Parity with rybka

[seemychess]

Their seems to be about 5 engines reaching parity with rybka. is rybka no longer all alone at the top?

[Erik Roggenburg]

Their seems to be about 5 engines reaching parity with rybka. is rybka no longer all alone at the top?

Which engines are those?

[Alessandro Scotti]

I see Rybka still comfortably on top of the rating lists but indeed some other engines are getting closer.

One funny thing is just a few years ago I remember a discussion over strong engines improving only 30 elo or so in a year because it was "exceptionally difficult" to improve an engine at the top. Then came Fruit and then came Rybka and now we're seeing huge improvements again.

This has to mean something though I'm not really sure what...

[Sylwy]

I see Rybka still comfortably on top of the rating lists but indeed some other engines are getting closer.

One funny thing is just a few years ago I remember a discussion over strong engines improving only 30 elo or so in a year because it was "exceptionally difficult" to improve an engine at the top. Then came Fruit and then came Rybka and now we're seeing huge improvements again.

This has to mean something though I'm not really sure what...

Clones again ?????????

[swami]

I see Rybka still comfortably on top of the rating lists but indeed some other engines are getting closer.

One funny thing is just a few years ago I remember a discussion over strong engines improving only 30 elo or so in a year because it was "exceptionally difficult" to improve an engine at the top. Then came Fruit and then came Rybka and now we're seeing huge improvements again.

This has to mean something though I'm not really sure what...

My thoughts exactly. Now,I'm having a vulcan mind meld with you,Alessandro.

[swami]

I see Rybka still comfortably on top of the rating lists but indeed some other engines are getting closer.

One funny thing is just a few years ago I remember a discussion over strong engines improving only 30 elo or so in a year because it was "exceptionally difficult" to improve an engine at the top. Then came Fruit and then came Rybka and now we're seeing huge improvements again.

This has to mean something though I'm not really sure what...

Clones again ?????????

Not exactly, I would guess that commercial authors like Frans for example, learned something from much improved open source engines like Toga,Glaurung etc

They surely know the way to use ideas instead of copy pasting because they are expert themselves.

[Ovyron]

This has to mean something though I'm not really sure what...

New techniques were discovered regarding computer chess, and these techniques destroy the old approaches.

Also, the hardware got faster, so it means newer possibilities not available before. For example, the latest ProDeo is the strongest on actual hardware, because it uses some new tricks, but you can test this ProDeo on ancient hardware (Like at 1Ghz) and this ProDeo will perform worse than other ProDeos. That means that at older hardware no more strength could have been added, it's the new hardware that opens such doors to new approaches.

[Uri Blass]

I see Rybka still comfortably on top of the rating lists but indeed some other engines are getting closer.

One funny thing is just a few years ago I remember a discussion over strong engines improving only 30 elo or so in a year because it was "exceptionally difficult" to improve an engine at the top. Then came Fruit and then came Rybka and now we're seeing huge improvements again.

This has to mean something though I'm not really sure what...

Clones again ?????????

No need for clones.
Good programmers can learn from stronger programs.

A simple example:
if you see f(0)=0 f(1)=1 f(2)=4 f(3)=9 f(4)=16 then you can guess f(x)

chess program is simply a function that gives output for every input.
No need for source code of rybka in order to learn from rybka and if you see output for many positions then you can learn from it and improve your program by adding knowledge that rybka has and your program does not have.

Uri

[Karmazen & Oliver]

I see Rybka still comfortably on top of the rating lists but indeed some other engines are getting closer.

One funny thing is just a few years ago I remember a discussion over strong engines improving only 30 elo or so in a year because it was "exceptionally difficult" to improve an engine at the top. Then came Fruit and then came Rybka and now we're seeing huge improvements again.

This has to mean something though I'm not really sure what...

Clones again ?????????

No need for clones.
Good programmers can learn from stronger programs.

A simple example:
if you see f(0)=0 f(1)=1 f(2)=4 f(3)=9 f(4)=16 then you can guess f(x)

chess program is simply a function that gives output for every input.
No need for source code of rybka in order to learn from rybka and if you see output for many positions then you can learn from it and improve your program by adding knowledge that rybka has and your program does not have.

Uri

YES ¡ and the knowledge have not © ...

[Ovyron]

I think that the secret of Rybka was already guessed, and that if Vas stops updating Rybka, she could be surpassed soon.