The blunders in 1995 chess programs, Deep Blue vs fritz

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

Moderators: hgm, Rebel, chrisw

kgburcham
Posts: 2016
Joined: Sun Feb 17, 2008 4:19 pm

The blunders in 1995 chess programs, Deep Blue vs fritz

Post by kgburcham »

several blunders in this game on both sides. Reminds me of the "Special Olympics", but everyone has to start somewhere.
Here are two blunders.

Fritz 90mhz P5

Description given in 1995:
Deep Blue Prototype consists of an IBM RS/6000 workstation with 14 chess search engines as slave processors. Each processor contains a VLSI chip for move generation, as well as additional hardware for search and evaluation. Each Deep Thought 2 processor searches about 500,000 positions per second standalone, or about 400,000 positions per second as a slave processor. (This is about 1/10th of the projected speed of the Deep Blue single-processor currently in fabrication.) The 14- processor Deep Thought 2 typically searches between 3 and 5 million positions per second. When conducting a search, the search tree near the root position is processed on the host workstation, and includes selective search extension algorithms such as singular extensions. The deepest nodes in the search tree are handled by the slave search engines which usually do 4-ply alpha-beta searches.


Before 14.Kh1
[D] r2qkbr1/5p1p/p1npb3/1p1Np2Q/4Pp2/N2B4/PPP2PPP/R4RK1 w q -

Houdini 2.0c Pro x64

25/70 6:13 -0.34 14.c4 b4 15.Nc2 Bg4 16.Qxh7 Rg6 17.g3 Ra7 (9.256.205.965) 24786
26/70 10:05 -0.40-- 14.c4 b4 (15.103.370.825) 24954
26/70 12:49 -0.34 14.h3 Rg5 15.Nc7+ Kd7 16.Nxe6 fxe6 17.Qd1 Qf6 (19.356.005.581) 25139
27/70 15:46 -0.27++ 14.h3 (23.882.787.393) 25234
27/70 16:36 -0.25 14.h3 Rg5 15.Nc7+ Kd7 16.Nxe6 fxe6 17.Qd1 Qf6 (25.183.376.965) 25266
28/70 19:42 -0.29 14.h3 Rg5 15.Nc7+ Kd7 16.Nxe6 fxe6 17.Qd1 Qf6 (29.986.230.198) 25367
29/70 26:33 -0.29 14.h3 Rg5 15.Nc7+ Kd7 16.Nxe6 fxe6 17.Qd1 Qf6 (40.830.387.885) 25624
29/70 32:01 -0.29 14.h3 Rg5 15.Nc7+ Kd7 16.Nxe6 fxe6 17.Qd1 Qf6 (49.550.288.729) 25783


After 14.Kh1
[D] r2qkbr1/5p1p/p1npb3/1p1Np2Q/4Pp2/N2B4/PPP2PPP/R4R1K b q -

Houdini 2.0c Pro x64

17/42 0:01 -0.66++ 14...Bg4 (26.333.255) 17661
17/48 0:02 -1.08++ 14...Bg4 (39.275.796) 18344
17/53 0:02 -1.04 14...Bg4 15.Qxh7 Rg6 16.h3 Be6 17.Nxb5 axb5 18.Bxb5 Rc8 (44.299.212) 18162
19/53 0:03 -1.04 14...Bg4 15.Qxh7 Rg6 16.h3 Be6 17.Nxb5 axb5 (70.601.390) 19854
20/53 0:05 -1.13++ 14...Bg4 (113.385.230) 21213
20/67 0:29 -1.76 14...Bg4 15.Qxh7 Rg6 16.Nxb5 axb5 17.Bxb5 Rc8 18.Nb4 Bd7 (700.552.368) 23617
21/67 0:31 -1.76 14...Bg4 15.Qxh7 Rg6 16.Nxb5 axb5 17.Bxb5 Rc8 (742.298.006) 23606
22/67 0:36 -1.88++ 14...Bg4 (860.329.484) 23716
23/67 0:51 -1.99 14...Bg4 15.Qxh7 Rg6 16.Nxb5 axb5 17.Bxb5 Rc8 (1.236.985.144) 23891
24/67 1:09 -2.04 14...Bg4 15.Qxh7 Rg6 16.Nxb5 axb5 17.Bxb5 Rc8 (1.699.368.222) 24359
25/67 1:24 -2.04 14...Bg4 15.Qxh7 Rg6 16.Nxb5 axb5 17.Bxb5 Rc8 (2.082.558.861) 24581
26/67 2:25 -2.16++ 14...Bg4 (3.662.770.877) 25157
26/75 3:51 -2.93++ 14...Bg4 (5.868.594.904) 25297
26/80 7:52 -3.27 14...Bg4 15.Qxh7 Rg7 16.Qh8 Rg6 17.c3 Rh6 (12.070.540.282) 25541
27/80 8:18 -3.27 14...Bg4 15.Qxh7 Rg7 16.Qh8 Rg6 17.c3 Rh6 (12.728.926.923) 25552
28/80 11:56 -3.24 14...Bg4 15.Qxh7 Rg7 16.Qh8 Rg6 17.c3 Rh6 (18.509.761.359) 25833
29/80 16:34 -3.24 14...Bg4 15.Qxh7 Rg7 16.Qh8 Rg6 17.c3 Rh6 (25.968.492.761) 26101
29/80 18:42 -3.24 14...Bg4 15.Qxh7 Rg7 16.Qh8 Rg6 17.c3 Rh6 18.Nf6+ Rxf6 (29.367.044.068) 26173


Before 16.c4
[D] 2rqkb2/5p1p/p1npb1r1/1p1Np3/4Pp2/N2B4/PPP2PPP/R2Q1R1K w - -

23/51 0:07 -0.40++ 16.f3 (179.187.608) 23271
23/52 0:10 -0.46 16.f3 Qh4 17.Qe1 Qh5 18.Rf2 Rh6 19.Qg1 Bxd5 (253.156.244) 23555
24/57 0:17 -0.46 16.f3 Qh4 17.Qe1 Qh5 18.Rf2 Rh6 19.Qg1 Bxd5 (410.916.589) 23375
25/60 1:08 -0.39++ 16.f3 (1.700.980.096) 24710
25/60 1:10 -0.45 16.f3 Qh4 17.Qe1 Qh5 18.Rf2 Rh6 19.Qg1 Bxd5 (1.733.781.555) 24680
26/60 1:33 -0.52-- 16.f3 Qh4 (2.289.750.958) 24483
26/62 2:26 -0.53 16.f3 Qh4 17.Qe1 Qh5 18.Rf2 Rh6 19.Qg1 Bxd5 (3.616.166.958) 24673
27/70 3:39 -0.53 16.f3 Rh6 17.Rf2 Qh4 18.Qg1 Be7 19.Nxe7 Nxe7 (5.435.735.283) 24778
27/70 4:03 -0.53 16.f3 Rh6 17.Rf2 Qh4 18.Qg1 Be7 19.Nxe7 Nxe7 (6.046.509.939) 24860

After 16.c4

17/50 0:01 -5.63 16...Qh4 17.g3 Qh3 18.Qf3 Bg4 19.Qg2 Qh5 (25.195.321) 19996
18/50 0:02 -5.68 16...Qh4 17.Be2 Rh6 18.Bh5 Bxd5 19.exd5 Rxh5 20.h3 Ne7 (48.219.401) 21364
19/50 0:03 -5.68 16...Qh4 17.Be2 Rh6 18.Bh5 Bxd5 19.exd5 Rxh5 20.h3 Ne7 (74.855.416) 22245
20/50 0:05 -5.68 16...Qh4 17.Be2 Rh6 18.Bh5 Bxd5 19.exd5 Rxh5 20.h3 Ne7 (122.388.660) 23061
21/55 0:06 -5.41-- 16...Qh4 17.Be2 (155.203.041) 22644
21/55 0:10 -5.33 16...Qh4 17.Be2 Rh6 18.Bh5 Bxd5 19.Qg4 Qxh5 (254.959.985) 23195
22/56 0:17 -5.58++ 16...Qh4 (426.752.219) 23709
22/63 0:23 -5.66 16...Qh4 17.Be2 Rh6 18.Bh5 Bxd5 19.Qg4 Qxh5 (566.000.102) 23849
23/72 0:36 -5.71 16...Qh4 17.Be2 Rh6 18.Bh5 Bxd5 19.Qg4 Qxh5 (869.519.673) 24145
24/72 1:06 -5.80 16...Qh4 17.Be2 Rh6 18.Bh5 Bxd5 19.Qg4 Qxh5 (1.631.522.066) 24624
25/72 1:55 -5.86 16...Qh4 17.Be2 Rh6 18.Bh5 Bxd5 19.Qg4 Qxh5 (2.885.841.944) 25030
26/72 4:23 -6.11++ 16...Qh4 (6.721.822.015) 25556
26/72 8:09 -6.60++ 16...Qh4 (12.673.599.238) 25874
26/72 8:53 -6.60 16...Qh4 17.Be2 Rh6 18.Bh5 Bxd5 19.Qg4 Qxh5 (13.800.805.364) 25891


[Event "Hong Kong WCCC"]
[Site "Hong Kong WCCC"]
[Date "1995.??.??"]
[Round "5"]
[White "Deep Blue (Computer)"]
[Black "Fritz (Computer)"]
[ECO "B33"]
[Result "0-1"]

1. e4 c5 2. Nf3 Nc6 3. d4 cxd4 4. Nxd4 Nf6 5. Nc3 e5
6. Ndb5 d6 7. Bg5 a6 8. Na3 b5 9. Bxf6 gxf6 10. Nd5 f5
11. Bd3 Be6 12. Qh5 f4 13. O-O Rg8 14. Kh1 Rg6 15. Qd1 Rc8
16. c4 Qh4 17. g3 Qh3 18. Qd2 f3 19. Rg1 Rh6 20. Qxh6 Qxh6
21. cxb5 Bxd5 22. exd5 Nb4 23. Bf5 Rc5 24. bxa6 Nxa6
25. Nc2 Qd2 26. Ne1 Rxd5 27. Nxf3 Qxf2 28. Be4 Ra5 29. Rg2
Qe3 30. Re1 Qh6 31. Bc6+ Kd8 32. a3 f5 33. Rc2 Rc5 34. Rxc5
Nxc5 35. Rf1 Be7 36. a4 f4 37. gxf4 Qxf4 38. Rg1 Nxa4
39. b4 Qxb4 0-1
Top
rbarreira
Posts: 900
Joined: Tue Apr 27, 2010 3:48 pm

Re: The blunders in 1995 chess programs, Deep Blue vs fritz

Post by rbarreira »

Feng-Hsiung Hsu posted some interesting notes on this championship, in regard to the c4 move he said:

http://groups.google.com/group/rec.game ... fac76c9032
Instead of c4? allowing Qh5, the immediate g3 appears to hold--
might still be lost against the likes of Kasparov, but I doubt Fritz can push
it through. The machine did play g3, but the phone was disconnected at the
worst possible time, and when it was restarted, it did not have enough time
to rediscover the move.
What does Houdini think about g3?
Top
kgburcham
Posts: 2016
Joined: Sun Feb 17, 2008 4:19 pm

Re: The blunders in 1995 chess programs, Deep Blue vs fritz

Post by kgburcham »

rbarreira wrote:Feng-Hsiung Hsu posted some interesting notes on this championship, in regard to the c4 move he said:

http://groups.google.com/group/rec.game ... fac76c9032
Instead of c4? allowing Qh5, the immediate g3 appears to hold--
might still be lost against the likes of Kasparov, but I doubt Fritz can push
it through. The machine did play g3, but the phone was disconnected at the
worst possible time, and when it was restarted, it did not have enough time
to rediscover the move.
What does Houdini think about g3?


[D] 2rqkb2/5p1p/p1npb1r1/1p1Np3/4Pp2/N2B4/PPP2PPP/R2Q1R1K w - -

28 29:42 -0.61 16.f3 Rh6 17.Rf2 Qh4 (45.850.264.470) 25718
28 29:42 -1.20 16.g3 Rh6 17.Rg1 Qg5 (45.850.264.470) 25718
28 29:42 -1.39 16.g4 Bxg4 17.f3 Bh3 18.Rg1 Qh4 19.Bf1 f5 (45.850.264.470) 25718
28 29:42 -1.65 16.c3 Rh6 17.g3 Qg5 18.Nc2 Ne7 19.Ncb4 Nxd5 (45.850.264.470) 25718
Top
zullil
Posts: 6442
Joined: Tue Jan 09, 2007 12:31 am
Location: PA USA
Full name: Louis Zulli

Re: The blunders in 1995 chess programs, Deep Blue vs fritz

Post by zullil »

rbarreira wrote:Feng-Hsiung Hsu posted some interesting notes on this championship, in regard to the c4 move he said:

http://groups.google.com/group/rec.game ... fac76c9032
Instead of c4? allowing Qh5, the immediate g3 appears to hold--
might still be lost against the likes of Kasparov, but I doubt Fritz can push
it through. The machine did play g3, but the phone was disconnected at the
worst possible time, and when it was restarted, it did not have enough time
to rediscover the move.
What does Houdini think about g3?
[D]2rqkb2/5p1p/p1npb1r1/1p1Np3/4Pp2/N2B2P1/PPP2P1P/R2Q1R1K b - - 0 16

Stockfish-2.2.2 (8 threads) says White is losing after Rh6:

Code: Select all

Searching: 2rqkb2/5p1p/p1npb1r1/1p1Np3/4Pp2/N2B2P1/PPP2P1P/R2Q1R1K b - - 0 16
infinite: 1 ponder: 0 time: 0 increment: 0 moves to go: 0
 1   +0.44   00:01     163  Bh3 
 2   +0.36   00:01    1546  Bxd5 exd5 
 3   +0.36   00:01    2754  Nd4 c3 Bxd5 exd5 
 4   +0.48   00:01    4758  Bh3 Rg1 Qg5 c4 fxg3 fxg3 
 5   +0.65   00:01    7238  Ne7 c4 Bh3 Rg1 Nxd5 cxd5 fxg3 fxg3 
 6   +0.65   00:01    9822  Ne7 c4 Bh3 Rg1 Nxd5 cxd5 fxg3 fxg3 
 7   +0.44   00:01   22562  Ne7 Nb4 fxg3 fxg3 Bh3 Rf2 Qb6 
 8   +0.28   00:01   52690  Ne7 Nb4 fxg3 fxg3 Qa5 c3 Bh3 Rg1 f5 Qh5 fxe4 Qxh3 
                            exd3 Nxd3 
 9   +0.57   00:01  165632  b4 Nb1 Qa5 gxf4 Bxd5 exd5 Qxd5+ f3 Nd4 Bxg6 hxg6 
                            fxe5 dxe5 
10   +0.32   00:01  495480  Bh3 Rg1 fxg3 Rxg3 Qh4 c4 Ne7 Nxe7 Bxe7 cxb5 axb5 
                            Nxb5 
11   +0.24   00:01  965260  Bh3 Rg1 Nd4 gxf4 Qh4 Rg3 exf4 Rxh3 Qxh3 Nxf4 Qf3+ 
                            Qxf3 Nxf3 Nxg6 hxg6 
12   +0.48   00:01   1475K  Ne7 c4 Bh3 Rg1 Nxd5 exd5 fxg3 Bxg6 g2+ Rxg2 Bxg2+ 
                            Kxg2 Qg5+ Kf1 hxg6 cxb5 
13   +0.40   00:01   2469K  Ne7 c4 bxc4 Bxc4 Nxd5 Bxd5 Bxd5 Qxd5 Rc5 Qb3 d5 gxf4 
                            Rb6 Qd3 dxe4 Qxe4 
14   +0.40   00:01   2718K  Ne7 c4 bxc4 Bxc4 Nxd5 Bxd5 Bxd5 Qxd5 Rc5 Qb3 d5 gxf4 
                            Rb6 Qd3 dxe4 Qxe4 
15   +0.40   00:01   3695K  Ne7 c4 bxc4 Bxc4 Nxd5 Bxd5 Bxd5 Qxd5 Rc5 Qb3 d5 gxf4 
                            Rb6 Qd3 dxe4 Qxe4 Qd5 
16   +0.40   00:01   4328K  Ne7 c4 bxc4 Bxc4 Nxd5 Bxd5 Bxd5 Qxd5 Rc5 Qb3 d5 gxf4 
                            Rb6 Qd3 dxe4 Qxe4 Qd5 
17   +0.57   00:02   7119K  Ne7 c4 bxc4 Bxc4 Nxd5 Bxd5 Bxd5 Qxd5 Rc5 Qb3 d5 gxf4 
                            Rb6 Qa4+ Qd7 Qxd7+ Kxd7 Rfd1 exf4 b3 
18   +0.20   00:05  29232K  Bh3 c4 Bxf1 Bxf1 b4 Nc2 Rb8 Bh3 Rh6 Bf5 a5 Ne1 fxg3 
                            fxg3 b3 Nf3 bxa2 Rxa2 Bg7 Qa4 
19   +0.24   00:10  67106K  Qg5 c4 Ne7 Nb6 Rc6 Nd5 bxc4 Bxc4 Qg4 Qxg4 Bxg4 f3 
                            Bh3 Rg1 fxg3 Rxg3 Be6 Kg2 Bh6 Rxg6 hxg6 Nf6+ Kd8 
                            Bxe6 fxe6 
20   +0.24   00:11  78039K  Qg5 c4 Ne7 Nb6 Rc6 Nd5 bxc4 Bxc4 Qg4 Qxg4 Bxg4 f3 
                            Bh3 Rg1 fxg3 Rxg3 Be6 Kg2 Bh6 Rxg6 hxg6 Nf6+ Kd8 
                            Bxe6 fxe6 
21   +0.40   00:13  94815K  Rh6 c4 b4 Nc2 Qg5 Kg1 a5 b3 Ne7 Nb6 Rc6 Nd5 Bh3 Re1 
                            Nxd5 exd5 fxg3 fxg3 Rc7 Ne3 Rf6 Qe2 Bh6 Kh1 
22   +1.21   00:32 238306K  Qg5 c3 Rh6 Kg1 Ne7 Re1 Nxd5 exd5 fxg3 fxg3 Bg4 Qc2 
                            Be7 Qg2 Bh3 Qf2 Rf6 Qc2 h5 Bf1 Bg4 Bg2 Qf5 Qxf5 Rxf5 
23   +1.49   00:40 307215K  Qg5 c3 Rh6 Kg1 Ne7 Re1 Nxd5 exd5 fxg3 fxg3 Bg4 Qc2 
                            Be7 Qg2 Bh3 Qe2 f5 Qf2 f4 Be4 Rf6 Bxh7 Qh5 Be4 Rf7 
                            Qe2 Bg4 Qf2 fxg3 Qxg3 
24   +1.70   01:10 544211K  Qg5 Nxb5 axb5 Bxb5 Bg4 f3 Bd7 Rg1 fxg3 Rxg3 Qh4 Rxg6 
                            hxg6 a4 Rb8 c3 Bh6 Qe1 Qxe1+ Rxe1 Kd8 Nf6 Bg5 Nxd7 
                            Kxd7 Kg2 Bf4 Rd1 Rh8 h3 f5 
25   +1.78   02:29   1191M  Qg5 Rg1 Rh6 Rg2 Bh3 c3 Ne7 Be2 f5 exf5 Rc5 Nxe7 Bxe7 
                            Rg1 Bxf5 g4 Qh4 Rg2 Bd7 Qd2 Qh3 f3 Rc8 c4 Bc6 cxb5 
26   +1.45   03:18   1589M  Qg5 Rg1 Rh6 Rg2 Bh3 Nb1 Be7 Nd2 Rb8 Qg1 Rg6 Bf1 
                            Bxg2+ Qxg2 Qh6 Rd1 fxg3 fxg3 Bd8 Nf3 Ne7 Nxe7 Bxe7 
                            Qf2 Kf8 c4 bxc4 Bxc4 Qh5 b3 f5 exf5 Qxf5 Bxa6 
27   +1.33   03:41   1770M  Qg5 Rg1 Rh6 Rg2 Bh3 Nb1 Bg7 Nd2 Bxg2+ Kxg2 Ne7 Nxe7 
                            Qxe7 Qg4 Qe6 Qxe6+ fxe6 Nf3 Bf6 b3 Kf7 a4 Be7 axb5 
                            axb5 Ra5 b4 Ra4 Rb8 Bc4 fxg3 
28   +1.29   05:35   2711M  Qg5 Rg1 Rh6 Rg2 Bh3 Nb1 Bg7 Nd2 Bxg2+ Kxg2 Ne7 Nxe7 
                            Qxe7 Qg4 Qe6 Qxe6+ fxe6 b4 fxg3 hxg3 Bf8 a4 d5 axb5 
                            Bxb4 Nf3 axb5 Nxe5 Bc3 
29   +1.25   06:14   3032M  Qg5 Rg1 Rh6 Rg2 Bh3 Nb1 Bg7 Nd2 Bxg2+ Kxg2 Ne7 Nxe7 
                            Qxe7 Qg4 Qe6 Qxe6+ fxe6 b4 fxg3 hxg3 Bf8 a4 d5 axb5 
                            Bxb4 Nf3 axb5 Nxe5 Bc3 Bxb5+ Kf8 Nd7+ Ke7 Nb6 
30   +1.98   16:16   8038M  Rh6 Kg1 Qg5 c3 Ne7 Nc2 Nxd5 exd5 Bg4 
31   +1.94   38:53  19498M  Rh6 c3 Qg5 Kg1 Ne7 Nc2 Nxd5 exd5 Bg4 f3 Bd7 Qd2 Qg8 
                            g4 f5 g5 Be7 Kh1 Bxg5 Qg2 Rg6 a4 Bd8 Qh3 e4 Be2 bxa4 
                            Nb4 Rb8 Rfb1 Rb6 Bxa6 e3 Bd3 Qf7 Bc2 Rb7 Bd3 
32   +1.94   41:58  21048M  Rh6 c3 Qg5 Kg1 Ne7 Nc2 Nxd5 exd5 Bg4 f3 Bd7 Qd2 Qg8 
                            g4 f5 g5 Be7 Kh1 Bxg5 Qg2 Rg6 a4 Bd8 Qh3 e4 Be2 bxa4 
                            Nb4 Rb8 Rfb1 Rb6 Bxa6 e3 Bd3 Qf7 Bc2 Rb7 Bd3 
33   +1.86   48:34  24395M  Rh6 c3 Qg5 Kg1 Ne7 Nc2 Nxd5 exd5 Bg4 f3 Bd7 Qd2 Qg8 
                            g4 f5 g5 Rg6 Be2 Be7 Kh1 Bxg5 a4 Bd8 a5 Rc5 Nb4 Bc8 
                            Bd3 Qf7 Qc2 Bh4 Rg1 Kf8 Rg2 Kg8 b3 
34   +1.86 1:22:18  41675M  Rh6 Kg1 Qg5 c3 Ne7 Nc2 Nxd5 exd5 Bg4 f3 Bd7 Qd2 
35   +1.98 3:30:32 108318M  Rh6 Rg1 Nd4 c3 Qg5 Rg2 f3 cxd4 fxg2+ Kxg2 f5 Kg1 
                            Bxd5 exd5 e4 f4 Qh5 Qxh5+ Rxh5 Be2 Rh6 Kg2 Kd7 Rd1 
                            Bg7 g4 Bf6 g5 Rg6
Top
attakinski
Posts: 39
Joined: Sat May 21, 2011 12:23 pm

Re: The blunders in 1995 chess programs, Deep Blue vs fritz

Post by attakinski »

funny.. I made the book manually and never before saw the remarks of David Byrne and Hsu.

In my opinion is missing a game from a candidates match (it was a game from Vaganian) a sign of bad preparation by the Deep Blue team.
Top

Return to “Computer Chess Club: General Topics”