GM Miguel Illescas vs Deep Junior 11

[pedrox]

On Thursday, June 5, 17.30 pm in Barcelona, Grandmaster Chess and current champion of Spain, Miguel Illescas, he will face in the form of rapid chess to software program Deep Junior, five times World Champion in the form of computers.

With the enormous strength of practice game of existing programmes, it would be a losing battle, if not because Miguel will be supported by a programme of basic chess, installed on a laptop running. Thus form a hybrid man-machine to grapple with the Deep Junior program, which will run on a powerful platform multiprocessor.

Rhythm game 20 minutes 10 seconds per move, 2 games

------------------------------------------------------------------

In my view I do not like as humans playing with advantages over when they do so with a human contricante. That is not liked by example that Kramnik could access the openings book of Fritz or Illescas now has a basic programme of tactic.

1. And that program is ?
2. Who will win?

Pedro

[Tony Thomas]

Advice Miguel to use Rybka.. Rybka on a lap top is a very decent opponent (Rybka might not win, but it wont lose that badly) for any Junior running on a multi-processor computer.

[Matthias Gemuh]

1. And that program is ?
2. Who will win?

Pedro

1. Danasah
2. Junior


Matthias.

[M ANSARI]

I like to think of a GM playing chess as someone who does not have good vision. When he plays a chess engine, his disadvantage is that he cannot see tactically many things ... thus by giving him glasses (which is help of a rudimentary engine which allows him to avoid simple tactical oversights) his vision is greatly improved and thus he can play very strong chess. Today ... for a GM to play a strong Rybka like engine unassisted is truly a challenging feat. Having a simple tactical engine evens the chances. I would guess the GM here will have a good result.

[Ovyron]

I think that Junior 11, even if being much stronger than Junior 10, would be no match for Rybka. So if the human uses Rybka in the match (trying to improve on Rybka moves, not using Rybka to improve his moves) he could win.

[pedrox]

1. And that program is ?
2. Who will win?

Pedro

1. Danasah
2. Junior


Matthias.

Hi Matthias, if the program basic was danasah or biglion, then there would be no need for a GM

Pedro

[peter]

Junior 11?
R.P.

[Terry McCracken]

I like to think of a GM playing chess as someone who does not have good vision. When he plays a chess engine, his disadvantage is that he cannot see tactically many things ... thus by giving him glasses (which is help of a rudimentary engine which allows him to avoid simple tactical oversights) his vision is greatly improved and thus he can play very strong chess. Today ... for a GM to play a strong Rybka like engine unassisted is truly a challenging feat. Having a simple tactical engine evens the chances. I would guess the GM here will have a good result.

No, GM's have excellent vision, but not perfect vision. But the best players on their best days can and do make computers look like the dumb machines they are.

Even lowly experts have scored points against engines, albeit not too often.

There are also techniques that make drawing computers not too difficult and closed positions are the computers worse nightmares, if they had any.

Machines are almost all speed and tactics and that is often enough for them to win.

[M ANSARI]

Ofcourse GM's have good tactical vision ... but really against a good engine on good hardware they must be tactically 400 or 500 ELO points weaker. Some GM's have the ability of playing computers much better than their rating. I was looking at some of the games in a recent tourney and got a chance to analyze a few games won by GM Mecking from Brazil. He seems to have a style that would be very successful against computers.

[Uri]

As was already proved from the Roman Dzindzichashvili vs. Rybka match, chess programs still have weaknesses that can be exploited and can still be beaten by a skilled human GM. Chess programs' weaknesses are especially in closed positions where there are not many tactical chances, in understanding compensation and in strategical endgames.