Fritz12 Pentium 4 Kills Rybka 3 running on Extreme X9775@3.2

[Father]

INTEL (R)
PENTIUM (R) 4CPU 1.60 GHz
1.60GHz, 640 MB de RAM
Versus
Rybka 3
Intel(R) Core(TM)2 Extreme CPU X9775


I have a question please:
My old computer, b]INTEL (R)
PENTIUM (R) 4CPU 1.60 GHz
1.60GHz, 640 MB de RAM [/b]
has killed Rybka 3 running on Intel(R) Core(TM)2 Extreme CPU X9775 [/b]



Summary. An old hardware running with the beutiful Queen Fritz 12, has won to Rybka 3 running on a super hardware... I would like to know why? Tnhaks in advance... This game is absolutly against the statistics and the logical

[d] 2r2rk1/pb1nqpp1/1ppbpn1p/3p4/2PPP3/1PNB1N2/PBQ2PPP/3R1RK1 b - - 0 13[d]

[d]2r2rk1/pb2qpp1/1p2pn1p/2pP4/2P5/bPBB1N2/P1Q2PPP/3R1RK1 b - - 0 18[d]

[d]3r1r1k/pb2NpB1/1p5p/2p5/1n2Bq2/bP6/P1Q2PPP/3RR1K1 b - - 0 25[d]

[d]3r4/3qk3/pp3p2/2p1bB1p/3N4/1P4PQ/5P1P/4R1K1 b - - 0 47[d]

[d]7r/7P/8/7R/4k1K1/6P1/4B3/8 b - - 0 83[d]







[Event "Rated game, 3m + 0s"]
[Site "Engine Room"]
[Date "2010.03.17"]
[Round "?"]
[White "Father, Fritz 12"]
[Black "Rybka33, Rybka 3"]
[Result "1-0"]
[ECO "D45"]
[WhiteElo "2099"]
[BlackElo "2742"]
[Annotator "0.37;-0.15"]
[PlyCount "165"]
[EventDate "2010.03.17"]
[TimeControl "180"]

{Rybka 3 (8 cores): 14.9 plies; 479kN/s Intel(R) Core(TM)2 Extreme CPU X9775
@ 3.20GHz, Fritz12.ctg, 100 MB} 1. d4 {B 0} c6 {B 0 (Cf6)} 2. Nf3 {B 0} d5 {B
0 (Cf6)} 3. c4 {B 0} Nf6 {B 0} 4. Nc3 {B 0} e6 {B 0} 5. e3 {B 0} Nbd7 {B 0} 6.
Qc2 {B 0} Bd6 {B 0} 7. b3 {B 0} O-O {B 0} 8. Be2 {B 0} b6 {B 0} 9. O-O {B 0}
Bb7 {B 0} 10. Bb2 {B 0} Qe7 {B 0} 11. Rad1 {B 0} h6 {B 0 (Tad8)} 12. Bd3 {0.37/
12 7} Rac8 {-0.15/13 4 (Tad8)} 13. e4 {0.42/14 23 (h3)} dxe4 {-0.09/12 3 (e5)}
14. Nxe4 {0.44/14 3 (Lxe4)} Nxe4 {0.03/14 2} 15. Bxe4 {0.44/13 0} Nf6 {0.10/15
10} 16. Bd3 {0.54/13 1} Ba3 {0.07/15 6 (Tfd8)} 17. Bc3 {0.67/13 6 (Se5)} c5 {
0.00/13 2 (Ab4)} 18. d5 {0.50/13 6} exd5 {0.00/15 0} 19. Rfe1 {0.20/12 4} Qd6 {
0.00/15 0} 20. cxd5 {0.46/12 3} Nxd5 {0.19/15 0} 21. Be4 {0.92/12 2} Rcd8 {0.
19/15 0 (c4)} 22. Nh4 {0.47/12 2} Qf4 {0.19/16 20} 23. Nf5 {0.40/15 0 (Lxd5)}
Nb4 {0.14/15 19} 24. Ne7+ {0.56/16 0} Kh8 {0.14/14 0} 25. Bxg7+ {0.37/16 14}
Kxg7 {0.26/16 0} 26. Qc3+ {0.34/13 1} Qf6 {0.26/16 16} (26... f6 {0} 27. Bxb7 {
2}) 27. Nf5+ {0.41/15 0} Kg6 {0.26/16 14} 28. Qh3 {0.50/15 0} h5 {0.11/12 1} (
28... Bxe4 {0} 29. Qxh6+ {2} Kxf5 {6} 30. g4+ {0} Ke5 {11} 31. Rxe4+ {0} Kxe4 {
4} 32. Qe3# {0}) 29. Nh4+ {0.46/12 1 (Lb1)} Kg7 {0.12/14 1} 30. Bxb7 {0.57/13 1
} Rxd1 {0.12/14 1} 31. Rxd1 {0.60/13 1} Rd8 {0.04/16 9} 32. Rf1 {0.59/14 0} Kf8
{0.04/15 0} 33. Nf3 {0.69/13 2 (Sf5)} Qh6 {0.26/13 2} 34. Qh4 {0.49/13 2} Re8 {
0.21/14 0} 35. Ng5 {0.36/14 7} f6 {0.27/15 0 (Ab2)} 36. Ne4 {0.38/12 7} Bb2 {
0.24/15 0 (Te7)} 37. a3 {0.89/12 3} Nc2 {0.27/15 6 (Cd3)} 38. Nd6 {0.44/13 6
(b4)} Rd8 {0.17/13 4} 39. Nf5 {0.48/14 0 (Sc4)} Qh7 {0.15/13 1} 40. Be4 {0.38/
14 9} Nxa3 {0.33/15 0} 41. g3 {0.35/13 4 (Df4)} a6 {0.34/13 2 (Ae5)} 42. Re1 {
0.54/12 2 (Df4)} Be5 {0.54/12 2 (Ac3)} 43. Qh3 {1.03/12 4} Nb5 {0.64/14 0 (Df7)
} 44. Nd4 {1.22/12 1 (Sh4)} Qd7 {0.19/9 0} 45. Ne6+ {0.69/13 6} Ke7 {0.96/14 0}
46. Bf5 {0.82/14 4} Nd4 {0.96/14 0} 47. Nxd4 {1.09/13 2} Qxd4 {0.96/14 5} 48.
Qxh5 {0.55/13 0} c4 {1.45/12 3 (Dd2)} 49. bxc4 {1.97/10 1 (Dh7+)} Qxc4 {1.74/
12 7} 50. f4 {2.24/13 0} Qc5+ {1.74/12 1} 51. Kh1 {2.17/13 1 (Kf1)} Qf2 {1.58/
11 1 (Dd5+)} 52. Qh7+ {2.75/12 2 (De2)} Kf8 {2.34/11 1} 53. Qh6+ {2.75/12 0}
Ke8 {2.53/11 0} 54. Bg6+ {2.75/12 0} Kd7 {2.69/12 0} 55. Qh7+ {1.88/12 3} Kc6 {
2.69/12 1} 56. Rc1+ {2.13/12 0} Kb5 {3.04/13 0} 57. Bd3+ {2.88/12 0} Ka5 {3.06/
13 2} 58. Rf1 {2.54/14 0} Qb2 {2.64/13 4 (De3)} 59. fxe5 {3.47/10 0} fxe5 {3.
46/14 2} 60. Rb1 {3.53/12 4} Qd4 {3.33/16 0} 61. Qf7 {3.73/13 1} Qd5+ {3.33/14
2} 62. Qxd5+ {3.84/15 0} Rxd5 {3.33/15 3} 63. Ra1+ {3.88/15 0} Kb4 {3.89/16 0}
64. Bxa6 {3.86/13 0} Ra5 {3.96/16 2 (b5)} 65. Rb1+ {3.79/13 1} Kc5 {3.96/16 0}
66. Be2 {3.90/13 6 (Ld3)} Ra2 {3.29/12 1} 67. Rb5+ {4.25/13 0} Kd4 {3.30/14 0}
68. Bg4 {4.43/13 0 (Lf3)} e4 {3.50/13 1} 69. Rb4+ {4.57/13 0} Ke3 {3.47/13 2}
70. Rxb6 {4.74/14 0} Ra1+ {4.11/14 2} 71. Kg2 {4.79/15 0} Ra2+ {4.28/16 1} 72.
Kh3 {4.83/13 0} Kf2 {4.28/15 0 (Tf2)} 73. Be6 {5.09/15 3} Rd2 {4.54/17 0} 74.
Bc4 {5.11/14 1} Rd8 {4.54/16 0 (Td4)} 75. Kg4 {5.46/13 1 (Tb2+)} Ke3 {4.83/11
0 (e3)} 76. h4 {6.15/12 1} Kd2 {5.03/10 0 (Rd4)} 77. h5 {6.68/12 1 (Tb4)} e3 {
4.90/9 2 (Rc3)} 78. h6 {7.55/12 1} Kc3 {6.25/10 0 (Td4+)} 79. Ba6 {7.89/13 1
(Le2)} Kd4 {5.73/8 0} 80. h7 {8.16/14 0} Rh8 {6.81/10 0} 81. Rh6 {9.68/12 0
(Tb7)} Ke4 {7.40/9 0 (Rd5)} 82. Rh5 {10.32/14 0 (Le2)} e2 {8.25/10 0 (Rd4)} 83.
Bxe2 {10.70/13 0 Rybka33,Rybka 3 gibt auf (Lag: Av=0.54s, max=1.3s)} 1-0

[Father]

Doctor
Robert Hyatt
My old hardware, a Pentium 4, over 6 years old, running with my new Fritz 12, has won the modern Rybka 3 running in a fashionable hardware; the old hardware has killed a better hardware and better software on it... This is really extrange for me... out of the Maths.And the game was really nice.
Thanks Doctor Robert for an explanation that you could give to me.
With best respect,
Pablo Ignacio Restrepo
Usser playchess.com as Father
March 17, 2010
Earth Planet

[Milos]

Doctor
Robert Hyatt
My old hardware, a Pentium 4, over 6 years old, running with my new Fritz 12, has won the modern Rybka 3 running in a fashionable hardware; the old hardware has killed a better hardware and better software on it... This is really extrange for me... out of the Maths.And the game was really nice.
Thanks Doctor Robert for an explanation that you could give to me.
With best respect,
Pablo Ignacio Restrepo
Usser playchess.com as Father
March 17, 2010
Earth Planet

I'm not Bob, still...
Statistics is for most of the ppl quite obscure. That doesn't mean it's hard or ugly...
Let's see. Imagine Rybka wins 90% of the matches in its super fast hardware against poor Fritz. This would mean, poor Fritz would win 1 point in every 10 games on average. Let assume that actually it wins 1 game and has 2 draws in 20 games played. This would mean that the actual chance of you seeing a win is 5%. Quite possible don't you agree?
Now, lets see what does 90% of win points mean... 90% is equal to 380 elo points of difference. Your time control of 3m/game on extreme hardware is somewhat faster than CCRL 40/4. Still there Rybka 3 is 250 Elo stronger than Fritz 12. So, we can say it's a good approximation.
Now your hardware is about 4 times stronger for Rybka compared to Fritz's hardware. This gives around 120-130 Elo points.
Now, lets add this together. 250elo+130elo=380elo.
And up there we have, how much...380elo. What a coincidence... .

[Vinvin]

I would like to know why?

The answer is easy : simply because sometime, the weakest player win.


[Event "Euro Clubs Cup (MEN)"]
[Site "Rethymnon, Crete (GRE)"]
[Date "2003.10.03"]
[EventDate "?"]
[Round "6"]
[Result "1-0"]
[White "Alexander Huzman"]
[Black "Garry Kasparov"]
[ECO "D45"]
[WhiteElo "2574"]
[BlackElo "2830"]
[PlyCount "43"]

1. Nf3 d5 2. d4 Nf6 3. c4 c6 4. Nc3 e6 5. e3 a6 6. b3 Bb4
7. Bd2 O-O 8. Bd3 Nbd7 9. Qc2 Bd6 10. Ne2 c5 11. O-O b6
12. cxd5 exd5 13. Ng3 Bb7 14. Nf5 Bc7 15. dxc5 bxc5 16. b4 c4
17. Be2 Ne4 18. Bc3 Nxc3 19. Qxc3 Nf6 20. Rfd1 Bc8 21. Rxd5
Qe8 22. Bxc4 1-0

[Father]

Hello Milos.
No no Milos.!!! Rybka 3 played in a super hardware, Fritz 12 played in my old Pentium 4, 6 years old!!! Then Rybka 3 had a better hardware that my Fritz 12 in an old hardware, but Fritz 12 won !!!

[Milos]

Hello Milos.
No no Milos.!!! Rybka 3 played in a super hardware, Fritz 12 played in my old Pentium 4, 6 years old!!! Then Rybka 3 had a better hardware that my Fritz 12 in an old hardware, but Fritz 12 won !!!

You didn't get it right?
Ok let's try again.
Your 6 years old single core cpu is around 20 times slower than your super extreme 4 core cpu. This you can check with any benchmark.
Now, if you look CCRL rating lists you will notice that Rybka 3 on 4 cpus is 250 elo points stronger than fritz 12 on a same hardware but single cpu.
That 6 years old P4 that you have is additionally 4-5 times slower than a single core of your super extreme hardware. This accounts for additional 130 elo points. In total 380 elo.
Even with this poor hardware, you poor Fritz has still 5% chance to win against might Rybka on mighty hardware.
I don't know if are able to process this information, but 5% winning chance is not a miracle. It's not even a coincidence. It's a something to be expected to see once in 2h of playing (with 3min/game TC).

[Father]

Milos.

I dont understand, how and why, Fritz 12 has produced a better and wonderful play that the game that Rybka 3 did. Fritz 12 played as a Super G.M. This game is an example of perfection. It was more than a simple coincidence...

This game has not been the sumarie of a long term in a end game plan or way. The game was a real example of a superior concept on strategic and tactics plans.

[Graham Banks]

Milos.

I dont understand, how and why, Fritz 12 has produced a better and wonderful play that the game that Rybka 3 did. Fritz 12 played as a Super G.M. This game is an example of perfection. It was more than a simple coincidence...

This game has not been the sumarie of a long term in a end game plan or way. The game was a real example of a superior concept on strategic and tactics plans.

As Milos said, these things happen from time to time.

[Milos]

I dont understand, how and why, Fritz 12 has produced a better and wonderful play that the game that Rybka 3 did. Fritz 12 played as a Super G.M. This game is an example of perfection. It was more than a simple coincidence...

The game is just an example of few crappy moves chosen by Rybka. Nothing more, nothing less. Why? Well, these are chess engines we are talking about, no miracle, no game plan, no ingenuity. It's just a brute force search and a good evaluation. Sorry to disappoint you, but this is all that chess engines are... The interpretation by humans is what gives them their soul and smartness, objectively they don't have any of it.

[Father]

Graham. This is new for me. I am not an expert in this knowlodge.