Father's monkey tricks are over....

[Dr.Wael Deeb]

http://rybkaforum.net/cgi-bin/rybkaforu ... ?tid=17547

As I wrote many moons before,a simple gambit opening book will burn him like a dry wood

[gerold]

http://rybkaforum.net/cgi-bin/rybkaforu ... ?tid=17547

As I wrote many moons before,a simple gambit opening book will burn him like a dry wood

Maybe.Time will tell I think he may fine a way to work around
that.

Best to you,
Gerold.

[Nimzovik]

Perhaps Herr Dr..... However would a simple gambit opening book work against a G.M.? This is an issue. If a GM has also Pablo's technique in his repertoire then would the machine fare as well? This is the philosophical question here. Pablo presents a certain 'pattern recognition' that the machine des not recognize(fortressess, blocakade etc.) either due to the 'horizon effect' or it's 'search' or it's 'contempt' factor. Thus Pablo's thesis of the "two sided coin" of which he often speaks is apparantly ....valid. Some prgrammers, if I recalll correctly, have indicated that some positional concepts are difficult to instll in a program. Blocked pawn chains is one of them. This is a good thing. For it means chess is not 'solved' by the silicon monsters yet and we still have a chance. I am not one of them however.

[Arturo Ochoa]

Perhaps Herr Dr..... However would a simple gambit opening book work against a G.M.? This is an issue.

No, it is not. Currently, the silicon is a superior tactician than a GM, who is a human being. In open positions, the silicon is superior. A gambit breaks the position, creates unbalanced and dynamic conditions that can’t be handled in 180 seconds, even if you bet in the time bug of the software.

[M ANSARI]

The person being talked about is not a GM, but rather a chess player that knows how engines behave and uses that fact along with a known GUI bug to win lost positions on time. While I have no problem with that and I think a win is a win, there are obviously very easy ways to circumvent that. It is just that engine authors are usually not interested in devising an engine just for that. I think a highly speculative book that forces an open position, or even an engine that has extremely high contempt would be successful as I doubt the player would be able to win a game against these engines on today's hardware even a knight down.

Having said that Pablo is not doing anything wrong and is just playing with the conditions that are out there. I personally don't think it interesting at all, but I certainly see nothing wrong with it. I have to admit there was a time I also used to play engines by closing positions and trying to use better positional understanding to grind a win (not a time win). Those times are long over and today's engines are just too good that I have a hard time even against my Shredder iPhone.

[Look]

http://rybkaforum.net/cgi-bin/rybkaforu ... ?tid=17547

As I wrote many moons before,a simple gambit opening book will burn him like a dry wood

Hi Dr,

I dont think that managing to draw against these monsters is a monkey trick, even if it is now and then. Having said this I think producing engines with would beat humans more easily should be possible. A crude Idea is preventing pawn blockades; say by giving negative scores to blocked pawns. There are more ideas too that one can think of. But overall they should not reduce engines strength considerably.

[Nimzovik]

Perhaps Herr Dr..... However would a simple gambit opening book work against a G.M.? This is an issue.

No, it is not. Currently, the silicon is a superior tactician than a GM, who is a human being. In open positions, the silicon is superior. A gambit breaks the position, creates unbalanced and dynamic conditions that can’t be handled in 180 seconds, even if you bet in the time bug of the software.

I wonder if this is true? If it were absolutely true then all computers would play nothing but gambits in tournaments against humans. Any tweaking of a program books or other method causes a trade off. Also note that I observe that the time bug issue is not the central issue. The fact that the final positions are dead draws is a real factor in this discussion. The issue is the fact that such a lower rated player (whatever the win /loss percentage may be) can do what he did to the machine -in such an embarassing manner too I might add. I reiterate the fortress and closed positions that the current programs do not understand completely is the issue. Yes one can indeed tweak a program to avoid at all costs such positions. However u would then experience a weakness in the program in some other area. I say again. I wonder how a modern day Nimzovich would fare against these things? It would be interesting....

[Albert Silver]

http://rybkaforum.net/cgi-bin/rybkaforu ... ?tid=17547

As I wrote many moons before,a simple gambit opening book will burn him like a dry wood

AFAIK, all you need to do is trade off 2-3 pawns and say goodbye to blockades.

[michiguel]

The person being talked about is not a GM, but rather a chess player that knows how engines behave and uses that fact along with a known GUI bug to win lost positions on time. While I have no problem with that and I think a win is a win, there are obviously very easy ways to circumvent that. It is just that engine authors are usually not interested in devising an engine just for that.

Because nowadays, engine authors are not interested in Artificial Intelligence anymore. Crying that a human is using premove is really silly. Make the engine a premover!

Miguel

I think a highly speculative book that forces an open position, or even an engine that has extremely high contempt would be successful as I doubt the player would be able to win a game against these engines on today's hardware even a knight down.

Having said that Pablo is not doing anything wrong and is just playing with the conditions that are out there. I personally don't think it interesting at all, but I certainly see nothing wrong with it. I have to admit there was a time I also used to play engines by closing positions and trying to use better positional understanding to grind a win (not a time win). Those times are long over and today's engines are just too good that I have a hard time even against my Shredder iPhone.

[Dr.Wael Deeb]

http://rybkaforum.net/cgi-bin/rybkaforu ... ?tid=17547

As I wrote many moons before,a simple gambit opening book will burn him like a dry wood

AFAIK, all you need to do is trade off 2-3 pawns and say goodbye to blockades.

Exactly