Armando Great outplayed Deep Fritz 11 in Playchess

[Harvey Williamson]

All,
According to this thread on Rybka Forum Armando Great was using an engine for this game (he is "Footbolo" nick in the engine room) but the GUI failed to display that fact ?? Look down towards the bottom of the thread.....


http://rybkaforum.net/cgi-bin/rybkaforu ... ?tid=10762

He is not that Armando. I can confirm this guy is a GM who wants to keep his identity secret.

[Dr.Wael Deeb]

All,
According to this thread on Rybka Forum Armando Great was using an engine for this game (he is "Footbolo" nick in the engine room) but the GUI failed to display that fact ?? Look down towards the bottom of the thread.....


http://rybkaforum.net/cgi-bin/rybkaforu ... ?tid=10762

I really doubt he used an engine. Really, give the guy some credit. I believe he won against Rybka honestly. It's hard for me to believe he cheated and got his 3217 rating dishonestly.

As Terry McCracken also pointed out in this thread, our software clearly its limits.

This means that from time to time or in some positions it will make weak moves. And no wonder that a strong human player like Oligarckh or Armando Great will be able to exploit this limits and defeat Rybka from time to time even if only 1 game out of 10.

Only a fool can disagree with this statement....but as you said Zlatnik:
From time to time....let's be more optimistic and say 2 wins and two draws at best for the human in a 10 games match....
Dr.D

[Cubeman]

I play on ICC and have a blitz rating that hovers around the 1800-1900.
I have a FIDE rating of approx 2050.
How can a person, even a top GM who is lets say 2800 standard OTB play chess, let alone blitz get rating of 3200?
Doesn't blitz make humans play weaker.Or is there no correlation between the Playchess rating and Fide ratings.
Also how long has ICC and Playchess be going for?I guess over 10 years and has any one tracked the best human player ratings during that time.Have the best human ratings during that time increased in line with the improved chess software.

[George Tsavdaris]

Some say he used Rybka but I doubt it because some of his moves in the game were clearly human moves which no computer currently plays.

This is not a proof that he used no computer. He may used computer for his 95% of the moves and played by himself for the rest 5%.
Or he may used the computer for blunder checking the moves and this is what he actually did i guess.
Or of course if he is a GM then he may won the game genuinely but this is just ONE game so heavily generalizing to this is completely stupid.

I believe computers need to be at least 1000 times faster and software more intelligent to even match the best human players.

I believe that computer with current hardware and current software have already matched best humans and they are better than the best humans.

This just proves that the best chess humans are still better and will outplay the computer most of the times.

No. This does not prove at all what you say.
How can you be so blind to believe this? From ONE GAME? In which the human most probably used a computer for blunder checking? Let's be serious.

Look for example results in ICC(look here for example), look at various Rybka games against IMs and GMs, look at the last 5 year games of computers against GMs and see the results.
A cosmogony for you i guess will occur!

Even I and I am a weak player could get a good position against Rybka in the Ruy Lopez (C65). Rybka had a pinned knight on f6 and played the weakening move g5. From there I got a better position but didn't play on.

I myself played 1.e4 against Rybka, a move that has very good results as white so i had a good position, but i didn't play on.

[gerold]

Armando Great outplayed Deep Fritz 11 yesterday in Playchess. Deep Fritz simply pushed his kingside pawns, weakening them and in the end lost them. So in the end Armando Great got 4 connected passed pawns.

Here's the game:

[Event "3m + 1s, rated"]
[Site "?"]
[Date "2009.04.24"]
[Round "?"]
[White "Comp Fritz, Deep Fritz 11"]
[Black "Armando Great"]
[Result "0-1"]
[ECO "D11"]
[WhiteElo "3333"]
[BlackElo "3217"]
[PlyCount "90"]
[EventDate "2009.04.24"]
[EventType "blitz"]

1. Nf3 d5 2. d4 Nf6 3. c4 c6 4. e3 Bg4 5. h3 Bh5 6. Nc3 e6 7. g4 Bg6 8. Ne5 Nbd7 9. Nxg6 hxg6 10. Bd3 Bd6 11. c5 Bc7 12. Bd2 e5 13. Qb3 b5 14. g5 Nh5 15. a4 exd4 16. exd4 b4 17. Ne2 a5 18. O-O-O O-O 19. f4 Re8 20. Rhf1 Nf8 21. h4 Qd7 22. f5 Rxe2 23. Bxe2 Ng3 24. Qd3 gxf5 25. Rfe1 Ne4 26. h5 Re8 27. h6 g6 28. Bf3 Ne6 29. b3 Kh7 30. Bxe4 dxe4 31. Qc4 Rd8 32. Kb1 Qxd4 33. Qxd4 Rxd4 34. Bc1 Be5 35. Rf1 Rxd1 36. Rxd1 f4 37. Rd7 Nxg5 38. Re7 Bf6 39. Rc7 e3 40. Kc2 e2 41. Bd2 f3 42. Kd3 Bc3 43. Re7 Bxd2 44. Kxd2 Kxh6 45. Ke1 Ne6 Comp Fritz,Deep Fritz 11 gibt auf (Lag: Av=0.45s, max=2.9s) 0-1

Some say he used Rybka but I doubt it because some of his moves in the game were clearly human moves which no computer currently plays. Armando Great is clearly one of the best chess players, definitely on the level of Vugar Gashimov and Raffael who also play there sometimes.

I believe computers need to be at least 1000 times faster and software more intelligent to even match the best human players. This just proves that the best chess humans are still better and will outplay the computer most of the times. I believe computers will one day equal the best chess humans but they will never be better. Human will always be one level higher than computer in some way.

Even I and I am a weak player could get a good position against Rybka in the Ruy Lopez (C65). Rybka had a pinned knight on f6 and played the weakening move g5. From there I got a better position but didn't play on.

Why you didn/t play on. These computers will trick you,
They can read your mind

[Lion]

Kramnik already played Deep Fritz and lost convincelly !

Humans can still do learns things to computers, but even free chess programs on slow hardware are stronger than best humans.

the race is finished.

[Dr.Wael Deeb]

Kramnik already played Deep Fritz and lost convincelly !

Humans can still do learns things to computers, but even free chess programs on slow hardware are stronger than best humans.

the race is finished.

Beautifully said
Dr.D

[MattieShoes]

Ratings on ICC and FIDE are not very comparable. It's a different pool of players and a different rating system.

Even if the average rating is the same in both places (I have no idea whether it is), the ratings on ICC are obviously more spread out, resulting in more players at both extremes. I haven't checked ICC but blitz ratings tend to be more spread out than standard ones too, so this magnifies the differences at both ends of the spectrum too.

[Terry McCracken]

Kramnik already played Deep Fritz and lost convincelly !


False! Kramnik in a better position overlooked or better forgot the Mate in One.

Humans can still do learns things to computers, but even free chess programs on slow hardware are stronger than best humans.

False! Weak programs on weak hardware are just that, weak!

the race is finished.

For you that happened a longtime ago.