Which is the weaker engine that can find the draw in game #2

[ZirconiumX]

FF likes Kf1 for the first 26 plies, then changes to Kh2:

Score : +1.64
Depth : 27/1
Time : 00:02:09
Nodes : 30970000
N/sec : 238536
1. Kh2 Rb8 2. Ra6 Qxc6 3. dxc6 Bc7 4. Ra7 Rc8 5. Rb7 h5 6. Kg3 h4 7. Kg4 Ke7 8. Rxb5

(No - I have no idea how a simplistic program like FF can get the same depth as houdini in half the time)

Matthew:out

[kgburcham]

It seems Deep Blue did not blunder with 44.Kf1 and this move did not play into a drawn position. It was move 45.Ra6 was not best move.
So 45.Qd7 was best move.
I will play out Houdinis best line from 45.Qd7
I used to have the logs for this game but i cannot find them.
score before 45.Ra6 3.16
score after 45.Ra6 .46
so now the next question is, Is the Houdini 44.Kf1 line correct and is it a winning line?
and also, Is 45.Ra6 a blunder?

[D] R7/1r3kp1/1qQb1p1p/1p1PpP2/1Pp1B3/2P4P/6P1/6K1 w - -


CPU0: GenuineIntel Intel64 Family 6 Model 44 Stepping 0 3911 MHz
CPU1: GenuineIntel Intel64 Family 6 Model 44 Stepping 0 3911 MHz
CPU2: GenuineIntel Intel64 Family 6 Model 44 Stepping 0 3911 MHz
CPU3: GenuineIntel Intel64 Family 6 Model 44 Stepping 0 3911 MHz
CPU4: GenuineIntel Intel64 Family 6 Model 44 Stepping 0 3911 MHz
CPU5: GenuineIntel Intel64 Family 6 Model 44 Stepping 0 3911 MHz
CPU6: GenuineIntel Intel64 Family 6 Model 44 Stepping 0 3911 MHz
CPU7: GenuineIntel Intel64 Family 6 Model 44 Stepping 0 3911 MHz
CPU8: GenuineIntel Intel64 Family 6 Model 44 Stepping 0 3911 MHz
CPU9: GenuineIntel Intel64 Family 6 Model 44 Stepping 0 3911 MHz
CPU10: GenuineIntel Intel64 Family 6 Model 44 Stepping 0 3911 MHz
CPU11: GenuineIntel Intel64 Family 6 Model 44 Stepping 0 3911 MHz
12 processor(s) found, POPCNT available
NUMA configuration with 2 node(s), offset 0
4096 MB Large Page Hash
Nalimov 6 men EGTB available - 512 MB cache
Engine: Houdini 2.0c Pro x64 (4096 MB)
by Robert Houdart

multi pv over 15 hours with 12 cores overnight

39 947:21 +3.56 44.Kh1 Rb8 45.Qd7+ Kg8 46.Ra7 Bf8 47.d6 Kh7 48.Bd5 Qe3 49.Qf7 Qc1+ 50.Kh2 Qf4+ 51.g3 Qf2+ 52.Bg2 e4 53.Qc7 Qb6 54.d7 Qxc7 55.Rxc7 Rd8 56.Bxe4 Bd6 57.Rb7 (2015.647.812.884) 35460 TB:2.710
39 947:21 +3.16 44.Kh2 Rb8 45.Qd7+ Kg8 46.Ra7 Bf8 47.Qf7+ Kh8 48.d6 Qf2 49.Bd5 Qf4+ 50.g3 Qf2+ 51.Bg2 Qb6 52.Rd7 e4 53.Bxe4 Qf2+ 54.Bg2 Ra8 55.Qd5 Re8 56.Rf7 Kh7 57.Rxf6 (2015.647.812.884) 35460 TB:2.710
39 947:21 +3.16 44.Kf1 Rb8 45.Qd7+ Kg8 46.Ra7 Bf8 47.Qf7+ Kh8 48.d6 Qe3 49.Bd5 Qc1+ 50.Kf2 Qd2+ 51.Kg1 Qc1+ 52.Kh2 Qf4+ 53.g3 Qf2+ 54.Bg2 Qb6 55.Rd7 e4 56.Bxe4 Qf2+ 57.Bg2 (2015.647.812.884) 35460 TB:2.710
38 947:21 +1.81 44.Qxb6 Rxb6 45.Kf2 Ke7 46.Kf3 Bc7 47.Ra7 Kd8 48.Kg4 Rd6 49.Kh5 Kc8 50.Ra8+ Bb8 51.Ra5 Rd7 52.Rxb5 Ba7 53.Kg6 Bf2 54.d6 Rxd6 55.Kxg7 h5 56.g4 hxg4 57.hxg4 (2015.647.812.884) 35460 TB:2.710
best move: Kg1-h1 time: 947:19.656 min n/s: 35.460.000 CPU 99.8% n/s(1CPU): 35.531.062 nodes: 2015.647.812.884 TB: 2.710

--------------------------------------------------------
33/90 41:43 +0.46 45...Qe3 46.Qxd6 Re8 47.Qc7+ Re7 48.Qc6 Qxe4 49.d6 Qd3+ 50.Kg1 Re8 51.Ra1 h5 52.Kh2 h4 53.Qxb5 Rd8 54.Qc6 Kf8 55.Qc7 Qxd6 56.Qxc4 e4+ 57.Kg1 Re8 58.Re1 Qe5 (90.912.489.213) 36310 TB:3

[Event "?"]
[Site "?"]
[Date "1997.05.04"]
[Round "2"]
[White "Deep Blue"]
[Black "Kasparov"]
[Result "1-0"]

1. e4 e5 2. Nf3 Nc6 3. Bb5 a6 4. Ba4 Nf6 5. O-O Be7 6. Re1
b5 7. Bb3 d6 8. c3 O-O 9. h3 h6 10. d4 Re8 11. Nbd2 Bf8
12. Nf1 Bd7 13. Ng3 Na5 14. Bc2 c5 15. b3 Nc6 16. d5 Ne7
17. Be3 Ng6 18. Qd2 Nh7 19. a4 Nh4 20. Nxh4 Qxh4 21. Qe2
Qd8 22. b4 Qc7 23. Rec1 c4 24. Ra3 Rec8 25. Rca1 Qd8 26. f4
Nf6 27. fxe5 dxe5 28. Qf1 Ne8 29. Qf2 Nd6 30. Bb6 Qe8
31. R3a2 Be7 32. Bc5 Bf8 33. Nf5 Bxf5 34. exf5 f6 35. Bxd6
Bxd6 36. axb5 axb5 37. Be4 Rxa2 38. Qxa2 Qd7 39. Qa7 Rc7
40. Qb6 Rb7 41. Ra8+ Kf7 42. Qa6 Qc7 43. Qc6 Qb6+ 44. Kf1
Rb8 45. Ra6

[zullil]

What is Houdini and Critter analysis on this position with a very fast system?
[D] R7/1r3kp1/1qQb1p1p/1p1PpP2/1Pp1B3/2P4P/6P1/6K1 w - - 11 44

Here's what Critter-1.6a (8 threads, 4GB hash) gives after an hour and about 42 billion nodes searched:

32/82 57:57 41652M 11976K +2.77 Kh1 Rb8 Qd7+ Kg8 Ra7 Bf8 d6 Kh7 Bd5 Qe3 Qf7 Qc1+ Kh2 Qf4+ g3 Qf2+ Bg2 e4 Qc7 Qb6 d7 Qxc7 Rxc7 Rd8 Bxe4 Bd6 Rb7 Be5 Bc6 Bxc3 Bxb5 Bxb4 Bxc4 Bd6 Bf7 h5 Kg2 Kh6 Kf3 Kg5 Bg6 Kh6 Ra7 Kg5 h4+ Kh6 Be8 Kh7 Ra6 Bb4 Rb6 Ba5 Ra6 Bc7 Rc6 Ba5 Rc5 Bb6 Bg6+ Kh6 Rc8 Ba5 Be8 Bb6 Rc6 Bd4 Rc8 Bb6

[pichy]

What is Houdini and Critter analysis on this position with a very fast system?
[D] R7/1r3kp1/1qQb1p1p/1p1PpP2/1Pp1B3/2P4P/6P1/6K1 w - - 11 44

Here's what Critter-1.6a (8 threads, 4GB hash) gives after an hour and about 42 billion nodes searched:

32/82 57:57 41652M 11976K +2.77 Kh1 Rb8 Qd7+ Kg8 Ra7 Bf8 d6 Kh7 Bd5 Qe3 Qf7 Qc1+ Kh2 Qf4+ g3 Qf2+ Bg2 e4 Qc7 Qb6 d7 Qxc7 Rxc7 Rd8 Bxe4 Bd6 Rb7 Be5 Bc6 Bxc3 Bxb5 Bxb4 Bxc4 Bd6 Bf7 h5 Kg2 Kh6 Kf3 Kg5 Bg6 Kh6 Ra7 Kg5 h4+ Kh6 Be8 Kh7 Ra6 Bb4 Rb6 Ba5 Ra6 Bc7 Rc6 Ba5 Rc5 Bb6 Bg6+ Kh6 Rc8 Ba5 Be8 Bb6 Rc6 Bd4 Rc8 Bb6

In order to know for certain that today's programs are way better than Deep Blue we need to know in each game where there was a critical move made by Deep Blue that Today's top programs can still make a better move using the same time that it took Deep Blue to select that particular move. For Instance if it took DB 1 minute or 2 minutes to select Kf1? then Houdini and Critter 1.6a might also select the same Kf1 move using the same time. We can NOT provide Critter and Houdini an hour (60 minutes to analyze) when DB made the move in 1 0r 2 minutes. Therefore, we need to look at DB Log and find out how long it took DB to make certain inferior move.

[zullil]

What is Houdini and Critter analysis on this position with a very fast system?
[D] R7/1r3kp1/1qQb1p1p/1p1PpP2/1Pp1B3/2P4P/6P1/6K1 w - - 11 44

Here's what Critter-1.6a (8 threads, 4GB hash) gives after an hour and about 42 billion nodes searched:

32/82 57:57 41652M 11976K +2.77 Kh1 Rb8 Qd7+ Kg8 Ra7 Bf8 d6 Kh7 Bd5 Qe3 Qf7 Qc1+ Kh2 Qf4+ g3 Qf2+ Bg2 e4 Qc7 Qb6 d7 Qxc7 Rxc7 Rd8 Bxe4 Bd6 Rb7 Be5 Bc6 Bxc3 Bxb5 Bxb4 Bxc4 Bd6 Bf7 h5 Kg2 Kh6 Kf3 Kg5 Bg6 Kh6 Ra7 Kg5 h4+ Kh6 Be8 Kh7 Ra6 Bb4 Rb6 Ba5 Ra6 Bc7 Rc6 Ba5 Rc5 Bb6 Bg6+ Kh6 Rc8 Ba5 Be8 Bb6 Rc6 Bd4 Rc8 Bb6

In order to know for certain that today's programs are way better than Deep Blue we need to know in each game where there was a critical move made by Deep Blue that Today's top programs can still make a better move using the same time that it took Deep Blue to select that particular move. For Instance if it took DB 1 minute or 2 minutes to select Kf1? then Houdini and Critter 1.6a might also select the same Kf1 move using the same time. We can NOT provide Critter and Houdini an hour (60 minutes to analyze) when DB made the move in 1 0r 2 minutes. Therefore, we need to look at DB Log and find out how long it took DB to make certain inferior move.

Based on

I wonder if you let it analyze up to 80/80 close to 80 ply since Deep Blue was able to calculate 2,000,000 positions per second, if it will go back and select f1 as the best move, and what about critter 1.6a what does it say?

I thought you wanted to know if Critter would switch back to Kf1 after a long search. So I ran a long search and only posted the final iteration.

Here's the beginning of the search. It seems unlikely that Critter would choose Kf1, since that move is rejected almost instantly:

Code: Select all

Critter v1.6a 64-bit, by Richard Vida
hardware POPCNT enabled
setoption name Hash value 4096
setboard R7/1r3kp1/1qQb1p1p/1p1PpP2/1Pp1B3/2P4P/6P1/6K1 w - - 11 44
info string Opening book: "book.cbk" - 57853 entries [native]
info string GTB Init OK (5 piece set)
info string Using 8 thread(s), 4096 MB hash LP=off
go infinite
 2/ 4  00:00      162   81000  +0.54 Kf1 Qxc6 dxc6 Rc7
 2/ 4  00:00      226  113000  +0.71 Qxb6 Rxb6 Ra7+ Be7
 3/ 4  00:00      284  142000  +0.71 Qxb6 Rxb6 Ra7+ Be7
 4/ 6  00:00      471  157000  +0.87 Qxb6 Rxb6 Ra7+ Be7 Kf2 Rd6
 4/ 8  00:00     1038  173000  +0.91 Kh1 Qxc6 dxc6 Rc7 Ra5 Ke7 Rxb5
 5/10  00:00     1545  193125  +0.91 Kh1 Qxc6 dxc6 Rc7 Ra5 Ke7 Rxb5
 6/10  00:00     2060  206000  +1.11 Kh1 Qxc6 dxc6 Rc7 Ra5 Ke7 Kh2 Kd8 Rxb5 Kc8
 6/10  00:00     2845  237083  +1.58 Kf1 Qxc6 dxc6 Rc7 Ra5 Ke7 Kf2 Kd8 Rxb5 Kc8
 7/10  00:00     3434  245285  +1.58 Kf1 Qxc6 dxc6 Rc7 Ra5 Ke7 Kf2 Kd8 Rxb5 Kc8
 8/12+ 00:00     4286  267875  +1.67 Kf1 Qxc6 dxc6 Rc7 Ra5 Ke7 Kf2 Kd8 Rxb5 Kc8
 8/12  00:00     4763  280176  +1.72 Kf1 Qxc6 dxc6 Rc7 Ra5 Ke7 Kf2 Kd8 Kf3 h5 Rxb5 Kc8
 9/15- 00:00    10233  319781  +1.61 Kf1 Qxc6 dxc6 Rc7 Ra5 Ke7 Kf2 Kd8 Kf3 h5 Rxb5 Kc8
 9/16+ 00:00    21372  339238  +1.82 Kh2 Qxc6 dxc6 Rc7 Kg3 Ke7 Ra5 Kd8 Rxb5 Kc8 Ra5
 9/18  00:00    34242  372195  +1.77 Kh2 Qxc6 dxc6 Rc7 Kg3 h5 Ra5 Ke7 Kf3 Kd8 Bd5 e4+ Ke3 Be5 Rxb5 Bxc3 Kxe4 Be5 Rb8+ Rc8
10/27- 00:00    49330  418050  +1.66 Kh2 Qxc6 dxc6 Rc7 Kg3 h5 Ra5 Ke7 Kf3 Kd8 Bd5 e4+ Ke3 Be5 Rxb5 Bxc3 Kxe4 Be5 Rb8+ Rc8
10/27- 00:00    61119  446124  +1.50 Kh2 Qxc6 dxc6 Rc7 Kg3 h5 Ra5 Ke7 Kf3 Kd8 Bd5 e4+ Ke3 Be5 Rxb5 Bxc3 Kxe4 Be5 Rb8+ Rc8
10/27  00:00    75907  465687  +1.34 Kh2 Rb8 Qd7+ Kg8 Ra7 Bf8 Qe6+ Kh7 Qxb6 Rxb6 Kg3 Bd6 Kf3 Kg8 Ke3 Bb8 Re7 Kf8
10/27  00:00   107750  533415  +1.57 Kf1 Rb8 Ra6 Qxc6 dxc6 Kf8 Ra7 Rc8 Rb7 Bc7 Bd5 Rd8 Bxc4 bxc4 Rxc7
11/27+ 00:00   117357  545846  +1.66 Kf1 Rb8 Ra6 Qxc6 dxc6 Kf8 Ra7 Rc8 Rb7 Bc7 Bd5 Rd8 Bxc4 bxc4 Rxc7
11/27- 00:00   158798  590327  +1.47 Kf1 Rb8 Ra6 Qxc6 dxc6 Kf8 Ra7 Rc8 Rb7 Bc7 Bd5 Rd8 Bxc4 bxc4 Rxc7
11/27  00:00   201526  621993  +1.76 Kf1 Rb8 Ra6 Qxc6 dxc6 Kf8 Ra7 Rd8 Rb7 Rc8 Rxb5 Bc7 Rb7 Ke7 Ra7 Kd6 b5
12/27  00:00   205255  623875  +1.76 Kf1 Rb8 Ra6 Qxc6 dxc6 Kf8 Ra7 Rd8 Rb7 Rc8 Rxb5 Bc7 Rb7 Ke7 Ra7 Kd6 b5
13/27  00:00   281821  701047  +1.77 Kf1 Rb8 Ra6 Qxc6 dxc6 Rc8 Ra7+ Rc7 Ra5 Ke7 Kf2 Rc8 Rxb5 Bc7 Rb7 Kd6 Kf3 Rb8 Ra7 Rd8
14/29+ 00:00   401550  798310  +1.88 Kf1 Rb8 Ra6 Qxc6 dxc6 Rc8 Ra7+ Rc7 Ra5 Ke7 Kf2 Rc8 Rxb5 Bc7 Rb7 Kd6 Kf3 Rb8 Ra7 Rd8
14/29  00:00   474691  864646  +1.83 Kf1 Rb8 Ra6 Qxc6 dxc6 Rc8 Ra7+ Rc7 Ra5 Ke7 Ke2 Rc8 Rxb5 Bc7 Rc5 Bb6 Rxc4 Kd6 b5 Kc7 Ra4
15/29  00:00   689758   1006K  +1.87 Kf1 Rb8 Ra6 Qxc6 dxc6 Rc8 Ra7+ Rc7 Ra5 Ke7 Ke2 Rc8 Rxb5 Bc7 Rc5 Rd8 Rxc4 Kd6 b5 Rb8 Rb4 Ba5 Rb3 Kc7
16/38- 00:01    2753K   1743K  +1.75 Kf1 Rb8 Ra6 Qxc6 dxc6 Rc8 Ra7+ Rc7 Ra5 Ke7 Ke2 Rc8 Rxb5 Bc7 Rc5 Rd8 Rxc4 Kd6 b5 Rb8 Rb4 Ba5 Rb3 Kc7
16/38- 00:01    3782K   2045K  +1.58 Kf1 Rb8 Ra6 Qxc6 dxc6 Rc8 Ra7+ Rc7 Ra5 Ke7 Ke2 Rc8 Rxb5 Bc7 Rc5 Rd8 Rxc4 Kd6 b5 Rb8 Rb4 Ba5 Rb3 Kc7
16/51  00:02    6699K   2615K  +1.40 Kf1 Rb8 Qd7+ Kg8 Ra7 Bf8 Qf7+ Kh7 d6 Qe3 Bd5 Qd3+ Kg1 Qe3+ Kh2 Qf4+ Kg1 Qxf5 Qc7 Rc8 Qb7
16/51  00:02    8001K   2848K  +1.67 Kh1 Rb8 Ra6 Qxc6 dxc6 Rc8 Ra7+ Rc7 Ra5 Ke7 Kh2 h5 Kg3 Rc8 Rxb5 Rb8 Rd5 Rd8 b5 Bc7 Rc5 Rb8
17/51+ 00:02    8763K   2998K  +1.77 Kh1 Rb8 Ra6 Qxc6 dxc6 Rc8 Ra7+ Rc7 Ra5 Ke7 Kh2 h5 Kg3 Rc8 Rxb5 Rb8 Rd5 Rd8 b5 Bc7 Rc5 Rb8 Kf3
17/51  00:02    8918K   3018K  +1.76 Kh1 Rb8 Ra6 Qxc6 dxc6 Kf8 Ra7 Rc8 Rd7 Bc7 Kh2 Bb6 Rb7 Bc7 Rxb5 Ke7 Rb7 Kd6 Ra7 Rb8 Kg3 Rd8 Kf3 Rb8 Ra2 Rd8 Rd2+ Ke7
18/51  00:03    9964K   3208K  +1.76 Kh1 Rb8 Ra6 Qxc6 dxc6 Kf8 Ra7 Rc8 Rd7 Bc7 Kh2 Bb6 Rb7 Bc7 Rxb5 Ke7 Rb7 Kd6 Ra7 Rb8 Kg3 Rd8 Kf3 Rc8 Ra1 Bb6 b5
19/51  00:03   12508K   3619K  +1.71 Kh1 Rb8 Ra6 Qxc6 dxc6 Kf8 Ra7 Rc8 Rd7 Bc7 Kh2 Bb6 Rb7 Bc7 Rxb5 Ke7 Rb7 Kd6 Ra7 Rb8 Kg3 Rd8 Kf3 h5 g3 Rb8 h4
20/51  00:03   16703K   4201K  +1.71 Kh1 Rb8 Qd7+ Kg8 Ra7 Bf8 Qf7+ Kh8 d6 Qe3 Bd5 Qc1+ Kh2 Qf4+ g3 Qd2+ Bg2 e4 d7 Qe3 Rc7 Qxc3 Qe8
21/51  00:05   30714K   5533K  +1.78 Kh1 Rb8 Ra6 Qxc6 dxc6 Kf8 Ra7 Rc8 Rd7 Bc7 Kh2 Bb6 Rb7 Bc7 Rxb5 Ke7 Rb7 Kd6 Kg3 Rb8 Ra7 Rc8 Kf3 Rd8 g3 Rb8 h4
22/51+ 00:07   44766K   6380K  +1.89 Kh1 Rb8 Ra6 Qxc6 dxc6 Kf8 Ra7 Rc8 Rd7 Bc7 Kh2 Bb6 Rb7 Bc7 Rxb5 Ke7 Rb7 Kd6 Kg3 Rb8 Ra7 Rc8 Kf3 Rd8 g3 h5 h4 Rb8 Ra2 Rd8 Rd2+ Ke7
22/51  00:08   53865K   6714K  +1.93 Kh1 Rb8 Ra6 Qxc6 dxc6 Kf8 Ra7 Rc8 Rd7 Bc7 Kh2 Bb6 Rb7 Bc7 Rxb5 Ke7 Kg3 h5 Rc5 Rd8 Rxc4 Rd2 b5 Kd6 Rb4 Kc5 Kf3 Kb6 g3
23/52  00:10   81825K   7558K  +1.97 Kh1 Rb8 Ra6 Qxc6 dxc6 Kf8 Ra7 Rc8 Rd7 Bc7 Kh2 Bb6 Rb7 Bc7 Rxb5 Ke7 Kg3 Rb8 Rc5 Bd6 Rxc4 Kd8 Kf3 Kc7 Bd5 Bf8 Rg4 h5 Rg6 Ra8 Ke4 Ra3 Rg3
24/53  00:13  113139K   8112K  +1.97 Kh1 Rb8 Ra6 Qxc6 dxc6 Kf8 Ra7 Rc8 Rd7 Bc7 Kh2 Bb6 Rb7 Bc7 Rxb5 Ke7 Kg3 Rb8 Rc5 Bd6 Rxc4 Kd8 Kf3 Kc7 Bd5 Bf8 Rg4 h5 Rg6 Ra8 Ke4 Ra3 Rg3
25/56  00:19  166714K   8705K  +1.97 Kh1 Rb8 Ra6 Qxc6 dxc6 Kf8 Ra7 Rc8 Rd7 Bc7 Kh2 Bb6 Rb7 Bc7 Rxb5 Ke7 Kg3 Rb8 Rc5 Bd6 Rxc4 Kd8 Kf3 Kc7 Bd5 Bf8 Rg4 h5 Rg6 Ra8 Ke4 Ra3 Rg3
26/59  00:30  289543K   9482K  +1.93 Kh1 Rb8 Ra6 Qxc6 dxc6 Kf8 Ra7 Rc8 Rd7 Bc7 Kh2 Bb6 Rb7 Bc7 Rxb5 Ke7 Kg3 Rb8 Rc5 Bd6 Rxc4 Kd8 Kf3 Kc7 Bd5 Bf8 Rg4 Ra8 Ke4 Ra1 b5 Rb1 c4 Re1+ Kd3 Rd1+ Ke3 Re1+ Kf2 h5
27/61  01:06  669383K   9994K  +1.91 Kh1 Rb8 Qd7+ Kg8 Ra7 Bf8 d6 Qe3 Bd5+ Kh7 Qf7 Qc1+ Kh2 Qf4+ Kg1 Qxf5 Qc7 Rc8 Qb7
27/72  02:00    1246M  10373K  +1.92 Kh2 Rb8 Qd7+ Kg8 Ra7 Bf8 d6 Qf2 Qf7+ Kh7 Bd5 Qf4+ Kg1 Qxf5 Qc7 Rc8 Qb7 Kh8 d7 Rd8 Ra8 Rxa8 Qxa8 Qb1+ Kh2
28/72  02:21    1475M  10437K  +1.96 Kh2 Rb8 Qd7+ Kg8 Ra7 Bf8 Qf7+ Kh8 d6 Qe3 Bd5 Qf4+ g3 Qf2+ Bg2 Qb6 Rd7 e4 Bxe4 Qf2+ Bg2 Qxf5 Rb7 Rd8 Qc7 Rxd6 Qb8 Kg8 Qe8 Rd2 Qf7+ Kh7 Qxf8 Rxg2+
29/73  03:24    2174M  10652K  +1.98 Kh2 Rb8 Qd7+ Kg8 Ra7 Bf8 Qf7+ Kh8 d6 Qe3 Bd5 Qf4+ g3 Qf2+ Bg2 Qb6 Rd7 e4 Bxe4 Qe3 Bg2 Qxc3 Rb7 Rxb7 Qxf8+ Kh7 Bxb7 Qxb4 Bg2 c3 Qc8 Qxd6 Qxc3 b4 Qb3 Qb6 h4 Qc5 Qa2 Qe5 Qc4 Qe1 Kh3 Qd2 Qe6 Qd1 Bd5 Qf1+ Kg4 Qd1+ Kf4 Qc1+ Kf3 Qf1+ Ke3 Qe1+ Kd4 Qc3+ Ke4
30/76  05:53    3862M  10916K  +2.10 Kh2 Rb8 Qd7+ Kg8 Ra7 Bf8 Qf7+ Kh8 d6 Qe3 Bd5 Qf4+ g3 Qf2+ Bg2 Qb6 Rd7 e4 Bxe4 Qf2+ Bg2 Qxf5 Rb7 Rd8 Qc7 Rxd6 Rb8 Rd2 Rxf8+ Kh7 Qb7 Qe5 Qa8 Re2 Rh8+ Kg6 Qf3 h5 h4 Rc2 Rf8 Re2 Qb7 Kh7 Qf7 Rd2
31/80  09:58    6674M  11151K  +2.22 Kh2 Rb8 Qd7+ Kg8 Ra7 Bf8 Qf7+ Kh8 d6 Qe3 Bd5 Qf4+ g3 Qf2+ Bg2 Qb6 Rd7 e4 Bxe4 Qe3 Bg2 Qxc3 Rb7 Rxb7 Qxf8+ Kh7 Bxb7 Qxb4 Bg2 c3 Qa8 c2 d7 Qd2 d8=Q Qxd8 Qxd8 c1=Q Qd3 b4 Bd5 Qb2+ Kg1 Qa1+ Kf2 h5 Bb3 Qb2+ Qc2 Qd4+ Kf3 Qg1 Qc8 Qf1+ Ke4 Qe1+ Kf4

[pichy]

What is Houdini and Critter analysis on this position with a very fast system?
[D] R7/1r3kp1/1qQb1p1p/1p1PpP2/1Pp1B3/2P4P/6P1/6K1 w - - 11 44

Here's what Critter-1.6a (8 threads, 4GB hash) gives after an hour and about 42 billion nodes searched:

32/82 57:57 41652M 11976K +2.77 Kh1 Rb8 Qd7+ Kg8 Ra7 Bf8 d6 Kh7 Bd5 Qe3 Qf7 Qc1+ Kh2 Qf4+ g3 Qf2+ Bg2 e4 Qc7 Qb6 d7 Qxc7 Rxc7 Rd8 Bxe4 Bd6 Rb7 Be5 Bc6 Bxc3 Bxb5 Bxb4 Bxc4 Bd6 Bf7 h5 Kg2 Kh6 Kf3 Kg5 Bg6 Kh6 Ra7 Kg5 h4+ Kh6 Be8 Kh7 Ra6 Bb4 Rb6 Ba5 Ra6 Bc7 Rc6 Ba5 Rc5 Bb6 Bg6+ Kh6 Rc8 Ba5 Be8 Bb6 Rc6 Bd4 Rc8 Bb6

In order to know for certain that today's programs are way better than Deep Blue we need to know in each game where there was a critical move made by Deep Blue that Today's top programs can still make a better move using the same time that it took Deep Blue to select that particular move. For Instance if it took DB 1 minute or 2 minutes to select Kf1? then Houdini and Critter 1.6a might also select the same Kf1 move using the same time. We can NOT provide Critter and Houdini an hour (60 minutes to analyze) when DB made the move in 1 0r 2 minutes. Therefore, we need to look at DB Log and find out how long it took DB to make certain inferior move.

Based on

I wonder if you let it analyze up to 80/80 close to 80 ply since Deep Blue was able to calculate 2,000,000 positions per second, if it will go back and select f1 as the best move, and what about critter 1.6a what does it say?

I thought you wanted to know if Critter would switch back to Kf1 after a long search. So I ran a long search and only posted the final iteration.

Here's the beginning of the search. It seems unlikely that Critter would choose Kf1, since that move is rejected almost instantly:

Code: Select all

Critter v1.6a 64-bit, by Richard Vida
hardware POPCNT enabled
setoption name Hash value 4096
setboard R7/1r3kp1/1qQb1p1p/1p1PpP2/1Pp1B3/2P4P/6P1/6K1 w - - 11 44
info string Opening book: "book.cbk" - 57853 entries [native]
info string GTB Init OK (5 piece set)
info string Using 8 thread(s), 4096 MB hash LP=off
go infinite
 2/ 4  00:00      162   81000  +0.54 Kf1 Qxc6 dxc6 Rc7
 2/ 4  00:00      226  113000  +0.71 Qxb6 Rxb6 Ra7+ Be7
 3/ 4  00:00      284  142000  +0.71 Qxb6 Rxb6 Ra7+ Be7
 4/ 6  00:00      471  157000  +0.87 Qxb6 Rxb6 Ra7+ Be7 Kf2 Rd6
 4/ 8  00:00     1038  173000  +0.91 Kh1 Qxc6 dxc6 Rc7 Ra5 Ke7 Rxb5
 5/10  00:00     1545  193125  +0.91 Kh1 Qxc6 dxc6 Rc7 Ra5 Ke7 Rxb5
 6/10  00:00     2060  206000  +1.11 Kh1 Qxc6 dxc6 Rc7 Ra5 Ke7 Kh2 Kd8 Rxb5 Kc8
 6/10  00:00     2845  237083  +1.58 Kf1 Qxc6 dxc6 Rc7 Ra5 Ke7 Kf2 Kd8 Rxb5 Kc8
 7/10  00:00     3434  245285  +1.58 Kf1 Qxc6 dxc6 Rc7 Ra5 Ke7 Kf2 Kd8 Rxb5 Kc8
 8/12+ 00:00     4286  267875  +1.67 Kf1 Qxc6 dxc6 Rc7 Ra5 Ke7 Kf2 Kd8 Rxb5 Kc8
 8/12  00:00     4763  280176  +1.72 Kf1 Qxc6 dxc6 Rc7 Ra5 Ke7 Kf2 Kd8 Kf3 h5 Rxb5 Kc8
 9/15- 00:00    10233  319781  +1.61 Kf1 Qxc6 dxc6 Rc7 Ra5 Ke7 Kf2 Kd8 Kf3 h5 Rxb5 Kc8
 9/16+ 00:00    21372  339238  +1.82 Kh2 Qxc6 dxc6 Rc7 Kg3 Ke7 Ra5 Kd8 Rxb5 Kc8 Ra5

I don't think that Kf1 was a Blunder by Deep Blue, since after 45...Qe3 46.Qxd6 Re8 47.Qd7+ Re7 48.Qc6 Qxe4 49.d6 Qd3+ 50.Kg1 Re8 51.Ra1 there is no perpetual check, but Kh2 is a better move that is all.

[Uri Blass]

What is Houdini and Critter analysis on this position with a very fast system?
[D] R7/1r3kp1/1qQb1p1p/1p1PpP2/1Pp1B3/2P4P/6P1/6K1 w - - 11 44

Here's what Critter-1.6a (8 threads, 4GB hash) gives after an hour and about 42 billion nodes searched:

32/82 57:57 41652M 11976K +2.77 Kh1 Rb8 Qd7+ Kg8 Ra7 Bf8 d6 Kh7 Bd5 Qe3 Qf7 Qc1+ Kh2 Qf4+ g3 Qf2+ Bg2 e4 Qc7 Qb6 d7 Qxc7 Rxc7 Rd8 Bxe4 Bd6 Rb7 Be5 Bc6 Bxc3 Bxb5 Bxb4 Bxc4 Bd6 Bf7 h5 Kg2 Kh6 Kf3 Kg5 Bg6 Kh6 Ra7 Kg5 h4+ Kh6 Be8 Kh7 Ra6 Bb4 Rb6 Ba5 Ra6 Bc7 Rc6 Ba5 Rc5 Bb6 Bg6+ Kh6 Rc8 Ba5 Be8 Bb6 Rc6 Bd4 Rc8 Bb6

In order to know for certain that today's programs are way better than Deep Blue we need to know in each game where there was a critical move made by Deep Blue that Today's top programs can still make a better move using the same time that it took Deep Blue to select that particular move. For Instance if it took DB 1 minute or 2 minutes to select Kf1? then Houdini and Critter 1.6a might also select the same Kf1 move using the same time. We can NOT provide Critter and Houdini an hour (60 minutes to analyze) when DB made the move in 1 0r 2 minutes. Therefore, we need to look at DB Log and find out how long it took DB to make certain inferior move.

Based on

I wonder if you let it analyze up to 80/80 close to 80 ply since Deep Blue was able to calculate 2,000,000 positions per second, if it will go back and select f1 as the best move, and what about critter 1.6a what does it say?

I thought you wanted to know if Critter would switch back to Kf1 after a long search. So I ran a long search and only posted the final iteration.

Here's the beginning of the search. It seems unlikely that Critter would choose Kf1, since that move is rejected almost instantly:

Code: Select all

Critter v1.6a 64-bit, by Richard Vida
hardware POPCNT enabled
setoption name Hash value 4096
setboard R7/1r3kp1/1qQb1p1p/1p1PpP2/1Pp1B3/2P4P/6P1/6K1 w - - 11 44
info string Opening book: "book.cbk" - 57853 entries [native]
info string GTB Init OK (5 piece set)
info string Using 8 thread(s), 4096 MB hash LP=off
go infinite
 2/ 4  00:00      162   81000  +0.54 Kf1 Qxc6 dxc6 Rc7
 2/ 4  00:00      226  113000  +0.71 Qxb6 Rxb6 Ra7+ Be7
 3/ 4  00:00      284  142000  +0.71 Qxb6 Rxb6 Ra7+ Be7
 4/ 6  00:00      471  157000  +0.87 Qxb6 Rxb6 Ra7+ Be7 Kf2 Rd6
 4/ 8  00:00     1038  173000  +0.91 Kh1 Qxc6 dxc6 Rc7 Ra5 Ke7 Rxb5
 5/10  00:00     1545  193125  +0.91 Kh1 Qxc6 dxc6 Rc7 Ra5 Ke7 Rxb5
 6/10  00:00     2060  206000  +1.11 Kh1 Qxc6 dxc6 Rc7 Ra5 Ke7 Kh2 Kd8 Rxb5 Kc8
 6/10  00:00     2845  237083  +1.58 Kf1 Qxc6 dxc6 Rc7 Ra5 Ke7 Kf2 Kd8 Rxb5 Kc8
 7/10  00:00     3434  245285  +1.58 Kf1 Qxc6 dxc6 Rc7 Ra5 Ke7 Kf2 Kd8 Rxb5 Kc8
 8/12+ 00:00     4286  267875  +1.67 Kf1 Qxc6 dxc6 Rc7 Ra5 Ke7 Kf2 Kd8 Rxb5 Kc8
 8/12  00:00     4763  280176  +1.72 Kf1 Qxc6 dxc6 Rc7 Ra5 Ke7 Kf2 Kd8 Kf3 h5 Rxb5 Kc8
 9/15- 00:00    10233  319781  +1.61 Kf1 Qxc6 dxc6 Rc7 Ra5 Ke7 Kf2 Kd8 Kf3 h5 Rxb5 Kc8
 9/16+ 00:00    21372  339238  +1.82 Kh2 Qxc6 dxc6 Rc7 Kg3 Ke7 Ra5 Kd8 Rxb5 Kc8 Ra5

I don't think that Kf1 was a Blunder by Deep Blue, since after 45...Qe3 46.Qxd6 Re8 47.Qd7+ Re7 48.Qc6 Qxe4 49.d6 Qd3+ 50.Kg1 Re8 51.Ra1 there is no perpetual check, but Kh2 is a better move that is all.

The fact that there is no perpetual check does not mean that white can win and the material is equal.

I think that Ra6 is the blunder and better was Qd7+

[pichy]

What is Houdini and Critter analysis on this position with a very fast system?
[D] R7/1r3kp1/1qQb1p1p/1p1PpP2/1Pp1B3/2P4P/6P1/6K1 w - - 11 44

Here's what Critter-1.6a (8 threads, 4GB hash) gives after an hour and about 42 billion nodes searched:

32/82 57:57 41652M 11976K +2.77 Kh1 Rb8 Qd7+ Kg8 Ra7 Bf8 d6 Kh7 Bd5 Qe3 Qf7 Qc1+ Kh2 Qf4+ g3 Qf2+ Bg2 e4 Qc7 Qb6 d7 Qxc7 Rxc7 Rd8 Bxe4 Bd6 Rb7 Be5 Bc6 Bxc3 Bxb5 Bxb4 Bxc4 Bd6 Bf7 h5 Kg2 Kh6 Kf3 Kg5 Bg6 Kh6 Ra7 Kg5 h4+ Kh6 Be8 Kh7 Ra6 Bb4 Rb6 Ba5 Ra6 Bc7 Rc6 Ba5 Rc5 Bb6 Bg6+ Kh6 Rc8 Ba5 Be8 Bb6 Rc6 Bd4 Rc8 Bb6

In order to know for certain that today's programs are way better than Deep Blue we need to know in each game where there was a critical move made by Deep Blue that Today's top programs can still make a better move using the same time that it took Deep Blue to select that particular move. For Instance if it took DB 1 minute or 2 minutes to select Kf1? then Houdini and Critter 1.6a might also select the same Kf1 move using the same time. We can NOT provide Critter and Houdini an hour (60 minutes to analyze) when DB made the move in 1 0r 2 minutes. Therefore, we need to look at DB Log and find out how long it took DB to make certain inferior move.

Based on

I wonder if you let it analyze up to 80/80 close to 80 ply since Deep Blue was able to calculate 2,000,000 positions per second, if it will go back and select f1 as the best move, and what about critter 1.6a what does it say?

I thought you wanted to know if Critter would switch back to Kf1 after a long search. So I ran a long search and only posted the final iteration.

Here's the beginning of the search. It seems unlikely that Critter would choose Kf1, since that move is rejected almost instantly:

Code: Select all

Critter v1.6a 64-bit, by Richard Vida
hardware POPCNT enabled
setoption name Hash value 4096
setboard R7/1r3kp1/1qQb1p1p/1p1PpP2/1Pp1B3/2P4P/6P1/6K1 w - - 11 44
info string Opening book: "book.cbk" - 57853 entries [native]
info string GTB Init OK (5 piece set)
info string Using 8 thread(s), 4096 MB hash LP=off
go infinite
 2/ 4  00:00      162   81000  +0.54 Kf1 Qxc6 dxc6 Rc7
 2/ 4  00:00      226  113000  +0.71 Qxb6 Rxb6 Ra7+ Be7
 3/ 4  00:00      284  142000  +0.71 Qxb6 Rxb6 Ra7+ Be7
 4/ 6  00:00      471  157000  +0.87 Qxb6 Rxb6 Ra7+ Be7 Kf2 Rd6
 4/ 8  00:00     1038  173000  +0.91 Kh1 Qxc6 dxc6 Rc7 Ra5 Ke7 Rxb5
 5/10  00:00     1545  193125  +0.91 Kh1 Qxc6 dxc6 Rc7 Ra5 Ke7 Rxb5
 6/10  00:00     2060  206000  +1.11 Kh1 Qxc6 dxc6 Rc7 Ra5 Ke7 Kh2 Kd8 Rxb5 Kc8
 6/10  00:00     2845  237083  +1.58 Kf1 Qxc6 dxc6 Rc7 Ra5 Ke7 Kf2 Kd8 Rxb5 Kc8
 7/10  00:00     3434  245285  +1.58 Kf1 Qxc6 dxc6 Rc7 Ra5 Ke7 Kf2 Kd8 Rxb5 Kc8
 8/12+ 00:00     4286  267875  +1.67 Kf1 Qxc6 dxc6 Rc7 Ra5 Ke7 Kf2 Kd8 Rxb5 Kc8
 8/12  00:00     4763  280176  +1.72 Kf1 Qxc6 dxc6 Rc7 Ra5 Ke7 Kf2 Kd8 Kf3 h5 Rxb5 Kc8
 9/15- 00:00    10233  319781  +1.61 Kf1 Qxc6 dxc6 Rc7 Ra5 Ke7 Kf2 Kd8 Kf3 h5 Rxb5 Kc8
 9/16+ 00:00    21372  339238  +1.82 Kh2 Qxc6 dxc6 Rc7 Kg3 Ke7 Ra5 Kd8 Rxb5 Kc8 Ra5

I don't think that Kf1 was a Blunder by Deep Blue, since after 45...Qe3 46.Qxd6 Re8 47.Qd7+ Re7 48.Qc6 Qxe4 49.d6 Qd3+ 50.Kg1 Re8 51.Ra1 there is no perpetual check, but Kh2 is a better move that is all.

The fact that there is no perpetual check does not mean that white can win and the material is equal.

I think that Ra6 is the blunder and better was Qd7+

I strongly agree with you Uri, but how many engines can select Qd7+! besised Houdini, Critter 1.6a and possibly Komodo

[Uri Blass]

What is Houdini and Critter analysis on this position with a very fast system?
[D] R7/1r3kp1/1qQb1p1p/1p1PpP2/1Pp1B3/2P4P/6P1/6K1 w - - 11 44

Here's what Critter-1.6a (8 threads, 4GB hash) gives after an hour and about 42 billion nodes searched:

32/82 57:57 41652M 11976K +2.77 Kh1 Rb8 Qd7+ Kg8 Ra7 Bf8 d6 Kh7 Bd5 Qe3 Qf7 Qc1+ Kh2 Qf4+ g3 Qf2+ Bg2 e4 Qc7 Qb6 d7 Qxc7 Rxc7 Rd8 Bxe4 Bd6 Rb7 Be5 Bc6 Bxc3 Bxb5 Bxb4 Bxc4 Bd6 Bf7 h5 Kg2 Kh6 Kf3 Kg5 Bg6 Kh6 Ra7 Kg5 h4+ Kh6 Be8 Kh7 Ra6 Bb4 Rb6 Ba5 Ra6 Bc7 Rc6 Ba5 Rc5 Bb6 Bg6+ Kh6 Rc8 Ba5 Be8 Bb6 Rc6 Bd4 Rc8 Bb6

In order to know for certain that today's programs are way better than Deep Blue we need to know in each game where there was a critical move made by Deep Blue that Today's top programs can still make a better move using the same time that it took Deep Blue to select that particular move. For Instance if it took DB 1 minute or 2 minutes to select Kf1? then Houdini and Critter 1.6a might also select the same Kf1 move using the same time. We can NOT provide Critter and Houdini an hour (60 minutes to analyze) when DB made the move in 1 0r 2 minutes. Therefore, we need to look at DB Log and find out how long it took DB to make certain inferior move.

Based on

I wonder if you let it analyze up to 80/80 close to 80 ply since Deep Blue was able to calculate 2,000,000 positions per second, if it will go back and select f1 as the best move, and what about critter 1.6a what does it say?

I thought you wanted to know if Critter would switch back to Kf1 after a long search. So I ran a long search and only posted the final iteration.

Here's the beginning of the search. It seems unlikely that Critter would choose Kf1, since that move is rejected almost instantly:

Code: Select all

Critter v1.6a 64-bit, by Richard Vida
hardware POPCNT enabled
setoption name Hash value 4096
setboard R7/1r3kp1/1qQb1p1p/1p1PpP2/1Pp1B3/2P4P/6P1/6K1 w - - 11 44
info string Opening book: "book.cbk" - 57853 entries [native]
info string GTB Init OK (5 piece set)
info string Using 8 thread(s), 4096 MB hash LP=off
go infinite
 2/ 4  00:00      162   81000  +0.54 Kf1 Qxc6 dxc6 Rc7
 2/ 4  00:00      226  113000  +0.71 Qxb6 Rxb6 Ra7+ Be7
 3/ 4  00:00      284  142000  +0.71 Qxb6 Rxb6 Ra7+ Be7
 4/ 6  00:00      471  157000  +0.87 Qxb6 Rxb6 Ra7+ Be7 Kf2 Rd6
 4/ 8  00:00     1038  173000  +0.91 Kh1 Qxc6 dxc6 Rc7 Ra5 Ke7 Rxb5
 5/10  00:00     1545  193125  +0.91 Kh1 Qxc6 dxc6 Rc7 Ra5 Ke7 Rxb5
 6/10  00:00     2060  206000  +1.11 Kh1 Qxc6 dxc6 Rc7 Ra5 Ke7 Kh2 Kd8 Rxb5 Kc8
 6/10  00:00     2845  237083  +1.58 Kf1 Qxc6 dxc6 Rc7 Ra5 Ke7 Kf2 Kd8 Rxb5 Kc8
 7/10  00:00     3434  245285  +1.58 Kf1 Qxc6 dxc6 Rc7 Ra5 Ke7 Kf2 Kd8 Rxb5 Kc8
 8/12+ 00:00     4286  267875  +1.67 Kf1 Qxc6 dxc6 Rc7 Ra5 Ke7 Kf2 Kd8 Rxb5 Kc8
 8/12  00:00     4763  280176  +1.72 Kf1 Qxc6 dxc6 Rc7 Ra5 Ke7 Kf2 Kd8 Kf3 h5 Rxb5 Kc8
 9/15- 00:00    10233  319781  +1.61 Kf1 Qxc6 dxc6 Rc7 Ra5 Ke7 Kf2 Kd8 Kf3 h5 Rxb5 Kc8
 9/16+ 00:00    21372  339238  +1.82 Kh2 Qxc6 dxc6 Rc7 Kg3 Ke7 Ra5 Kd8 Rxb5 Kc8 Ra5

I don't think that Kf1 was a Blunder by Deep Blue, since after 45...Qe3 46.Qxd6 Re8 47.Qd7+ Re7 48.Qc6 Qxe4 49.d6 Qd3+ 50.Kg1 Re8 51.Ra1 there is no perpetual check, but Kh2 is a better move that is all.

The fact that there is no perpetual check does not mean that white can win and the material is equal.

I think that Ra6 is the blunder and better was Qd7+

I strongly agree with you Uri, but how many engines can select Qd7+! besised Houdini, Critter 1.6a and possibly Komodo

I expect every strong engine to find it if you give it enough time.

[pichy]

It seems Deep Blue did not blunder with 44.Kf1 and this move did not play into a drawn position. It was move 45.Ra6 was not best move.
So 45.Qd7 was best move.
I will play out Houdinis best line from 45.Qd7
I used to have the logs for this game but i cannot find them.
score before 45.Ra6 3.16
score after 45.Ra6 .46
so now the next question is, Is the Houdini 44.Kf1 line correct and is it a winning line?
and also, Is 45.Ra6 a blunder?

WHAT DID YOU FOUND OUT?

[D] R7/1r3kp1/1qQb1p1p/1p1PpP2/1Pp1B3/2P4P/6P1/6K1 w - -


CPU0: GenuineIntel Intel64 Family 6 Model 44 Stepping 0 3911 MHz
CPU1: GenuineIntel Intel64 Family 6 Model 44 Stepping 0 3911 MHz
CPU2: GenuineIntel Intel64 Family 6 Model 44 Stepping 0 3911 MHz
CPU3: GenuineIntel Intel64 Family 6 Model 44 Stepping 0 3911 MHz
CPU4: GenuineIntel Intel64 Family 6 Model 44 Stepping 0 3911 MHz
CPU5: GenuineIntel Intel64 Family 6 Model 44 Stepping 0 3911 MHz
CPU6: GenuineIntel Intel64 Family 6 Model 44 Stepping 0 3911 MHz
CPU7: GenuineIntel Intel64 Family 6 Model 44 Stepping 0 3911 MHz
CPU8: GenuineIntel Intel64 Family 6 Model 44 Stepping 0 3911 MHz
CPU9: GenuineIntel Intel64 Family 6 Model 44 Stepping 0 3911 MHz
CPU10: GenuineIntel Intel64 Family 6 Model 44 Stepping 0 3911 MHz
CPU11: GenuineIntel Intel64 Family 6 Model 44 Stepping 0 3911 MHz
12 processor(s) found, POPCNT available
NUMA configuration with 2 node(s), offset 0
4096 MB Large Page Hash
Nalimov 6 men EGTB available - 512 MB cache
Engine: Houdini 2.0c Pro x64 (4096 MB)
by Robert Houdart

multi pv over 15 hours with 12 cores overnight

39 947:21 +3.56 44.Kh1 Rb8 45.Qd7+ Kg8 46.Ra7 Bf8 47.d6 Kh7 48.Bd5 Qe3 49.Qf7 Qc1+ 50.Kh2 Qf4+ 51.g3 Qf2+ 52.Bg2 e4 53.Qc7 Qb6 54.d7 Qxc7 55.Rxc7 Rd8 56.Bxe4 Bd6 57.Rb7 (2015.647.812.884) 35460 TB:2.710
39 947:21 +3.16 44.Kh2 Rb8 45.Qd7+ Kg8 46.Ra7 Bf8 47.Qf7+ Kh8 48.d6 Qf2 49.Bd5 Qf4+ 50.g3 Qf2+ 51.Bg2 Qb6 52.Rd7 e4 53.Bxe4 Qf2+ 54.Bg2 Ra8 55.Qd5 Re8 56.Rf7 Kh7 57.Rxf6 (2015.647.812.884) 35460 TB:2.710
39 947:21 +3.16 44.Kf1 Rb8 45.Qd7+ Kg8 46.Ra7 Bf8 47.Qf7+ Kh8 48.d6 Qe3 49.Bd5 Qc1+ 50.Kf2 Qd2+ 51.Kg1 Qc1+ 52.Kh2 Qf4+ 53.g3 Qf2+ 54.Bg2 Qb6 55.Rd7 e4 56.Bxe4 Qf2+ 57.Bg2 (2015.647.812.884) 35460 TB:2.710
38 947:21 +1.81 44.Qxb6 Rxb6 45.Kf2 Ke7 46.Kf3 Bc7 47.Ra7 Kd8 48.Kg4 Rd6 49.Kh5 Kc8 50.Ra8+ Bb8 51.Ra5 Rd7 52.Rxb5 Ba7 53.Kg6 Bf2 54.d6 Rxd6 55.Kxg7 h5 56.g4 hxg4 57.hxg4 (2015.647.812.884) 35460 TB:2.710
best move: Kg1-h1 time: 947:19.656 min n/s: 35.460.000 CPU 99.8% n/s(1CPU): 35.531.062 nodes: 2015.647.812.884 TB: 2.710

--------------------------------------------------------
33/90 41:43 +0.46 45...Qe3 46.Qxd6 Re8 47.Qc7+ Re7 48.Qc6 Qxe4 49.d6 Qd3+ 50.Kg1 Re8 51.Ra1 h5 52.Kh2 h4 53.Qxb5 Rd8 54.Qc6 Kf8 55.Qc7 Qxd6 56.Qxc4 e4+ 57.Kg1 Re8 58.Re1 Qe5 (90.912.489.213) 36310 TB:3

[Event "?"]
[Site "?"]
[Date "1997.05.04"]
[Round "2"]
[White "Deep Blue"]
[Black "Kasparov"]
[Result "1-0"]

1. e4 e5 2. Nf3 Nc6 3. Bb5 a6 4. Ba4 Nf6 5. O-O Be7 6. Re1
b5 7. Bb3 d6 8. c3 O-O 9. h3 h6 10. d4 Re8 11. Nbd2 Bf8
12. Nf1 Bd7 13. Ng3 Na5 14. Bc2 c5 15. b3 Nc6 16. d5 Ne7
17. Be3 Ng6 18. Qd2 Nh7 19. a4 Nh4 20. Nxh4 Qxh4 21. Qe2
Qd8 22. b4 Qc7 23. Rec1 c4 24. Ra3 Rec8 25. Rca1 Qd8 26. f4
Nf6 27. fxe5 dxe5 28. Qf1 Ne8 29. Qf2 Nd6 30. Bb6 Qe8
31. R3a2 Be7 32. Bc5 Bf8 33. Nf5 Bxf5 34. exf5 f6 35. Bxd6
Bxd6 36. axb5 axb5 37. Be4 Rxa2 38. Qxa2 Qd7 39. Qa7 Rc7
40. Qb6 Rb7 41. Ra8+ Kf7 42. Qa6 Qc7 43. Qc6 Qb6+ 44. Kf1
Rb8 45. Ra6