Stockfish 1.5 : 2821

[Dann Corbit]

My first matches were not so stunning at all (single PC and blitz):

SF1.5 - Rybka3 3,5 - 16,5
SF1.5 - Naum4 5,5 - 14,5
SF1.5 - SF1.4 9 - 11

In test suites new Stockfish is slightly faster than old. BTW I wonder how Marco can improve engine without compiler (if I understand right)?

Jouni

"Beware of bugs in the above code; I have only proved it correct, not tried it." -- Donald Knuth

[George Tsavdaris]

For the 1st rating SF 1.5 played against weak engines !

It doesn't really matter. In fact it may be(because of the drawish nature of practical Chess) harder against weak engines to perform so good to keep a high enough rating.

[mcostalba]

In test suites new Stockfish is slightly faster than old. BTW I wonder how Marco can improve engine without compiler (if I understand right)?

I have the compiler....but I am not able to use it

I have Windows Vista 32 bit with MSVC and Linux 32 bit with gcc and icc (Intel compiler).

I do my compiles for my tests and for developing of course, but Jim does much better pgo compile then me, I am still not able to find the secret recipe to make the pgo really fast, is a kind of black art and Jim is the master here.

So the public final release is up to Jim becasue he is able to build a faster binary then I am.

[Michael Sherwin]

My first matches were not so stunning at all (single PC and blitz):

SF1.5 - Rybka3 3,5 - 16,5
SF1.5 - Naum4 5,5 - 14,5
SF1.5 - SF1.4 9 - 11

In test suites new Stockfish is slightly faster than old. BTW I wonder how Marco can improve engine without compiler (if I understand right)?

Jouni

When the Spracklen's wrote Sargon they did not even own a computer!