Louguet endgame solution: how many hours on your computer?

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

Moderators: hgm, Rebel, chrisw

ernest
Posts: 2041
Joined: Wed Mar 08, 2006 8:30 pm

Louguet endgame solution: how many hours on your computer?

Post by ernest »

The Louguet LCTII test suite is 16 years old, but I found that at least that endgame position, LCTII.FIN.08, from the Spassky - Byrne 1974 game, holds its rank quite well (c5! wins).
Houdini 1.5 gtb: 13 hours on my dualcore
Deep Fritz 10: more than 2 hours on a friend's Quad
Shredder: does not find it after several hours...

How about you?

[D]2k5/p7/Pp1p1b2/1P1P1p2/2P2P1p/3K3P/5B2/8 w - - bm c5; id "LCTII.FIN.08"; c0 "Spassky - Byrne, 1974";
Top
yanquis1972
Posts: 1766
Joined: Wed Jun 03, 2009 12:14 am

Re: Louguet endgame solution: how many hours on your compute

Post by yanquis1972 »

critter really excels here --

[+1.61] d=34 1.c5 bxc5 2.Be1 Kb8 3.Kc4 Kc8 4.Ba5 Kb8 5.Kd3 Kc8 6.Ke3 Bd4 7.Ke2 Bf6 8.Kd3 Kb8 9.Kc4 Be7 10.b6 axb6 11.Bxb6 Bf8 12.Bd8 Ka7 13.Bxh4 Kxa6 14.Bg5 Kb7 15.h4 Kc7 16.h5 Kd7 17.h6 Ke8 18.Kb5 Kd7 19.h7 Bg7 20.Bh4 Ke8 21.Kc6 c4 22.Kxd6 Kf7 23.Be1 c3 24.Kc5 c2 25.Bd2 Kg6 26.Kc4 Kxh7 27.Kd3 Kg6 28.Kxc2 Kf7 29.Bc3 Bf8 30.Be5 (0:01:14) 539985kN
[+1.60] d=34 1.c5 dxc5 2.Kc4 (0:00:58) 421079kN
[+1.48] d=34 1.c5 dxc5 2.Kc4 (0:00:52) 364591kN
[+1.37] d=34 1.Bd4 Bxd4 2.Kxd4 Kd7 3.Kd3 Ke7 4.Kd2 Kf7 5.Ke2 Ke7 6.Kf2 Kd7 7.Kf3 Ke7 8.Kg2 Kd7 9.Kh1 Ke7 10.Kg1 Kf6 11.Kg2 Kf7 12.Kf3 Kg8 13.Ke3 Kf7 14.Kd2 Ke7 15.Ke1 Kf6 16.Kf1 Ke7 17.Ke2 Kd7 18.Kf2 Ke7 (0:00:28) 191351kN
[+1.37] d=33 1.Bd4 Bxd4 2.Kxd4 Kd7 3.Kd3 Ke7 4.Kd2 Kf7 5.Ke2 Ke7 6.Kd1 Kd8 7.Ke1 Kd7 8.Kf2 Ke7 9.Kf3 Kd7 10.Kg2 Ke7 11.Kf2 Kd7 12.Kf3 Ke7 13.Kg2 Kd7 14.Kf1 Kd8 (0:00:18) 124066kN
[+1.37] d=32 1.Bd4 Bxd4 2.Kxd4 Kd7 3.Kd3 Ke7 4.Kd2 Kf7 5.Ke2 Ke7 6.Kd3 Kd8 7.Ke3 Kd7 8.Kf2 Ke7 9.Kf3 Kd7 10.Kg2 Ke7 11.Kf1 Kf6 12.Kf2 (0:00:14) 99256kN
[+1.37] d=31 1.Bd4 Bxd4 2.Kxd4 Kd7 3.Kd3 Ke7 4.Kd2 Kf7 5.Ke2 Ke7 6.Kd3 Kd8 7.Ke3 Kd7 8.Kf2 Ke7 9.Ke1 Kf6 10.Kf1 Ke7 11.Ke2 Kd7 12.Kd2 Ke7 (0:00:02) 12556kN
[+1.37] d=30 1.Bd4 Bxd4 2.Kxd4 Kd7 3.Kd3 Ke7 4.Kd2 Kf7 5.Ke2 Ke7 6.Kd3 Kd8 7.Ke3 Kd7 8.Kf2 Ke7 9.Kf3 Kd7 10.Kg2 Ke7 11.Kf2 Kd7 12.Kf3 Ke7 13.Ke3 Kd8 14.Kf2 (0:00:01) 8585kN
[+1.37] d=29 1.Bd4 Bxd4 2.Kxd4 Kd7 3.Kd3 Ke7 4.Kd2 Kf7 5.Ke2 Ke7 6.Ke1 Kf6 7.Kf2 Ke7 8.Kf3 Kd7 9.Kg2 Ke7 10.Kf2 Kd7 11.Kf3 Ke7 12.Ke2 Kd7 13.Kd2 Ke7 14.Kd3 Kd8 15.Ke3 Kd7 16.Kf2 (0:00:01) 6366kN
[+1.37] d=28 1.Bd4 Bxd4 2.Kxd4 Kd7 3.Kd3 Ke7 4.Kd2 Kf7 5.Ke2 Ke7 6.Ke1 Kf6 7.Kf2 Ke7 8.Kf3 Kd7 9.Kg2 Ke7 10.Kf2 Kd7 11.Kf3 Ke7 12.Ke3 Kd8 13.Kf2 (0:00:01) 5084kN
[+1.37] d=27 1.Bd4 Bxd4 2.Kxd4 Kd7 3.Kd3 Ke7 4.Kd2 Kf7 5.Ke2 Ke7 6.Kd3 Kd8 7.Ke3 Kd7 8.Kf2 Ke7 9.Kf3 Kd7 10.Kg2 Ke7 11.Kf2 Kd7 12.Kf3 Ke7 13.Ke3 Kd8 14.Kf2 (0:00:00) 4141kN
[+1.37] d=26 1.Bd4 Bxd4 2.Kxd4 Kd7 3.Kd3 Ke7 4.Kd2 Kf7 5.Ke2 Ke7 6.Kf2 Kd7 7.Kf3 Ke7 8.Kg2 Kd7 9.Kf1 Kd8 10.Ke1 Kd7 11.Kd2 Ke7 12.Ke2 Kd7 13.Kf2 Ke7 14.Kf3 Kd7 15.Kg2 Ke7 (0:00:00) 1420kN
[+1.37] d=25 1.Bd4 Bxd4 2.Kxd4 Kd7 3.Kd3 Ke7 4.Kd2 Kf7 5.Ke2 Ke7 6.Ke1 Kf6 7.Kf2 Ke7 8.Kf3 Kd7 9.Kf2 (0:00:00) 994kN
[+1.37] d=24 1.Bd4 Bxd4 2.Kxd4 Kd7 3.Kd3 Ke7 4.Kd2 Kf7 5.Ke2 Ke7 6.Ke1 Kf6 7.Kf2 Ke7 8.Kf3 Kd7 9.Kf2 (0:00:00) 652kN
[+1.37] d=23 1.Bd4 Bxd4 2.Kxd4 Kd7 3.Kd3 Ke7 4.Kd2 Kf7 5.Ke2 Ke7 6.Kf2 Kd7 7.Kf3 Ke7 (0:00:00) 465kN
[+1.37] d=22 1.Bd4 Bxd4 2.Kxd4 Kd7 3.Kd3 Ke7 4.Kd2 Kf7 5.Ke2 Ke7 6.Kd3 Kd8 7.Ke3 Kd7 8.Kf3 Ke7 9.Ke3 Kd8 10.Kd2 Ke7 (0:00:00) 346kN
[+1.37] d=21 1.Bd4 Bxd4 2.Kxd4 Kd7 3.Kd3 Ke7 4.Kd2 Kf7 5.Ke2 Ke7 6.Ke1 Kf6 7.Kf2 Ke7 8.Kf3 Kd7 9.Kf2 (0:00:00) 243kN
[+1.37] d=20 1.Bd4 Bxd4 2.Kxd4 Kd7 3.Kd3 Ke7 4.Kd2 Kf7 5.Ke2 Ke7 6.Ke1 Kf6 7.Kf2 Ke7 8.Kf1 Kf6 (0:00:00) 174kN
[+1.37] d=19 1.Bd4 Bxd4 2.Kxd4 Kd7 3.Kd3 Ke7 4.Kd2 Kf7 5.Ke2 Ke7 6.Ke1 Kf6 7.Kf2 Ke7 8.Kf3 Kd7 9.Ke3 Ke7 10.Kf2 Kd7 11.Kf3 Ke7 12.Ke3 Kd8 13.Kf2 (0:00:00) 120kN
[+1.37] d=18 1.Bd4 Bxd4 2.Kxd4 Kd7 3.Kd3 Ke7 4.Kd2 Kf7 5.Ke2 Ke7 6.Kd3 Kd8 7.Ke3 Kd7 8.Kf2 Ke7 9.Kf3 Kd7 (0:00:00) 83kN
Top
zullil
Posts: 6442
Joined: Tue Jan 09, 2007 12:31 am
Location: PA USA
Full name: Louis Zulli

Re: Louguet endgame solution: how many hours on your compute

Post by zullil »

yanquis1972 wrote:critter really excels here --

[+1.48] d=34 1.c5 dxc5 2.Kc4 (0:00:52) 364591kN
Wow, 0.0144444444444... hours! :D
Top
zullil
Posts: 6442
Joined: Tue Jan 09, 2007 12:31 am
Location: PA USA
Full name: Louis Zulli

Re: Louguet endgame solution: how many hours on your compute

Post by zullil »

Stockfish-2.0.1 (8 threads) takes about .5 hours:

Code: Select all

info depth 53 seldepth 26 multipv 1 score cp 129 upperbound time 80590 nodes 424058207 nps 5261920 pv f2g1 c8c7 g1e3 c7b8 e3d2 b8c7 d3e3 c7d7 d2b4 d7c7 e3e2 f6d4 e2d3 d4h8 d3e2 h8g7 b4a3 g7f6 e2d3 f6a1 d3e3 a1c3 a3c1 c7c8 c1a3 c8c7 
info nodes 424058207 nps 5261920 time 80590
info currmove d3d2 currmovenumber 2
info currmove d3e2 currmovenumber 3
info currmove c4c5 currmovenumber 4
info depth 53 seldepth 13 multipv 1 score cp 153 lowerbound time 1746535 nodes 16207531675 nps 9279820 pv c4c5 d6c5 d5d6 f6b2 f2c5 b6c5 d6d7 c8d8 b5b6 a7b6 a6a7 c5c4 d3c4 
info depth 53 seldepth 11 multipv 1 score cp 169 lowerbound time 1981499 nodes 18005258002 nps 9086685 pv c4c5 b6c5 f2e1 f6d8 e1c3 c8c7 c3a5 c7d7 a5d8 d7d8 b5b6
Top
zullil
Posts: 6442
Joined: Tue Jan 09, 2007 12:31 am
Location: PA USA
Full name: Louis Zulli

Re: Louguet endgame solution: how many hours on your compute

Post by zullil »

Crafty-23.4 (8 threads) takes about 2/3 of an hour:

Code: Select all

               42    38:27     +1   1. c5!                            
               42    48:29   2.15   1. c5 bxc5 2. Be1 Kc7 3. Ba5+ Kb8 4.
                                    b6 axb6 5. Bxb6 Be7 6. Kc4 Bf8 7. Bd8
                                    Ka7 8. Bxh4 Kxa6 9. Bg5 Ka5 10. h4
                                    Kb6 11. h5 Kc7 12. h6 Kd7 13. h7 Bg7
                                    14. Bh4 Ke8 15. Kb5 Bh8 16. Be1 Ke7
                                    17. Kc6 Kf7 18. Kxd6 c4 19. Bb4 c3
                                    20. Ba3 c2 21. Bc1 Kf6 22. Kc5
               42->  48:30   2.15   1. c5 bxc5 2. Be1 Kc7 3. Ba5+ Kb8 4.
                                    b6 axb6 5. Bxb6 Be7 6. Kc4 Bf8 7. Bd8
                                    Ka7 8. Bxh4 Kxa6 9. Bg5 Ka5 10. h4
                                    Kb6 11. h5 Kc7 12. h6 Kd7 13. h7 Bg7
                                    14. Bh4 Ke8 15. Kb5 Bh8 16. Be1 Ke7
                                    17. Kc6 Kf7 18. Kxd6 c4 19. Bb4 c3
                                    20. Ba3 c2 21. Bc1 Kf6 22. Kc5
               43    60:09   1.93   1. c5 bxc5 2. Be1 Kc7 3. Ba5+ Kb8 4.
                                    b6 axb6 5. Bxb6 Be7 6. Kc4 Bf8 7. Bd8
                                    Ka7 8. Bxh4 Kxa6 9. Bg5 Ka5 10. h4
                                    Kb6 11. h5 Kc7 12. h6 Kd7 13. h7 Bg7
                                    14. Bh4 Ke8 15. Kb5 Bh8 16. Be1 Kf7
                                    17. Ba5 Ke7 18. Kc6 c4 19. Bc7 Kf6
                                    20. Bxd6 Kg6 21. Ba3 Kxh7 22. d6 c3
               43->  71:14   1.93   1. c5 bxc5 2. Be1 Kc7 3. Ba5+ Kb8 4.
                                    b6 axb6 5. Bxb6 Be7 6. Kc4 Bf8 7. Bd8
                                    Ka7 8. Bxh4 Kxa6 9. Bg5 Ka5 10. h4
                                    Kb6 11. h5 Kc7 12. h6 Kd7 13. h7 Bg7
                                    14. Bh4 Ke8 15. Kb5 Bh8 16. Be1 Kf7
                                    17. Ba5 Ke7 18. Kc6 c4 19. Bc7 Kf6
                                    20. Bxd6 Kg6 21. Ba3 Kxh7 22. d6 c3
Top
Robert Flesher
Posts: 1280
Joined: Tue Aug 18, 2009 3:06 am

Re: Louguet endgame solution: how many hours on your compute

Post by Robert Flesher »

Heya John, you may have had some SMP luck here. Try clearing the hash table and see if you can reproduce C5 again at 34 plies.
Top
ernest
Posts: 2041
Joined: Wed Mar 08, 2006 8:30 pm

Re: Louguet endgame solution: how many hours on your compute

Post by ernest »

Indeed, in my 1st Critter 0.90 (with gtb) try, it only gets it at depth 38...
But still, pretty good! :)
Top
kasinp
Posts: 251
Joined: Sat Dec 02, 2006 10:47 pm
Location: Toronto
Full name: Peter Kasinski

Re: Louguet endgame solution: how many hours on your compute

Post by kasinp »

Analysis by Zappa Mexico II:

1.Bd4 Bxd4 2.Kxd4 Kd7 3.Ke3 Ke8 4.Kd3 Ke7 5.Kd2 Kf7
+/- (1.30) Depth: 19/31 00:00:01 2298kN
+/- (1.19) Depth: 24/65 00:00:09 53271kN, tb=1
1.Ke2 Kc7 2.Kf3 Kc8 3.Be3 Kb8 4.Ke2 Bg7 5.Bf2 Bf6 6.Kd1 Kc7 7.Kd2 Kc8 8.Be1 Kd8 9.Kd3 Kd7 10.Bf2 Kc8 11.Ke2
+/- (1.30) Depth: 24/65 00:00:11 67792kN, tb=11
+/- (1.30) Depth: 24/65 00:00:11 67792kN, tb=11
1.c5 bxc5 2.Be1 Kc7 3.Ba5+ Kb8 4.Kc4 Be7 5.Bc3 Bd8 6.Bg7 Kc8 7.Bh6 Ba5 8.Kd3 Kc7 9.Bg5 Bb6 10.Bxh4 c4+ 11.Kxc4 Be3 12.Bf6 Kb8 13.h4
+- (1.69) Depth: 24/67 00:01:02 395mN, tb=237
1.c5 bxc5 2.Be1 Kc7 3.Ba5+ Kb8 4.Kc4 Be7 5.Bc3 Bd8 6.Bg7 Kc8 7.Bh6 Ba5 8.Kd3 Kc7 9.Bg5 Bb6 10.Bxh4 c4+ 11.Kxc4 Be3 12.Bf6 Kb8 13.h4
+- (1.69) Depth: 24/67 00:01:02 395mN, tb=237

i7 970 (6 cores) @ 3.2 GHz with 160 GB worth of 6-man tablebases.

Trusty Zappa remains my favorite for endgame evaluation.

PK
Top
Jouni
Posts: 3293
Joined: Wed Mar 08, 2006 8:15 pm

Re: Louguet endgame solution: how many hours on your compute

Post by Jouni »

How many of 100 EET positions are solved with Zappa and 60s limit?

Jouni
Top
kasinp
Posts: 251
Joined: Sat Dec 02, 2006 10:47 pm
Location: Toronto
Full name: Peter Kasinski

Re: Louguet endgame solution: how many hours on your compute

Post by kasinp »

Jouni wrote:How many of 100 EET positions are solved with Zappa and 60s limit?

Jouni
I don't think I have EET - if you could post a link I can give it a try.

PK
Top

Return to “Computer Chess Club: General Topics”