The Houdini chess engine has an Elo grade of 3300. Has a Grandmaster ever beaten it in any game? If so which grandmaster and when?
It's not so difficult to beat Houdini in correspondence chess. Computers still lack certain elements of positional understanding, and they can certainly be outplayed by a strong player. On rapid/blitz games though, it is a totally different story.
I'd be kind of surprised if humans do better in correspondence chess against a computer than in a 'normal' timed (but not rapid/blitz) game, because it can spend every second calculating (unless restricted in some way) whereas a human has the matter of life to attend to.
I guess you're surprised then.
1. In longer time controls, humans can avoid most major tactical oversights.
2. Computers cannot acquire positional grasp, no matter how long the think. If they can pick between a few moves of about equal value according to their algorithms, there is a serious chance picking up an inferior one.
Currently in ICCF word championships the usage of computers is allowed. While there have been a few weak players who have made good results, the vast majority of champions and top rated players are/were also very strong OTB.
I'd be surprised as well. I don't believe anyone can outplay Houdini or any top rated engine. Sure, computers have no real positional understanding, but that doesn't mean that a positionally good player will overcome it.
This only means that a stronger OTB player can understand and use the information coming out of the engine in a much better way than any ol' n00b with an engine. This doesn't mean that a human can outplay an engine. Not even close.
I think that if the engine is given enough time, it will outplay any human ;-)
Michael Adams lost 5.5-0.5 to a computer quite recently. I know that Kramnik lost 2-0 to Fritz in 2006. Houdini is a lot stronger than Fritz was then - Houdini's grade of 3300 is 500 points more than the grade of even a super-grandmaster. To beat a player 500 points higher is very rare.
I wondered if anyone had any concrete evidence of a GM beating Houdini (or Stockfish or any of the other super-engines) recently in any game at any time control - maybe they could post the score.
When they'll make the chess enginge that makes best moves possible?
While it's rather apparent that Kramnik has absolutely no motivation playing a correspondence match against a machine (he would win the match hands down of course) you may have a look at a slightly older article:
The engines were positionally dumb, and still are.
Hmm, but does strategy matter much when you can see 64 moves ahead in every single line? It seems to me that strategy is essentially a bunch of thinking shortcuts that humans use because we're so intellectually limited.
It's funny how many of you downplay/misunderstand the importance of positional thinking.
Think of it this way, computers, although they calculate truely massive amounts of lines unthinkably quickly, they too don't begin to scratch the surface of the game tree complexity.
Computers have gained tremendous strength this past decade due to improved algorithms that are better at taking into account positional elements. It's a shortcut for us both.
Even at my meager rating, I can suggest moves to Houdini that it doesn't like at first, but 5-10 moves down the road it decides are better (I go back to that move and suddenly my move is preferred). Obviously this is not the norm or I'd be much stronger, but to see it happen with any frequency brings you face to face with their limitations.
Computers can't be beaten in classical time controls because their tactics are so damn consistent. (And of course they make tactical/positional mistakes too, otherwise they'd never beat eachother). Humans can't match this kind of consistent quality OTB.
Local expert who is ~2400 USCF correspondence talks about how easy it is to beat the idiots who take Houdini or Rybkas top few moves and only ever play them. Correspondence chess is a totally different matter.
The link thet IM Pfren gives is from 2005 -
Things have moved on even from 2005 - the engines are better the humans are about the same.
Does anyone have evidence (rather than just opinion) of a GM beating a top engine (Elo > 3000)?
To my knowledge no such match (correspondence type) has ever been organized. So if you're looking for a game then you wont find one.
If you mean classical time controls, then of course the human will lose.
No, it doesn't. At least not yet. If that applied then the ICCF world champion would be a woodpusher. He isn't, the current one had a respectable 2260 FIDE ELO before turning completely to correspondence chess, and the same applies for the guy who finished second.
Correspondence chess is not really typical chess, since as IM pfren pointed out, the humans can use computers to help them anyway so I don't see how you can tell if humans can beat engines in correspondence chess.
My original question concerned normal OTB chess with any of the normal time controls.
I agree with padman.
Quote from wikipedia (under Houdini) -
When GM Peter Svidler was asked which one player he would choose to represent Earth in a hypothetical match against aliens, he answered "Houdini"
GM Svidler seems to agree too!
Okay. Then, surely, being able to see 64 moves ahead in all lines with 100% accuracy, plus having some positional understanding, is superior to being able to see a handful of moves ahead in some lines with reasonable accuracy, and having some position understanding.
It seems to me that strategy is simply long-term tactics, and engines are wonderful at tactics, whether in the short term or the long term.
If that applied then the ICCF world champion would be a woodpusher.
I don't think the claim "engines are stronger than humans" entails that a beginner using an engine will play just as well as a GM using an engine.
It could be that engines have a few key weaknesses, and GMs are really good at filling in the gaps here, but at the same time engines are so strong that, even with these weaknesses, they can beat any human player.
For once, pfren is well off target.
Humans have no chance against good software/good hardware combinations, at any time control.
All this talk that engines don't understand positional play is simply out of date nonsense.
p.s. I'm willing to back up my assertion.... how about this: I play you (pfren) two games of centaur chess, in which we both use computers. I'm a patzer compared to you OTB but suspect my rig (hardware/software) is superior. Bet you don't even come close to winning a game. Not even close.
Id like to see a human player take the top ranking on chess.com.