Arasan 8.0

[Jon Dart]

Geschrieben von:/Posted by: Jon Dart at 22 August 2004 23:52:32:

Arasan 8.0 is available from http://www.arasanchess.org.
Changes in Arasan 8.0:
1) Extended futility pruning is implemented now. Futility pruning and
razoring can be combined.
2) Passed pawns are scored higher.
3) "Capture" extension is now disabled by default.
4) Fixed some bugs in king-pawn hash table code
5) Minor changes to hash probe code.
6) Default tablebases cache size increased, changes to tablebase probe
logic. Generally this version does fewer probes.
7) New egtb code from Eugene Nalimov & Andrew Kadatch.
8) Significant changes to evaluation. Evaluation is simpler and
faster. A small evaluation cache is used.
9) Various speed enhancements, in move generator and elsewhere.
10) Some fixes to 50-move draw handling.
11) "makebook" program can now filter moves so that only moves
occurring with a minimum frequency will be included. New book
uses minimum frequency of 6 from a large PGN game collection.
12) a few minor GUI fixes.
--Jon

[Roger Brown]

Geschrieben von:/Posted by: Roger Brown at 23 August 2004 04:46:13:
Als Antwort auf:/In reply to: Arasan 8.0 geschrieben von:/posted by: Jon Dart at 22 August 2004 23:52:32:

Arasan 8.0 is available from http://www.arasanchess.org.

Hello Jon,
Thanks.
You are no longer publishing updates of the large Arasan book?
I did not see it on your downloads section.
Should I delete that one in favour of the smaller book that currently ships with the executable or are they both used now?
Later.

[Norm Pollock]

Geschrieben von:/Posted by: Norm Pollock at 23 August 2004 19:28:25:
Als Antwort auf:/In reply to: Arasan 8.0 geschrieben von:/posted by: Jon Dart at 22 August 2004 23:52:32:

Arasan 8.0 is available from http://www.arasanchess.org.
Changes in Arasan 8.0:
1) Extended futility pruning is implemented now. Futility pruning and
razoring can be combined.
2) Passed pawns are scored higher.
3) "Capture" extension is now disabled by default.
4) Fixed some bugs in king-pawn hash table code
5) Minor changes to hash probe code.
6) Default tablebases cache size increased, changes to tablebase probe
logic. Generally this version does fewer probes.
7) New egtb code from Eugene Nalimov & Andrew Kadatch.
8) Significant changes to evaluation. Evaluation is simpler and
faster. A small evaluation cache is used.
9) Various speed enhancements, in move generator and elsewhere.
10) Some fixes to 50-move draw handling.
11) "makebook" program can now filter moves so that only moves
occurring with a minimum frequency will be included. New book
uses minimum frequency of 6 from a large PGN game collection.
12) a few minor GUI fixes.
--Jon

Thanks. 8.0 looks much improved in preliminary tests. You did a good job!
-Norm

[Tom Likens]

Geschrieben von:/Posted by: Tom Likens at 23 August 2004 20:44:50:
Als Antwort auf:/In reply to: Arasan 8.0 geschrieben von:/posted by: Jon Dart at 22 August 2004 23:52:32:

Jon,
I've a couple of minor questions which I've inserted below:

Arasan 8.0 is available from http://www.arasanchess.org.
Changes in Arasan 8.0:
1) Extended futility pruning is implemented now. Futility pruning and
razoring can be combined.
2) Passed pawns are scored higher.
3) "Capture" extension is now disabled by default.
4) Fixed some bugs in king-pawn hash table code
5) Minor changes to hash probe code.
6) Default tablebases cache size increased, changes to tablebase probe
logic. Generally this version does fewer probes.
7) New egtb code from Eugene Nalimov & Andrew Kadatch.
8) Significant changes to evaluation. Evaluation is simpler and
faster. A small evaluation cache is used.
9) Various speed enhancements, in move generator and elsewhere.
10) Some fixes to 50-move draw handling.
11) "makebook" program can now filter moves so that only moves
occurring with a minimum frequency will be included. New book
uses minimum frequency of 6 from a large PGN game collection.
12) a few minor GUI fixes.
--Jon

What led you to turn off the capture extension?
I haven't been following this very closely, what is the benefit of the
new EGTB code? I'm curious if it's something I might want to implement.
This is interesting, since I'm in the middle of reducing my evaluation
as well, (I'm trying to place more of the "brains" in the search).
regards,
--tom

[Jon Dart]

Geschrieben von:/Posted by: Jon Dart at 23. August 2004 20:56:
Als Antwort auf:/In reply to: Re: Arasan 8.0 geschrieben von:/posted by: Tom Likens at 23 August 2004 20:44:50:

Jon,
I've a couple of minor questions which I've inserted below:
What led you to turn off the capture extension?

7) New egtb code from Eugene Nalimov & Andrew Kadatch.

I haven't been following this very closely, what is the benefit of the
new EGTB code? I'm curious if it's something I might want to implement.

Testing didn't show an improvement with it on. But it was never a big
factor anyway, the number of capture extensions was small.
Main thing for me is the new code compiles under VC++ .NET - I had problems
with the earlier code with that compiler.
--Jon

[Jon Dart]

Geschrieben von:/Posted by: Jon Dart at 23 August 2004 21:01:53:
Als Antwort auf:/In reply to: Re: Arasan 8.0 geschrieben von:/posted by: Roger Brown at 23 August 2004 04:46:13:

Arasan 8.0 is available from http://www.arasanchess.org.

Hello Jon,
Thanks.
You are no longer publishing updates of the large Arasan book?

I haven't removed it from the site: you can still download it via
http://www.arasanchess.org/bigbook.zip. But most people I think shouldn't
need it now - the default book shipped with Arasan 8.0 is bigger than that
supplied in earlier versions (over 50,000 moves now). The "bigbook" was
technically larger but many moves in it have low frequencies and Arasan's
standard book move selection code will cause them not to be played for
that reason. In contrast, most moves in the 8.0 book will be considered
playable, because they have at least a minimum frequency (6).
--Jon

[Roger Brown]

Geschrieben von:/Posted by: Roger Brown at 23 August 2004 21:20:15:
Als Antwort auf:/In reply to: Re: Arasan 8.0 geschrieben von:/posted by: Jon Dart at 23 August 2004 21:01:53:

The "bigbook" was

technically larger but many moves in it have low frequencies and Arasan's
standard book move selection code will cause them not to be played for
that reason. In contrast, most moves in the 8.0 book will be considered
playable, because they have at least a minimum frequency (6).
--Jon

I will promptly put the big book on ice and use the one shipped with the engine.
Thanks.
Later.