Komodo Dragon is Kicking Ass in TCEC

[Chessqueen]

That would be an Option to let all other engines use their own Opening Books except LCZero, Stockfish, Allies, or Stoofvlees (No Opening Books) What the result would be, better or worse for those programs without Opening Books ?

Sedat's book contests can be used to measure the value of a book.

In this contest it was about 15 Elo:
https://sites.google.com/site/computers ... book-cs-47

Note also that a lot of the books were worse than no book.

I do NOT know how exactly they conducted the test of 300 Blitz games at 3+2", but if somebody can get Stockfish NNUE, Komodo Dragon, LcZero, without Opening Books to compete against, Stoofvlees, Alliestein and Ethereal using their own Opening Book, NOW that would be more interesting to see than using Opening Books that give the advantage to White as in TCEC. What would the Outcome be?

There are so many moving targets in that test, I would have no way to interpret the outcome.
In Sedat's test, all the engines were identical, the only thing that changed is the book.
Hence, you can directly infer the effect of books one against another.

So if I understand correctly what they did in Sedat's test is for instance Used Komodo Dragon with Opening versus Komodo Dragon without opening Books, and Stockfish NNUE with Opening Books versus Stockfish NNUE without Opening Book and let them play Hundreds or several hundreds of games and came out that the difference was only 15 Elo points more in favor of using Opening Books

[Dann Corbit]

No.
The engine was the Cfish port of the nnue version of SF.
The very same engine is used for every single game as both white and black.
The only difference between the opponents is that they are issued different books.
In this way, we can measure the strength of the book because it is the only variable that moves.
Everything else is identical.

[Chessqueen]

No.
The engine was the Cfish port of the nnue version of SF.
The very same engine is used for every single game as both white and black.
The only difference between the opponents is that they are issued different books.
In this way, we can measure the strength of the book because it is the only variable that moves.
Everything else is identical.

Thanks for explaining Dann, I thought that the test was to see if for instance as you mentioned let say using the same engine Cfish for both side one did NOT use Opening Book and the other used one or different top opening books to see if the difference is larger than 30 Elo points.

[Dann Corbit]

No.
The engine was the Cfish port of the nnue version of SF.
The very same engine is used for every single game as both white and black.
The only difference between the opponents is that they are issued different books.
In this way, we can measure the strength of the book because it is the only variable that moves.
Everything else is identical.

Thanks for explaining Dann, I thought that the test was to see if for instance as you mentioned let say using the same engine Cfish for both side one did NOT use Opening Book and the other used one or different top opening books to see if the difference is larger than 30 Elo points.

Actually that is partially correct because one of the configurations was Cfish with no book.

[Chessqueen]

No.
The engine was the Cfish port of the nnue version of SF.
The very same engine is used for every single game as both white and black.
The only difference between the opponents is that they are issued different books.
In this way, we can measure the strength of the book because it is the only variable that moves.
Everything else is identical.

Thanks for explaining Dann, I thought that the test was to see if for instance as you mentioned let say using the same engine Cfish for both side one did NOT use Opening Book and the other used one or different top opening books to see if the difference is larger than 30 Elo points.

Actually that is partially correct because one of the configurations was Cfish with no book.

It would be nice if the TCEC team after the 224 games allow all programmers to include their own Chess Openings and add another 112 games to determine the TCEC S20 Winner of this year, making the total number of games equal to 336 (224 + 112) = 336 . We can probably request that and if there are lot of register TCEC members requesting, they can easily add those 112 games

[Cornfed]

Oh please, no...don't make it a 'war of differing opening books'.

[mwyoung]

Oh please, no...don't make it a 'war of differing opening books'.

I agree, but computer chess started with opening books as a major part to the engine package.

And TCEC solution is clearly imperfect. And will always in its current state favor the more tactical engines.

When I started testing in computer chess around 1980. You could not separate the opening book, from the core engine.

[Chessqueen]

Oh please, no...don't make it a 'war of differing opening books'.

I agree, but computer chess started with opening books as a major part to the engine package.

And TCEC solution is clearly imperfect. And will always in its current state favor the more tactical engines.

When I started testing in computer chess around 1980. You could not separate the opening book, from the core engine.

Your are correct and that is the way it should be a complete package, even thought Stockfish NNUE, LCZero, Komodo Dragon and Maybe Stoofvless will still be in the same position as there are ahead of the other 4 engines

[Cornfed]

Oh please, no...don't make it a 'war of differing opening books'.

I agree, but computer chess started with opening books as a major part to the engine package.

And TCEC solution is clearly imperfect. And will always in its current state favor the more tactical engines.

When I started testing in computer chess around 1980. You could not separate the opening book, from the core engine.

Your are correct and that is the way it should be a complete package, even thought Stockfish NNUE, LCZero, Komodo Dragon and Maybe Stoofvless will still be in the same position as there are ahead of the other 4 engines

TCEC's choices may not be 'perfect'...but it's better than a 'wild west' solution of hired opening books. No, if the 'theoretical equality' of unbalanced openings/defenses played by each engine from each side leans 'too far' toward certain engines...simply pressure TCEC with reasoned argument to include more balanced positions to start from or even more 'suspected' options for White (to allow the possibility for more Black wins).

But God no...not different opening books. Only each engine supplying their own hardware would be worse.

[mwyoung]

Oh please, no...don't make it a 'war of differing opening books'.

I agree, but computer chess started with opening books as a major part to the engine package.

And TCEC solution is clearly imperfect. And will always in its current state favor the more tactical engines.

When I started testing in computer chess around 1980. You could not separate the opening book, from the core engine.

Your are correct and that is the way it should be a complete package, even thought Stockfish NNUE, LCZero, Komodo Dragon and Maybe Stoofvless will still be in the same position as there are ahead of the other 4 engines

TCEC's choices may not be 'perfect'...but it's better than a 'wild west' solution of hired opening books. No, if the 'theoretical equality' of unbalanced openings/defenses played by each engine from each side leans 'too far' toward certain engines...simply pressure TCEC with reasoned argument to include more balanced positions to start from or even more 'suspected' options for White (to allow the possibility for more Black wins).

But God no...not different opening books. Only each engine supplying their own hardware would be worse.

What are you talking about! The result with TCEC is pure wild west opening books. Chess openings is not played in a vacuum. No interest or care is taking with TCEC. As stated by TCEC itself. TCEC only concern is making artificial wins. For the purpose of views. Who is better then the programmer themselves to make the book. And have the best interest of their labor of love.