Can Rybka beat Aronian or Anand at this time control ?

[M ANSARI]

I remember that match by Rebel against Anand. Rebel won quite easily. Rybka as an engine is dramatically (and I mean dramatically) stronger than the Rebel that beat Anand. Rybka on today's hardware (say a Quadcore running at 3.2Ghz) would be maybe 10X or even 20X+ faster. A blitz time control today would easily be a long time control just a few years ago ... and fast time controls favor the engine much much more than it favors the human. One more thing, Anand was one of the new generation of GM's that used computers very heavily in his preparation ... maybe even more than Kasparov did. Although he is a tremendously talented player who can probably get a win every once in a while against a fully loaded Rybka ... in a match he would not stand a chance. I would say Rybka is at 200 ELO points ... and I am being very conservative.

[pichy]

I remember that match by Rebel against Anand. Rebel won quite easily. Rybka as an engine is dramatically (and I mean dramatically) stronger than the Rebel that beat Anand. Rybka on today's hardware (say a Quadcore running at 3.2Ghz) would be maybe 10X or even 20X+ faster. A blitz time control today would easily be a long time control just a few years ago ... and fast time controls favor the engine much much more than it favors the human. One more thing, Anand was one of the new generation of GM's that used computers very heavily in his preparation ... maybe even more than Kasparov did. Although he is a tremendously talented player who can probably get a win every once in a while against a fully loaded Rybka ... in a match he would not stand a chance. I would say Rybka is at 200 ELO points ... and I am being very conservative.

After looking at the Livingston Computer Chess960 World Championship Match Rybka vs Shredder, I believe that Rybka is strong enough to give Aronian a Pawn Odd .

http://chesstigers.de/ccm7/ccm7_games.php#cwm

[M ANSARI]

I think we are reaching a point where humans cannot really compete with today's engines on very powerful hardware. I still think that the top humans have a much better understanding of chess in general and are better at evaluating a single position than any engine. It is just that humans are horribly weak in comparison to the top engines on powerful hardware. The best way to describe it would be to guess that a top engine on powerful hardware is maybe around 2300 to 2400 ELO strength positionally ... and 3500 ELO tactically. While a top 10 GM is probably 3000 ELO positionally and only 2600 ELO tactically.

I think the future of human vs engine play would be some sort of Advanced Chess ... where the human is allowed to use an engine on modest hardware (say one core) vs. a fully loaded engine.

[Uri Blass]

I think we are reaching a point where humans cannot really compete with today's engines on very powerful hardware. I still think that the top humans have a much better understanding of chess in general and are better at evaluating a single position than any engine. It is just that humans are horribly weak in comparison to the top engines on powerful hardware. The best way to describe it would be to guess that a top engine on powerful hardware is maybe around 2300 to 2400 ELO strength positionally ... and 3500 ELO tactically. While a top 10 GM is probably 3000 ELO positionally and only 2600 ELO tactically.

I think the future of human vs engine play would be some sort of Advanced Chess ... where the human is allowed to use an engine on modest hardware (say one core) vs. a fully loaded engine.

I disagree here.

All you need to play chess well is to be able to evaluate positions correctly.
You have 30 possible positions that you need to evaluate assuming having 30 legal moves and you only need to evaluate which position is best.

Humans simply fail in this task relative to computers so they cannot claim that they evaluate positions better than computers.

The fact that computers find better positional moves thanks to search is not relevant and the simple fact is that they find better positional moves.

If GM's are allowed to take back moves that are tactical mistakes(computer evaluation jump by at least 0.5 pawn) when they will not going to be allowed to take back positional mistakes then I guess that they are still going to lose against machines in a match.

of course they may get an isolated pawn because of some tactics that they did not see earlier that they cannot take back but I do not call it tactics because tactics is only about combination that win material for no significant compensation.

Uri

[Ryan Benitez]

Can Rybka beat Aronian or Anand at this time control ?

In FRC? Easily. Unlike normal chess, I think Rybka could give draw odds and still be a heavy favorite. The novel positions will hurt the GMs more than Rybka.

Agreed.Question is if it can beat any of these GM's with only rook pawn odd.

My chess enthusiast friend, Anand would even take Rybka apart in a game of 25 minutes per side in normal chess without any odd. Just because Rybka performance against other chess programs has been excellent, it has never played the best blitz human player in the world.

My patzer friend,Whatever,but that's your opinion

PS: I also wish to the Rybka team luck against Joel Benjamin playing vs Rybka with a pawn odd, since Rybka is considered to be much stronger than Deeper Blue and Joel had an excellent 2-0 record against Deep Blue; and Benjamin was one of the key GM in tunning Deeper Blue versus Kasparov. Anyway in my opinion Anand is a much better player than Joel at any given time control and should score better than against Rybka

http://rybkaforum.net/cgi-bin/rybkaforu ... l?tid=1715

This is a pure speculation about Anand,the guy was sweaped away by the old program Rebel a long time ago

No wonder why Rebel was playing on a real monster machine the equivalent of an Intel Pentium II at 550 MHz

PS: The real factor of the slightly lost match by Anand was that Anand at that time did NOT had enough experience playing against computer chess programs, Plus jeroem cooked some openings that he knew in advadvance how Anand would respond to it up to move 20 or so.

http://www.rebel.nl/chespres.htm

Anand needs experience again(until he becomes a grandpa and around that time he would need need even more experience to play like young Anand who once was defeated by Rebel )
Anand doesn't have enough experience playing computers
because Rybka team has a monster machine and strength far better than Rebel and its machine long time back and much improved anti anand book from experienced Jereon again

You lost the argument,Give up Jorge.

Regards.

Why be disrespectful to someone based on a hypothetical that will likely never happen? Have you even considered that play conditions may be a factor in the games of engines playing GM's? This topic can be argued until the end of time with no one person being right. Is it fair that the book builder has access to a large list of the human players games and the human not have access to the engines opening book? Maybe, maybe not. It depends how you look at it. It could be argued that the engine also has publicly available games as it could also be argued that the opening prep is the most important part in a big game and will not be the same due to the situation. No one has lost anything here or has to give up.

[Ryan Benitez]

Can Rybka beat Aronian or Anand at this time control ?

In FRC? Easily. Unlike normal chess, I think Rybka could give draw odds and still be a heavy favorite. The novel positions will hurt the GMs more than Rybka.

Agreed.Question is if it can beat any of these GM's with only rook pawn odd.

My chess enthusiast friend, Anand would even take Rybka apart in a game of 25 minutes per side in normal chess without any odd. Just because Rybka performance against other chess programs has been excellent, it has never played the best blitz human player in the world.

My patzer friend,Whatever,but that's your opinion

PS: I also wish to the Rybka team luck against Joel Benjamin playing vs Rybka with a pawn odd, since Rybka is considered to be much stronger than Deeper Blue and Joel had an excellent 2-0 record against Deep Blue; and Benjamin was one of the key GM in tunning Deeper Blue versus Kasparov. Anyway in my opinion Anand is a much better player than Joel at any given time control and should score better than against Rybka

http://rybkaforum.net/cgi-bin/rybkaforu ... l?tid=1715

This is a pure speculation about Anand,the guy was sweaped away by the old program Rebel a long time ago

No wonder why Rebel was playing on a real monster machine the equivalent of an Intel Pentium II at 550 MHz

PS: The real factor of the slightly lost match by Anand was that Anand at that time did NOT had enough experience playing against computer chess programs, Plus jeroem cooked some openings that he knew in advadvance how Anand would respond to it up to move 20 or so.

http://www.rebel.nl/chespres.htm

Anand needs experience again(until he becomes a grandpa and around that time he would need need even more experience to play like young Anand who once was defeated by Rebel )
Anand doesn't have enough experience playing computers
because Rybka team has a monster machine and strength far better than Rebel and its machine long time back and much improved anti anand book from experienced Jereon again

You lost the argument,Give up Jorge.

Regards.

Why be disrespectful to someone based on a hypothetical that will likely never happen? Have you even considered that play conditions may be a factor in the games of engines playing GM's? This topic can be argued until the end of time with no one person being right. Is it fair that the book builder has access to a large list of the human players games and the human not have access to the engines opening book? Maybe, maybe not. It depends how you look at it. It could be argued that the engine also has publicly available games as it could also be argued that the opening prep is the most important part in a big game and will not be the same due to the situation. No one has lost anything here or has to give up.