Cfish-54302a... Search of startpos to depth 50

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

Moderators: hgm, Rebel, chrisw

zullil
Posts: 6442
Joined: Tue Jan 09, 2007 12:31 am
Location: PA USA
Full name: Louis Zulli

Cfish-54302a... Search of startpos to depth 50

Post by zullil »

This one might seem particularly entertaining, at least to those who enjoy such things.

Cfish-54302a0a2ea77558c78697fba08b560d2133cb35
20 threads
32 GB hash

Code: Select all

info depth 50 seldepth 73 multipv 1 score cp 9 nodes 971948104778 nps 33406079 hashfull 999 tbhits 0 time 29094947 pv e2e4 e7e5 g1f3 b8c6 f1b5 g8f6 e1g1 f6e4 f1e1 e4d6 f3e5 f8e7 b5f1 c6e5 e1e5 e8g8 d2d4 d6e8 c2c4 e7f6 e5e1 d7d5 c4d5 d8d5 c1e3 c8e6 b1c3 d5d7 c3e4 a8d8 e4c5 d7c8 c5e6 c8e6 d1c2 e6b6 d4d5 b6b2 c2b2 f6b2 a1b1 b2c3 e1c1 c3a5 e3c5 a5b6 c5f8 g8f8 f1c4 e8f6 g2g3 f6d5 c1d1 c7c6 g1g2 f8e7 c4d5 d8d5 d1d5 c6d5
[pgn]
1. e4 e5 2. Nf3 Nc6 3. Bb5 Nf6 4. O-O Nxe4 5. Re1 Nd6 6. Nxe5 Be7 7. Bf1 Nxe5 8.
Rxe5 O-O 9. d4 Ne8 10. c4 Bf6 11. Re1 d5 12. cxd5 Qxd5 13. Be3 Be6 14. Nc3 Qd7
15. Ne4 Rd8 16. Nc5 Qc8 17. Nxe6 Qxe6 18. Qc2 Qb6 19. d5 Qxb2 20. Qxb2 Bxb2 21.
Rab1 Bc3 22. Rec1 Ba5 23. Bc5 Bb6 24. Bxf8 Kxf8 25. Bc4 Nf6 26. g3 Nxd5 27. Rd1
c6 28. Kg2 Ke7 29. Bxd5 Rxd5 30. Rxd5 cxd5
[/pgn]
fastgm
Posts: 818
Joined: Mon Aug 19, 2013 6:57 pm

Re: Cfish-54302a... Search of startpos to depth 50

Post by fastgm »

2 threads, 2 GB Hash

Code: Select all

info depth 50 seldepth 79 multipv 1 score cp 15 nodes 316243494695 nps 2069139 hashfull 999 tbhits 0 time 152838143 pv e2e4 c7c5 g1f3 d7d6 b1c3 a7a6 g2g3 g8f6 f1g2 b8c6 e1g1 g7g6 d2d3 f8g7 f1e1 c8g4 h2h3 g4f3 g2f3 a8b8 f3g2 e8g8 f2f4 f6d7 c3e2 b7b5 e1f1 b5b4 f4f5 c6d4 e2d4 g7d4 g1h2 a6a5 a1b1 a5a4 c1h6 d4g7 h6e3 e7e6 d1d2 d8e7 e3g5 g7f6 g5h6 f8e8 f5g6 h7g6 a2a3 b4a3 b2a3 d7e5 b1b8
20 threads, 32 GB Hash

Code: Select all

info depth 50 seldepth 73 multipv 1 score cp 9 nodes 971948104778 nps 33406079 hashfull 999 tbhits 0 time 29094947 pv e2e4 e7e5 g1f3 b8c6 f1b5 g8f6 e1g1 f6e4 f1e1 e4d6 f3e5 f8e7 b5f1 c6e5 e1e5 e8g8 d2d4 d6e8 c2c4 e7f6 e5e1 d7d5 c4d5 d8d5 c1e3 c8e6 b1c3 d5d7 c3e4 a8d8 e4c5 d7c8 c5e6 c8e6 d1c2 e6b6 d4d5 b6b2 c2b2 f6b2 a1b1 b2c3 e1c1 c3a5 e3c5 a5b6 c5f8 g8f8 f1c4 e8f6 g2g3 f6d5 c1d1 c7c6 g1g2 f8e7 c4d5 d8d5 d1d5 c6d5

Code: Select all

threads  depth     time ms     hh:mm:ss         nodes       cp       nps
---------------------------------------------------------------------------
   2       50    152.838.143   42:27:18   316.243.494.695   15    2.069.139
  20       50     29.094.947   08:04:55   971.948.104.778    9   33.406.079
So 16.14 x faster in nodes per second, 5.25 x faster in time and 3.07 x more total nodes.
zullil
Posts: 6442
Joined: Tue Jan 09, 2007 12:31 am
Location: PA USA
Full name: Louis Zulli

Re: Cfish-54302a... Search of startpos to depth 55

Post by zullil »

Now at depth 55:

Code: Select all

info depth 55 seldepth 66 multipv 1 score cp 11 nodes 2505861687141 nps 33694762 hashfull 999 tbhits 0 time 74369472 pv d2d4 d7d5 c2c4 e7e6 b1c3 g8f6 c4d5 e6d5 c1g5 c7c6 e2e3 f8d6 f1d3 e8g8 d1c2 h7h6 g5h4 f8e8 g1e2 d8e7 e1g1 b8d7 h2h3 d7f8 h4g3 b7b6 f1d1 c8d7 a1b1 a7a5 g3d6 e7d6 e2f4 f8e6 c3e2 a8c8 f4e6 d6e6 e2f4 e6e7 d3a6 c8b8 a6e2 f6e4 e2d3 b8c8 b1c1 e7g5 a2a3 c6c5 d3e4 d5e4 d4c5 c8c5 c2d2 c5c1 d2c1
[pgn]
1. d4 d5 2. c4 e6 3. Nc3 Nf6 4. cxd5 exd5 5. Bg5 c6 6. e3 Bd6 7. Bd3 O-O 8. Qc2
h6 9. Bh4 Re8 10. Nge2 Qe7 11. O-O Nbd7 12. h3 Nf8 13. Bg3 b6 14. Rfd1 Bd7 15.
Rab1 a5 16. Bxd6 Qxd6 17. Nf4 Ne6 18. Nce2 Rac8 19. Nxe6 Qxe6 20. Nf4 Qe7 21.
Ba6 Rb8 22. Be2 Ne4 23. Bd3 Rbc8 24. Rbc1 Qg5 25. a3 c5 26. Bxe4 dxe4 27. dxc5
Rxc5 28. Qd2 Rxc1 29. Qxc1
[/pgn]
fastgm
Posts: 818
Joined: Mon Aug 19, 2013 6:57 pm

Re: Cfish-54302a... Search of startpos to depth 55

Post by fastgm »

2 threads, 2 GB Hash

Code: Select all

info depth 55 seldepth 79 multipv 1 score cp 15 nodes 1845641519551 nps 2087001 hashfull 999 tbhits 0 time 884350926 pv e2e4 c7c5 g1f3 d7d6 f1b5 b8d7 e1g1 a7a6 b5d7 c8d7 d2d4 c5d4 d1d4 g8f6 b1c3 e7e6 f1d1 f8e7 e4e5 d6e5 f3e5 d7b5 d4d8 e7d8 a2a4 b5c6 e5c6 b7c6 c1f4 f6d5 f4g3 e8g8 g3d6 f8e8 a1b1 d5f6 d6c5 d8c7 b2b4 e8d8 g2g3 c7e5 c3e2 f6e4 c5e3 e4c3 e2c3 e5c3 d1d8 a8d8 b1b3 c3f6 g1g2 d8d1 b4b5 c6b5 a4b5 a6b5 b3b5 g8f8 b5b7 f8e8 b7a7 d1a1 a7a1 f6a1
20 threads, 32 GB Hash

Code: Select all

info depth 55 seldepth 66 multipv 1 score cp 11 nodes 2505861687141 nps 33694762 hashfull 999 tbhits 0 time 74369472 pv d2d4 d7d5 c2c4 e7e6 b1c3 g8f6 c4d5 e6d5 c1g5 c7c6 e2e3 f8d6 f1d3 e8g8 d1c2 h7h6 g5h4 f8e8 g1e2 d8e7 e1g1 b8d7 h2h3 d7f8 h4g3 b7b6 f1d1 c8d7 a1b1 a7a5 g3d6 e7d6 e2f4 f8e6 c3e2 a8c8 f4e6 d6e6 e2f4 e6e7 d3a6 c8b8 a6e2 f6e4 e2d3 b8c8 b1c1 e7g5 a2a3 c6c5 d3e4 d5e4 d4c5 c8c5 c2d2 c5c1 d2c1

Code: Select all

threads  depth     time ms     hh:mm:ss         nodes         cp      nps
-----------------------------------------------------------------------------
   2       55    884.350.926  245:39:11   1.845.641.519.551   15    2.087.001
  20       55     74.369.472   20:39:29   2.505.861.687.141   11   33.694.762
So 16.15 x faster in nodes per second, 11.89 x faster in time and 1.36 x more total nodes.
syzygy
Posts: 5557
Joined: Tue Feb 28, 2012 11:56 pm

Re: Cfish-54302a... Search of startpos to depth 55

Post by syzygy »

fastgm wrote:So 16.15 x faster in nodes per second, 11.89 x faster in time and 1.36 x more total nodes.
20 threads achieve 16.15x the nps of 2 threads on the same computer?
fastgm
Posts: 818
Joined: Mon Aug 19, 2013 6:57 pm

Re: Cfish-54302a... Search of startpos to depth 55

Post by fastgm »

syzygy wrote:20 threads achieve 16.15x the nps of 2 threads on the same computer?
No, it's not the same computer and not the same Stockfish version.
It's just a comparison with my test http://talkchess.com/forum/viewtopic.php?t=66340
Lyudmil Tsvetkov
Posts: 6052
Joined: Tue Jun 12, 2012 12:41 pm

Re: Cfish-54302a... Search of startpos to depth 50

Post by Lyudmil Tsvetkov »

1. c4 :D

SF says 1. e4 is best, Komodo 1. d4.

Whom to believe?
Henk
Posts: 7216
Joined: Mon May 27, 2013 10:31 am

Re: Cfish-54302a... Search of startpos to depth 50

Post by Henk »

Lyudmil Tsvetkov wrote:1. c4 :D

SF says 1. e4 is best, Komodo 1. d4.

Whom to believe?
1. c4 Nf6 2. Nc3 e6 is a draw because Nimzo Indian is good for black. I don't think g3 makes black scared to death.
User avatar
Ovyron
Posts: 4556
Joined: Tue Jul 03, 2007 4:30 am

Re: Cfish-54302a... Search of startpos to depth 50

Post by Ovyron »

At least it depends on your opening preparation and the one by your opponent. 1.c4 is good if you can transpose into a position that you have greatly analyzed, but that your opponent has to do it from scratch.

At least, that's what I gather from my correspondence games. Someone is beating me with 1.b4 :shock: - And I don't think 1.b4 is good at all, but this player apparently only plays 1.b4 and has great experience with it, and it shows, I entered his territory, and zero experience does hurt.

1.c4 can be the best move if you make it your pet line, and it will not be drawn because you can find positions that look drawn, but aren't, and positions that look advantageous for black, but are the reverse.

What I can say after seeing this is that chess is full of life, and far from death, the problem is people keep entering the same well known lines over and over, but instead of that you can get into some unexplored terrain, make it your own, and force your opponent to fight in it.
tpoppins
Posts: 919
Joined: Tue Nov 24, 2015 9:11 pm
Location: upstate

Re: Cfish-54302a... Search of startpos to depth 50

Post by tpoppins »

Henk wrote:1. c4 Nf6 2. Nc3 e6 is a draw because Nimzo Indian is good for black.
After 3.e4 there will be no Nimzo-Indian: 3...Bb4? 4.e5 Ng8 5.Qg4+/-