what is left in computer chess programming

[Henk]

If the tuner is not optimal you won't get optimal results. Automatic tuning is normally implemented in software. Hardware is more difficult too change So you must be very sure that no better software will ever be found. But you even don't know if you have the minimal set of necessary parameters to be tuned.

[Bloodbane]

Perfection.

[Henk]

Perfection.

We should be happy with good results. Stockfish is good enough.

[mclane]

That is because you are talking about orthodox Chess, a game that has been beaten to death. You would do well to expand your horizon, because almost anything else is much more interesting. Steak is nice, but who still wants to eat a steak that has been chewed to the bone?

As Long as chess is Not solved it is not dead.

In the moment there are no new ideas. There will be new ideas
With new people coming into the field.

These new people think different. And program different things.

[Henk]

That is because you are talking about orthodox Chess, a game that has been beaten to death. You would do well to expand your horizon, because almost anything else is much more interesting. Steak is nice, but who still wants to eat a steak that has been chewed to the bone?

As Long as chess is Not solved it is not dead.

In the moment there are no new ideas. There will be new ideas
With new people coming into the field.

These new people think different. And program different things.

As long as humans are interested in chess, chess is not dead. Problem with computer chess is that humans are getting more and more superfluous. It started when humans were losing games from computers all the time. So perhaps a popular game where computers lose from humans would be a more interesting field for research.

[mclane]

computerchess is not about humans.
it is about the strongest engine.

when people watch soccer or marathon or jumping high
the people watching do not do that sport themselves. maybe some.
but the mass not.

the people buying houdini or hiarcs or stockfish and komodo do not need to beat these programs in chess.

progress happens when people with different approaches come into the field and do something different.

as long as anybody copies from each other i don't think progress will happen.

what was it the genius made strong ? he did something different.
frans morsch did something different.
ed schroeder did something different.
mark uniacke did it different.
chris whittington did it different.

they all had different ideas. that was the reason their programs were so different but on the other hand it was very interesting to see those different approaches fight against each other.
gandalf e.g. did king attacks.
it evaluated these attacks statically. it was capable to open pawn blocked center for a king attack. i have seen it myself.

virtual chess (i am talking about the dos version) was also capable to sac material for an attack.

in he copy/past days we do not see progress because they all use the same sources.

[Graham Banks]

......in he copy/past days we do not see progress because they all use the same sources.

Depends how you define progress. Ratings are improving relatively quickly compared to over five years ago.
That is due to not only refining existing ideas, but trying new ones I suspect.

Not all of the rapidly improving engines are derivatives.
However, I agree that the more original engines generally progress at a better rate.

[Henk]

computerchess is not about humans.
it is about the strongest engine.

when people watch soccer or marathon or jumping high
the people watching do not do that sport themselves. maybe some.
but the mass not.

the people buying houdini or hiarcs or stockfish and komodo do not need to beat these programs in chess.

progress happens when people with different approaches come into the field and do something different.

as long as anybody copies from each other i don't think progress will happen.

what was it the genius made strong ? he did something different.
frans morsch did something different.
ed schroeder did something different.
mark uniacke did it different.
chris whittington did it different.

they all had different ideas. that was the reason their programs were so different but on the other hand it was very interesting to see those different approaches fight against each other.
gandalf e.g. did king attacks.
it evaluated these attacks statically. it was capable to open pawn blocked center for a king attack. i have seen it myself.

virtual chess (i am talking about the dos version) was also capable to sac material for an attack.

in he copy/past days we do not see progress because they all use the same sources.

When there are no humans involved computer chess is totally useless. For instance computers playing against each other in a closed room and no results saved, totally useless. Next step is humans watching chess games they do not understand. Third step humans playing against computers but losing all the time. .......... Final step humans playing chess in real against humans that smoke, drink, make noises or other irritating gestures.

[mclane]

humans are involved as operators. programmers. kiebitz.

[Henk]

humans are involved as operators. programmers. kiebitz.

Computer chess needs to be improved for analyzing correspondence chess games or just for theoretical or scientific purposes.