CSTAL has greater style than what has been since?

[S.Taylor]

http://64.68.157.89/forum/viewtopic.php?t=21299&start=0

Swami and others are suddenly praising CSTall, and even perhaps trying to persuade Chris to make a brilliant upgrade.
But I had thought, at any rate, that the latest one had been far overtaken by other programs, not only in ingenious attacking ability, but in it all being absolutely near perfect too, and standing the test of other engines and humans, which play technically near perfect (2700 elo and over).
So what can Cstal give us today, what we don't yet have?
OK, the last one you might admit is very weak for todays standards, but CAN Chris make an upgrade which competes with todays tops, AND has that great speculative style, far above all others? (and which might produce some immortal games, far more immortal and brilliant than earlier "immortal" games?).

Or what is it meant to be like? Is it meant to be able to make unexpected, stupid-looking sacrifices, and then unleash a great attack, due to the situation created by the sacrifice? Like making up a new Marshal attack or two, in every game?
OK, that's good too! But it must be real great stuff, almost irrefutable.

[Tony Thomas]

I do not have Cstal II but I doubt that it offers anything that other programs doesnt offer. It is human nature to praise things that are not around anymore, Toyota Supra (Car) comes to mind. People have a fascination towards how things would have ended up if they continued to become developed (Deep Blue for example). I am pretty sure you have seen what's stronger Deep Blue or Engine X threads.

[fern]

You are right. Cstal is by now a dinosaur. In my case I would like to see it again, in a new incarnation, because:
a) every new program excites me
b) new incarnation of old stuff -as Colossus- excites me even more
c) strength qualities cannot care me less. I am already superseded by Chessmaster 2100.
You know?
What I liked more of DOS Cstal was the face of the monster, making grimaces according the march of the game.

Yes, I am a silly boy regards
Fernando

[Dr.Wael Deeb]

You are right. Cstal is by now a dinosaur. In my case I would like to see it again, in a new incarnation, because:
a) every new program excites me
b) new incarnation of old stuff -as Colossus- excites me even more
c) strength qualities cannot care me less. I am already superseded by Chessmaster 2100.
You know?
What I liked more of DOS Cstal was the face of the monster, making grimaces according the march of the game.

Yes, I am a silly boy regards
Fernando

You are a great guy Fernando in my opinion

[swami]

Elo is not everything to me, Money is not everything to me.

Honestly, I think in human vs engine matches,

watching CSTal building up its own pieces in an innovative way, making wild sacrifices or advancing all the pieces one step more past the center ranks to attack your kingside in earlier stage of the game.

...is much better compared to, say,


typical chess engine playing typical computer style looking for combinations or positional advantages,
or accumulating small small advantages till the end to defeat human players.


If you are a 2600 player already, then you could look for ChessThinker or other such engines. Most of us are below 2400, So CSTal in Quad is still good for us.

Bottom line : High Entertainment value.

[Tony Thomas]

I think all this talk about CStal might have something to do with Swami, Thorsten and Chris running for election than anything.

Last minute propaganda regards
Prince Cinammon Booots

[S.Taylor]

Elo is not everything to me, Money is not everything to me.

Honestly, I think in human vs engine matches,

watching CSTal building up its own pieces in an innovative way, making wild sacrifices or advancing all the pieces one step more past the center ranks to attack your kingside in earlier stage of the game.

...is much better compared to, say,


typical chess engine playing typical computer style looking for combinations or positional advantages,
or accumulating small small advantages till the end to defeat human players.


If you are a 2600 player already, then you could look for ChessThinker or other such engines. Most of us are below 2400, So CSTal in Quad is still good for us.

Bottom line : High Entertainment value.

What _I_ would consider excellent, and well worth sponsoring (if Fernando would like to) is a program which can make unlikely looking sacrifices and through them to prove it by building up unbelievable attacks through instability that sacrtifice caused. Ir should be sacrifices oif all dimensions and proportions, and so stupid that a normal person would never think much into it, BUT a mechanized Tal should be thinking of it the whole time, and be ready to pounce, and build up for it all the rest of thet time.
BUT, IF any program could do this, then I don't think it could be less than about 3000 elo. It needs to have SO MUCH KNOWLEDGE AND IDEAS, so that it can know what to conjur up when where and how.
I don't mind if it can be beaten easily by Rybka, but besides that, it should be like the real Tal in his prime and more! Maybe it can do stupider things than Tal himself would have done, BUT they must all have very intelligent follow-ups, even if they fail to Rybka, but to little else.
Such a program would be a massive massive inspiration for humans!

[Dr.Wael Deeb]

Elo is not everything to me, Money is not everything to me.

Honestly, I think in human vs engine matches,

watching CSTal building up its own pieces in an innovative way, making wild sacrifices or advancing all the pieces one step more past the center ranks to attack your kingside in earlier stage of the game.

...is much better compared to, say,


typical chess engine playing typical computer style looking for combinations or positional advantages,
or accumulating small small advantages till the end to defeat human players.


If you are a 2600 player already, then you could look for ChessThinker or other such engines. Most of us are below 2400, So CSTal in Quad is still good for us.

Bottom line : High Entertainment value.

What _I_ would consider excellent, and well worth sponsoring (if Fernando would like to) is a program which can make unlikely looking sacrifices and through them to prove it by building up unbelievable attacks through instability that sacrtifice caused. Ir should be sacrifices oif all dimensions and proportions, and so stupid that a normal person would never think much into it, BUT a mechanized Tal should be thinking of it the whole time, and be ready to pounce, and build up for it all the rest of thet time.
BUT, IF any program could do this, then I don't think it could be less than about 3000 elo. It needs to have SO MUCH KNOWLEDGE AND IDEAS, so that it can know what to conjur up when where and how.
I don't mind if it can be beaten easily by Rybka, but besides that, it should be like the real Tal in his prime and more! Maybe it can do stupider things than Tal himself would have done, BUT they must all have very intelligent follow-ups, even if they fail to Rybka, but to little else.
Such a program would be a massive massive inspiration for humans!

Well said Shimon....
I remember now from the tons of materials I read about world chess champions that not all Tal's combinations and sacrifices were sound,at least after a deep study,but they were an excellent surprise weapon over the board....

[Eraserheads]

"When (then-world-champion Boris) Spassky offers you a piece, you might as well resign then and there. But when Tal offers you a piece, you would do well to keep playing, because then he might offer you another, and then another, and then ... who knows?"

SPASSKY = could be akin to Rybka, Fritz, Shredder, or any other 'typical' chess program.

TAL = CSTAL or the other 'intuitive' programs

[swami]

Tony,

Konjam adangu

You will get elected.