21  00h:00m:24s:920ms   8. Be3 Bb6 9. Qd2 Be6 10. a3 Qd7 11. Bb5 
  +0.00/21  00h:01m:04s:500ms   8. Bb3 Be6 9. Bxe6 fxe6 10. Na4 Bb6 11. c3 
  +0.00/21  00h:00m:43s:600ms   8. Bd2 Be6 9. Bxe6 fxe6 10. Na4 Bb6 11. c3 
  +0.00/21  00h:00m:34s:448ms   8. Nd5 Nxd5 9. Bxd5 Qf6 10. Be3 Bxe3 11. fxe3 
  +0.00/21  00h:00m:33s:766ms   8. a3 Be6 9. Bxe6 fxe6 10. Be3 Bb6 11. Qb1 

  +0.09/20  00h:00m:07s:509ms   8. Be3 Bb6 9. Qd2 Be6 10. a3 Bxe3 11. Qxe3 
  +0.00/20  00h:01m:11s:780ms   8. Nd5 Nxd5 9. Bxd5 Qf6 10. Be3 Bb6 11. Qe2 
  +0.00/20  00h:00m:41s:445ms   8. a3 Be6 9. Bxe6 fxe6 10. Be3 Bd4 11. Qd2 
  +0.00/20  00h:00m:27s:636ms   8. Bb3 Be6 9. Be3 Bb6 10. Qd2 Qd7 11. Ba4 
  +0.00/20  00h:00m:05s:366ms   8. Na4 Bb6 9. Bd2 Be6 10. Bxe6 fxe6 11. c3 

  +0.10/19  00h:00m:06s:366ms   8. Be3 Bb6 9. a3 Be6 10. Qd2 Bxe3 11. Qxe3 
  +0.00/19  00h:00m:18s:568ms   8. Bb3 Be6 9. Be3 Bb6 10. Qd2 Qd7 11. Ba4 
  +0.00/19  00h:00m:11s:209ms   8. Nd5 Be6 9. c3 Bxd5 10. Bxd5 Nxd5 11. exd5 
  +0.00/19  00h:00m:09s:145ms   8. Na4 Bb6 9. Bd2 Be6 10. Bxe6 fxe6 11. c3 
  +0.00/19  00h:00m:07s:089ms   8. a3 Be6 9. Bxe6 fxe6 10. Be3 Bd4 11. Qd2 

  +0.05/18  00h:00m:05s:928ms   8. Be3 Bb6 9. Bxb6 axb6 10. d4 exd4 11. Nxd4 
  +0.01/18  00h:00m:17s:872ms   8. Bb3 a5 9. Bd2 Be6 10. Be3 Bxe3 11. fxe3 
  +0.00/18  00h:00m:30s:710ms   8. Na4 Bb6 9. Bd2 Be6 10. Bxe6 fxe6 11. c3 
  +0.00/18  00h:00m:03s:322ms   8. Bd2 Be6 9. Bxe6 fxe6 10. Na4 Bb6 11. c3 
  -0.03/18  00h:00m:27s:891ms   8. a3 Be6 9. Bxe6 fxe6 10. Be3 Bb6 11. Qb1 
zullil
Posts: 6442
Joined: Tue Jan 09, 2007 12:31 am
Location: PA USA
Full name: Louis Zulli

Re: position and surprising opinion of stockfish

Post by zullil »

AdminX wrote:Komodo agrees that Na4 is best here. However it does not show Be3 or Qe1 in it's 1st six lines.

Code: Select all

Analysis by Komodo 9.02 64-bit:

1. =  (0.13): 8.Na4 Bb6 9.a3 Re8 10.c3 d5 11.exd5 Nxd5 12.Re1 Bf5 13.b4 Qd6 14.Qb3 Be6 15.Qc2 Bf5 16.Qa2 Be6 17.b5 Na5 18.Rxe5 Nxc4 19.dxc4 Ne7 20.Nxb6 axb6 21.Re1 Ng6 
2. =  (0.05): 8.a3 Nd4 9.b4 Nxf3+ 10.Qxf3 Bd4 11.Bd2 Be6 12.Rad1 Bxc3 13.Bxc3 Bxc4 14.dxc4 Qe8 15.Rfe1 Qc6 16.Qd3 a6 17.c5 Rfe8 18.cxd6 Rad8 19.f3 Rxd6 20.Qe3 Qa4 21.Rxd6 cxd6 22.Qb6 Qxc2 
3. =  (0.02): 8.a4 Be6 9.Bxe6 fxe6 10.Nb1 a5 11.Nbd2 Nh5 12.Nb3 Bb6 13.Kh1 Qe8 14.Bd2 Qg6 15.c3 Nf6 16.Qe2 d5 17.Rae1 dxe4 18.dxe4 Rad8 19.Be3 Bxe3 20.Qxe3 
4. =  (0.00): 8.Ne2 Na5 9.Bb3 Nxb3 10.axb3 Re8 11.Bd2 a6 12.Ng3 Bd7 13.Re1 Nh7 14.Be3 Bxe3 15.Rxe3 Qf6 16.Qe1 Nf8 17.Qb4 b6 18.Qc4 Ra7 19.Rxa6 Rxa6 20.Qxa6 Ne6 
5. =  (0.00): 8.Bb3 a5 9.a4 Re8 10.Re1 Bd7 11.Ne2 Nh5 12.Nc3 Nf6 
6. =  (0.00): 8.Nd5 Nxd5 9.Bxd5 Qf6 10.c3 a5 11.Qe2 a4 12.Be3 Bxe3 13.fxe3 Ne7 14.Bc4 a3 15.b4 Qg6 16.Qf2 Bxh3 17.Nh4 Qg5 18.Nf3 Qg6 

Here are Stockfish's best six:

Code: Select all

info depth 40 seldepth 55 multipv 1 score cp 16 nodes 55130148131 nps 21625958 hashfull 999 tbhits 0 time 2549258 pv c3a4 c6a5 a4c5 a5c4 c5a4 c4a5 c2c4 c7c5 f3h2 f6h7 f2f4 f7f5 a4c3 a5c6 c1e3 b7b6 d1f3 c6d4 e3d4 e5d4 c3d5 c8b7 e4f5 f8f5 f3e4 f5f7 a1e1 b7d5 e4d5 h7f6 d5e6 d8d7 e6d7 f7d7 h2g4 g8f7 g4f6 f7f6 g1f2 a8c8 f2f3 g7g6 b2b3 d7g7 f1g1 g7g8

info depth 40 seldepth 55 multipv 2 score cp 0 nodes 55130148131 nps 21625958 hashfull 999 tbhits 0 time 2549258 pv a2a3 c6d4 a3a4 a7a5 f3d4 c5d4 c4b3 f6d7 b3c4 d7f6

info depth 40 seldepth 55 multipv 3 score cp 0 nodes 55130148131 nps 21625958 hashfull 999 tbhits 0 time 2549258 pv c3d5 f6d5 c4d5 d8f6 c2c3 c5b6 a2a4 a7a5 c1e3 b6e3 f2e3 f6g6 g1h2 c6e7 d5c4 b7b6 d1e1 c8d7 f3h4 g6h5 g2g4 h5g5 h4f3 g5g6 f3h4 g6g5

info depth 40 seldepth 55 multipv 4 score cp 0 nodes 55130148131 nps 21625958 hashfull 999 tbhits 0 time 2549258 pv a2a4 c8e6 c4e6 f7e6 c3e2 a7a5 e2g3 c6d4 f3d2 d4c6 d2f3

info depth 40 seldepth 55 multipv 5 score cp 0 nodes 55130148131 nps 21625958 hashfull 999 tbhits 0 time 2549258 pv c4b3 a7a5 a2a4 f8e8 f1e1 c8d7 b3c4 c6e7 f3h4 e7c6 h4f3

info depth 40 seldepth 55 multipv 6 score cp 0 nodes 55130148131 nps 21625958 hashfull 999 tbhits 0 time 2549258 pv a1b1 a7a6 a2a4 c8d7 c1d2 a6a5 f1e1 c6d4 f3d4 c5d4 d1f3 d4c5 d2e3 c5b4 f3g3 f6h5 g3f3 h5f6
zullil
Posts: 6442
Joined: Tue Jan 09, 2007 12:31 am
Location: PA USA
Full name: Louis Zulli

Re: position and surprising opinion of stockfish

Post by zullil »

Uri Blass wrote:[d]r1bq1rk1/ppp2pp1/2np1n1p/2b1p3/2B1P3/2NP1N1P/PPP2PP1/R1BQ1RK1 w - - 2 8

I am surprised that stockfish find 8.Qe1 as better than 8.Be3
Indeed:

Code: Select all

info depth 38 seldepth 56 multipv 1 score cp -4 nodes 26289298536 nps 21844127 hashfull 998 tbhits 0 time 1203495 pv d1e1 a7a5 c1e3 a5a4 a2a3 c8d7 c4a2 f8e8 e1d2 c6d4 e3d4 c5d4 f3d4 e5d4 c3d5 f6d5 a2d5 c7c6 d5a2 c6c5 c2c3 d8f6 c3d4 f6d4 d2c3 d7b5 f1d1 d4c3 b2c3 b5c6 a1b1 b7b5 f2f4 c5c4 d3c4 c6e4 b1b4 e4g2 g1g2 e8e2 g2g3 e2a2

info depth 38 seldepth 56 multipv 2 score cp -20 nodes 26289298536 nps 21844127 hashfull 998 tbhits 0 time 1203495 pv c1e3 c5e3 f2e3 c6a5 c4b3 c8d7 d1e1 c7c6 c3e2 a5b3 a2b3 a7a5 e2g3 b7b5 e1c3 f8e8 f3h4 g7g6 c3e1 g8g7 h4f3 c6c5 f3d2 d7c6 g3e2 d8e7 e1g3 f6d7 f1f2 e8f8 a1f1 g7h7 f1a1 b5b4 g3f3 a5a4 b3a4 a8a4 a1a4 c6a4
User avatar
Marek Soszynski
Posts: 587
Joined: Wed May 10, 2006 7:28 pm
Location: Birmingham, England

Re: position and surprising opinion of stockfish

Post by Marek Soszynski »

What happens if you force Komodo to look at only two moves? That is, simultaneously 8.Qe1 and 8.Be3 as in the original question. You may be surprised.
Marek Soszynski
Lyudmil Tsvetkov
Posts: 6052
Joined: Tue Jun 12, 2012 12:41 pm

Re: position and surprising opinion of stockfish

Post by Lyudmil Tsvetkov »

I like best a3, followed by Nd5.

Interesting, who has the advantage after Na4 a6?
Henk
Posts: 7261
Joined: Mon May 27, 2013 10:31 am

Re: position and surprising opinion of stockfish

Post by Henk »

Code: Select all

Depth  Value   Time
  1    0.40   0.041         809   f1e1 
  2    0.00   0.075        2250   f1e1 f8e8 
  3    0.28   0.128        4745   f1e1 f8e8 c1d2 
  4    0.10   0.599       24927   d1e2 c6d4 f3d4 e5d4 
  5    0.05   1.401       61571   c3a4 c8e6 a4c5 d6c5 d1e2 
  6    0.00   2.016       90042   c1d2 f8e8 f1e1 c8d7 d1e2 d8e7 
  7    0.10   6.218      285139   c1e3 c5e3 f2e3 c8d7 d1e2 f8e8 a1d1 
  8    0.00   9.457      445553   c1e3 f8e8 e3c5 d6c5 f1e1 c8d7 d1e2 d8e7 
  9    0.23   14.759      719588   c3a4 c8e6 a4c5 d6c5 c4e6 f7e6 c1d2 f8e8 f1e1 
 10    0.13   19.926      975506   c3a4 c8e6 a4c5 d6c5 c4e6 f7e6 c1d2 d8e7 f1e1 a8d8 
 11    0.00   72.001     3389276   a2a3 c8d7 b2b4 c5b6 c1d2 d8e7 d1e2 a7a6 b4b5 a6b5 c3b5 
Lyudmil Tsvetkov
Posts: 6052
Joined: Tue Jun 12, 2012 12:41 pm

Re: position and surprising opinion of stockfish

Post by Lyudmil Tsvetkov »

My main line on Na4 goes like a6 Nc5 dc5 Be3 b5 Bb3 Qd6

[d]r1b2rk1/2p2pp1/p1nq1n1p/1pp1p3/4P3/1B1PBN1P/PPP2PP1/R2Q1RK1 w - - 0 5

here SF gives 50cps white advantage, but I fully can not understand where it comes from.
for me, it is simply 0.0.

With such an eval, SF can not go very far, certainly not past TCEC champion. :)
User avatar
cdani
Posts: 2204
Joined: Sat Jan 18, 2014 10:24 am
Location: Andorra

Re: position and surprising opinion of stockfish

Post by cdani »

Lyudmil Tsvetkov wrote:My main line on Na4 goes like a6 Nc5 dc5 Be3 b5 Bb3 Qd6

[d]r1b2rk1/2p2pp1/p1nq1n1p/1pp1p3/4P3/1B1PBN1P/PPP2PP1/R2Q1RK1 w - - 0 5

here SF gives 50cps white advantage, but I fully can not understand where it comes from.
for me, it is simply 0.0.

With such an eval, SF can not go very far, certainly not past TCEC champion. :)
Blind evaluation of bishop pair. Like Andscacs :-) I have to improve it.