GM says Rybka & Fritz weaker than best GMs in classical

[pedrox]

If GM are so superior strategy to top engines, then I will not be because GM Miguel Illescas + Junior 9 not win Deep Junior 11 and because no GM + computer wins tournaments Free of PlayChess.

[Terry McCracken]

If GM are so superior strategy to top engines, then I will not be because GM Miguel Illescas + Junior 9 not win Deep Junior 11 and because no GM + computer wins tournaments Free of PlayChess.

Uh??

[Uri Blass]

A GM can _way_ out-calculate a computer along sharp tactical lines for the most part,

That is not true for today's hardware and top engines !!!!!.

Matthias.

No, I'm afraid it is true, Bob is right.

There are many many such examples out of human play that prove this, Bob explained accurately why the GM's often fail and why.

No

I think that he is not right.
Today
The GM fail even to get a positional advantage against computers from the opening poosition and it means that A GM cannot even out-calculate a computer along positional lines most of the time.

The computer may play the good positional moves thanks to search and not thanks to superior evaluation but this does not change the fact that
rybka plays better positional moves relative to humans.

There are of course positions that rybka does not play well but humans usually cannot force them when rybka is using a book.

Uri

No Uri, you're wrong. Maybe you need to be a GM to realize this or have decades of experience dealing with GM's and computers.

Uri is a correspondence grandmaster,he covered his latest norm early this year....
The funny thing is that I was going to post a thread today asking where is Uri as he didn't post for a long time and bang,there he is

I have one GM norm and ICCF rating of 2611 but
I am not a correspondence GM.

My performance is not worse than GM's but I did not try to get the GM title.

One condition to get the GM title is to play in tournaments and I played in almost no tournaments

I have exactly 30 games in the ICCF list and only in one tournament I could get the GM norm and got it(the first 2 tournaments that I played for rating were 2 weak to get the GM norm) .

Based on my experience in performing at GM level I can say that I used the computer for hours in the root position and I also used it for hours in different positions from the root position.

I did not use the computer only to blunder check analysis and I think that today people who use the computer only in that way do not get high rating.

Uri

[Uri]

I think that what differentiates strong players from weak players is mainly memory skills. GMs have the ability to recall from memory board positions which they calculated or experienced before, something which weak players find quite difficult.

[bob]

A GM can _way_ out-calculate a computer along sharp tactical lines for the most part,

That is not true for today's hardware and top engines !!!!!.

Matthias.

Easy to say, but programs have a fixed and fairly shallow horizon. GMs sometimes calculate variations to 40-50 plies. Programs don't.

[Matthias Gemuh]

A GM can _way_ out-calculate a computer along sharp tactical lines for the most part,

That is not true for today's hardware and top engines !!!!!.

Matthias.

Easy to say, but programs have a fixed and fairly shallow horizon. GMs sometimes calculate variations to 40-50 plies. Programs don't.

If you mean in correspondence chess, aided by engines, you are right.
Otherwise there is no proof for your claim.

Statistics prove that you are wrong:
if the GMs were that strong, they would not be blundering as frequently as they do in important human-human tournaments.
Just pick _any_ one of such tournaments and count the blunders at _shallow_ depth. There is no way they can calculate variations to 40-50 plies without using engines.

Matthias.

[Uri Blass]

A GM can _way_ out-calculate a computer along sharp tactical lines for the most part,

That is not true for today's hardware and top engines !!!!!.

Matthias.

Easy to say, but programs have a fixed and fairly shallow horizon. GMs sometimes calculate variations to 40-50 plies. Programs don't.

If you mean in correspondence chess, aided by engines, you are right.
Otherwise there is no proof for your claim.

Statistics prove that you are wrong:
if the GMs were that strong, they would not be blundering as frequently as they do in important human-human tournaments.
Just pick _any_ one of such tournaments and count the blunders at _shallow_ depth. There is no way they can calculate variations to 40-50 plies without using engines.

Matthias.

Note also that if we talk about the nolot positions that were mentioned earlier in this thread then I have 2 comments

1)These positions are exception and GM usually do not find moves that are deep for computers to find.

2)The GM's did not calculate 40-50 plies forward and they practically guessed when they used selective search without seeing all lines.

There are clearly cases when their guess was wrong.

3)Today top programs solve part of the nolot positions quickly and I do not think they are easy for GM's assuming that the GM's did not see the positions earlier.

Uri

[Terry McCracken]

A GM can _way_ out-calculate a computer along sharp tactical lines for the most part,

That is not true for today's hardware and top engines !!!!!.

Matthias.

Easy to say, but programs have a fixed and fairly shallow horizon. GMs sometimes calculate variations to 40-50 plies. Programs don't.

If you mean in correspondence chess, aided by engines, you are right.
Otherwise there is no proof for your claim.

Statistics prove that you are wrong:
if the GMs were that strong, they would not be blundering as frequently as they do in important human-human tournaments.
Just pick _any_ one of such tournaments and count the blunders at _shallow_ depth. There is no way they can calculate variations to 40-50 plies without using engines.

Matthias.

IGM Alexander Kotov claimed in one position he looked 50 moves ahead, 100 plies!

You underestimate GM's, you underestimate them greatly.

[Matthias Gemuh]

A GM can _way_ out-calculate a computer along sharp tactical lines for the most part,

That is not true for today's hardware and top engines !!!!!.

Matthias.

Easy to say, but programs have a fixed and fairly shallow horizon. GMs sometimes calculate variations to 40-50 plies. Programs don't.

If you mean in correspondence chess, aided by engines, you are right.
Otherwise there is no proof for your claim.

Statistics prove that you are wrong:
if the GMs were that strong, they would not be blundering as frequently as they do in important human-human tournaments.
Just pick _any_ one of such tournaments and count the blunders at _shallow_ depth. There is no way they can calculate variations to 40-50 plies without using engines.

Matthias.

IGM Alexander Kotov claimed in one position he looked 50 moves ahead, 100 plies!

You underestimate GM's, you underestimate them greatly.

Anybody can look 8 plies ahead and claim to look 500.

[Uri Blass]

A GM can _way_ out-calculate a computer along sharp tactical lines for the most part,

That is not true for today's hardware and top engines !!!!!.

Matthias.

Easy to say, but programs have a fixed and fairly shallow horizon. GMs sometimes calculate variations to 40-50 plies. Programs don't.

If you mean in correspondence chess, aided by engines, you are right.
Otherwise there is no proof for your claim.

Statistics prove that you are wrong:
if the GMs were that strong, they would not be blundering as frequently as they do in important human-human tournaments.
Just pick _any_ one of such tournaments and count the blunders at _shallow_ depth. There is no way they can calculate variations to 40-50 plies without using engines.

Matthias.

IGM Alexander Kotov claimed in one position he looked 50 moves ahead, 100 plies!

You underestimate GM's, you underestimate them greatly.

Anybody can look 8 plies ahead and claim to look 500.

I can add that selective search to 500 plies often can miss important things.
A GM may look 20 plies forward and miss mate in 1 as happened to kramnik against Fritz(I do not know if he calculated 20 plies forward in his selective search and it is only a guess).

Looking too deep is simply not effective and it is better to see to small depth but to see everything.

Uri