Crafty 25.x ...?

[Matthias Gemuh]

...

That's my "CURRENT" problem ...
...

[BBauer]

Frank Quisinsky wrote:
...
That's my "CURRENT" problem ...
...

Yeah, this guy is heavy on wire and he wants his tournament equal to go lose.
Kind regards
Bernhard

[Dann Corbit]

Heck, I'm happy with the 100M nodes per second on this 20 core box as is.

Half way to Deeper Blue, IIRC.

[Frank Quisinsky]

was clear ...

[mhull]

Heck, I'm happy with the 100M nodes per second on this 20 core box as is.

Half way to Deeper Blue, IIRC.

IIRC, 100M was about the average for DB2. The peak was 200M nps.

[bob]

Heck, I'm happy with the 100M nodes per second on this 20 core box as is.

Half way to Deeper Blue, IIRC.

Actually Hsu claimed they averaged about 100M nodes per second. But I'll take MY 100M any day, since I don't have the hardware limit they had (no hashing in last N plies of search, not to mention a very brute-force approach compared to today.)

However they would STILL be strong as all hell today, in that they would not make ANY 12 ply or so mistakes, usually quite a bit deeper with their singular extension stuff. But the Micros have now clearly passed the DB hardware although it only took 20 years.

[jdart]

Older systems are more common outside of the U.S.

--Jon

[Vinvin]

Heck, I'm happy with the 100M nodes per second on this 20 core box as is.

Half way to Deeper Blue, IIRC.

IIRC, 100M was about the average for DB2. The peak was 200M nps.

I remember I read something like 1 billion node/sec but Hsu prefers to say 200Mn/sec. He would probably mean that 80% of nodes are uninteresting ...

[bob]

Heck, I'm happy with the 100M nodes per second on this 20 core box as is.

Half way to Deeper Blue, IIRC.

IIRC, 100M was about the average for DB2. The peak was 200M nps.

I remember I read something like 1 billion node/sec but Hsu prefers to say 200Mn/sec. He would probably mean that 80% of nodes are uninteresting ...

No, the math was 480 chess processors, each searching 2M nodes per second, a max of right at one billion nodes per second. Hsu and Campbell claimed an average 10% duty cycle for the hardware, that is it averaged 1/10th of that theoretical max through a game, hence the 100M he mentioned in his book...

In terms of chess, 99.999% of the nodes are "uninteresting". But you can't tell that until after you have searched 'em.

[Jim Ablett]

Hi,

Is there anywhere to get Crafty executables (Win32 and Win64)? There used to be a page at http://www.kikrtech.com/ but that has changed completely.

Really looking forward to Crafty 25!

Many thanks,
Graham

Latest Crafty JA builds will always be found here >

http://crafty0981.moonfruit.com/

Jim.