Rybka 3 VS Top Players in the World

Wouldn't it be awesome to see a match between Rybka 3 in one corner and a team of the top 5 or 10 rated players in the world in the other? 

Who do you think would win?

What are the chances of seeing something like this actually happen?

If Rybka 3 was not handicapped at all, playing on the hardware that the Rybka 3 team can afford, given ample time to analyze the positions, I can't see Rybka 3 losing.

the silicone monster would eat everyone

Well, I think to be fair, you'd have to play it at standard otb time controls, not 'until rybka is happy with its move' time controls.  I can't see it losing at the moment.  Who would beat it?  I cant see carlsen beating it with the dragon, or anand with his sharp play.  If I had to pick someone to beat it it would probably be kramnik, but not on current form.  Maybe Ivanchuk would be in with a shout. 

aren't there people that just dominate computers though?

i saw nakamura beat some computer in a blitz game

it was like 250 moves or something

I think a team of grandmasters, given enough time to prepare for the match, and off days in between games, would win against an unhandicapped Rybka.

There's a huge, huge difference between dominating computers and dominating Rybka 3 on optimal hardware with no handicaps. I believe that game was against crafty, who is rated in the 2600's compared to Rybka 3, rated in the 3200's.

Blitz is different.  You can employ external strategies to beat computers in blitz, like blocking up the position and playing fast, safe moves, where a computer will have to spend more time thinking since it has no intuition, and doesnt necessarily know that most moves will be ok.  At regular time controls, you just have to play good moves. 

Unfortunately blitz is not applicable in this scenario at all, since how can you get 5 - 10 people to agree on a move in the time restraints. And at regular time controls, you can't play good moves, you have to play perfect moves to draw. 

No you don't have to play perfect moves at all, just non-losing moves to draw, and good moves to draw with winning chances.  With perfect moves, with enough games in the match you would win against rybka 3 every time. 

It is pretty easy to crush silicone machines...

On a more serious note, I think most grandmasters would beat Rybka 3 in a game of two weeks per move. 

lol silicone machines.  For some reason I didnt see that before. 

Hmm, not too much confidence left among us humans.  I am not sure what time controls would be ideal in order to reach consensus among 5-10 people but I think they'd have a shot at getting some wins.  Especially if it's a long match I think they could discover and exploit weaknesses the program might have.  I'd like to see something like this.

Unfortunately understanding how the computer works isn't enough. It analyzes the position and calculates the best followup line to a certain depth. Now I know how it plays, to crush it you're going to have to analyze to a larger depth, are you saying that humans can easily see every line 20ply deep?

Keeping in mind that Rybka 3 still makes mistakes ... and that engines are best at exploiting tactical errors ... I'd figure a team of SuperGM's are unlikely to make a mistake that Rybka is capable of exploiting ... yet a team of humans is bound to be able to set up a nice endgame position and at least draw time after time ...

My opinion... if it was just a few matches, a draw ... if it was a 'play until someone wins' type tournament ... the GM's.

Computers analyze a position in a specific way. The only reason they can search 20 ply ahead is because they cut off moves that worsen the position for example. Computers will see that losing a queen is a bad line and might only look another 3-4 moves after the losing of a queen. Where as humans might see a forced mate in 9 after a queen sacrifice.

Find me a position with a forced mate in 9 that Rybka 3 won't find. 

um good luck trying to make on of those