Arasan 20.3

Discussion of anything and everything relating to chess playing software and machines.

Moderators: hgm, Harvey Williamson, bob

jdart
Posts: 3476
Joined: Fri Mar 10, 2006 4:23 am
Location: http://www.arasanchess.org

Arasan 20.3

Post by jdart » Sun Nov 05, 2017 2:14 am

Arasan 20.3 is available from https://www.arasanchess.org.

Changes in Arasan 20.3:

1) Change queen mobility scoring: square reach beyond 23 squares gets
same score.
2) Re-work scoring for material imbalance with pawns.
3) Bug fixes for EPD/FEN handling (mostly affects utility programs).
4) Bug fixes for bitbases. Mate scores from bitbases were not being
transmitted to the search. Also bitbase scores were being added to
other eval terms which are not relevant for KPK.
5) Syzygy code (Fathom) is imported as a git submodule now.
6) Code restructuring, move piece values into Param class.
7) Change eval scale so Pawn value = 128 (vs. 1000 in 20.2).
8) Implement UCI "searchmoves" option to "go" command.
9) Bug fixes in parameter tuner.
10) If special-case endgame detected, skip other endgame scoring and
just use the special-case code.
11) In UCI MultiPV mode, ensure bestmove that is eventually sent is the
highest-scoring move, and re-order PVs if necessary so that they are
in decreasing score order.
12) Implement BMI2 (PEXT/PDEP) attack computation.
13) Change evasion pruning in qsearch.
14) Pack hash entries into 2 64-bits words instead of 3.

This is close to, but not identical with, the version that is playing in TCEC 10.

In my testing it is measurably stronger than 20.2, but not by a whole lot.

--Jon

User avatar
Graham Banks
Posts: 30534
Joined: Sun Feb 26, 2006 9:52 am
Location: Auckland, NZ

Re: Arasan 20.3

Post by Graham Banks » Sun Nov 05, 2017 4:20 am

Hi Jon,

good to see further progress. :)

Graham.
My email addresses:
gbanksnz at gmail.com
gbanksnz at yahoo.co.nz

User avatar
MikeB
Posts: 2505
Joined: Thu Mar 09, 2006 5:34 am
Location: Pen Argyl, Pennsylvania

Re: Arasan 20.3

Post by MikeB » Sun Nov 05, 2017 4:21 am

jdart wrote:Arasan 20.3 is available from https://www.arasanchess.org.

Changes in Arasan 20.3:

1) Change queen mobility scoring: square reach beyond 23 squares gets
same score.
2) Re-work scoring for material imbalance with pawns.
3) Bug fixes for EPD/FEN handling (mostly affects utility programs).
4) Bug fixes for bitbases. Mate scores from bitbases were not being
transmitted to the search. Also bitbase scores were being added to
other eval terms which are not relevant for KPK.
5) Syzygy code (Fathom) is imported as a git submodule now.
6) Code restructuring, move piece values into Param class.
7) Change eval scale so Pawn value = 128 (vs. 1000 in 20.2).
8) Implement UCI "searchmoves" option to "go" command.
9) Bug fixes in parameter tuner.
10) If special-case endgame detected, skip other endgame scoring and
just use the special-case code.
11) In UCI MultiPV mode, ensure bestmove that is eventually sent is the
highest-scoring move, and re-order PVs if necessary so that they are
in decreasing score order.
12) Implement BMI2 (PEXT/PDEP) attack computation.
13) Change evasion pruning in qsearch.
14) Pack hash entries into 2 64-bits words instead of 3.

This is close to, but not identical with, the version that is playing in TCEC 10.

In my testing it is measurably stronger than 20.2, but not by a whole lot.

--Jon
Thanks once again Jon!

Hugo
Posts: 707
Joined: Tue Dec 01, 2009 10:10 am

Re: Arasan 20.3

Post by Hugo » Sun Nov 05, 2017 7:14 am

At the times, when Windows 95 came out, I started with computers, and verry quick I also startetd with chess engines, chess programms.
One of the first free chess engines I saw was Arasan. It had its own interface, ready to play. I liked IT !
I am impressed that it is still under developement. Evertime I hear from Arasan updates, I remember the times of Win95 and the many nights I spent with reinstalling OS :D

Clemens Keck

jdart wrote:Arasan 20.3 is available from https://www.arasanchess.org.

Changes in Arasan 20.3:

1) Change queen mobility scoring: square reach beyond 23 squares gets
same score.
2) Re-work scoring for material imbalance with pawns.
3) Bug fixes for EPD/FEN handling (mostly affects utility programs).
4) Bug fixes for bitbases. Mate scores from bitbases were not being
transmitted to the search. Also bitbase scores were being added to
other eval terms which are not relevant for KPK.
5) Syzygy code (Fathom) is imported as a git submodule now.
6) Code restructuring, move piece values into Param class.
7) Change eval scale so Pawn value = 128 (vs. 1000 in 20.2).
8) Implement UCI "searchmoves" option to "go" command.
9) Bug fixes in parameter tuner.
10) If special-case endgame detected, skip other endgame scoring and
just use the special-case code.
11) In UCI MultiPV mode, ensure bestmove that is eventually sent is the
highest-scoring move, and re-order PVs if necessary so that they are
in decreasing score order.
12) Implement BMI2 (PEXT/PDEP) attack computation.
13) Change evasion pruning in qsearch.
14) Pack hash entries into 2 64-bits words instead of 3.

This is close to, but not identical with, the version that is playing in TCEC 10.

In my testing it is measurably stronger than 20.2, but not by a whole lot.

--Jon

Dariusz
Posts: 76
Joined: Sat Jun 13, 2015 8:08 am
Location: Poland

Re: Arasan 20.3

Post by Dariusz » Sun Nov 05, 2017 5:21 pm

Thank You Jon for this amazing engine.

I will testing it today on my mac :-))
Regards, Dariusz

Damir
Posts: 1770
Joined: Mon Feb 11, 2008 2:53 pm

Re: Arasan 20.3

Post by Damir » Sun Nov 05, 2017 6:15 pm

Thanks Jon for continous improvement of your engine. Do you have any estimation how much this version is stronger than 10.2 version ? :) :)

jdart
Posts: 3476
Joined: Fri Mar 10, 2006 4:23 am
Location: http://www.arasanchess.org

Re: Arasan 20.3

Post by jdart » Sun Nov 05, 2017 6:32 pm

estimation how much this version is stronger than 10.2 version
I guess you mean 20.2. I would guess +10 ELO or so on an arbitrary rating scale where 20.2 is about 2400 ELO. So a bit more on the CCRL or CEGT scale, probably.

Also note: I don't really put any effort into optimizing or testing the MacOS version. All I do is build it and do a little sanity checking.

--Jon

Damir
Posts: 1770
Joined: Mon Feb 11, 2008 2:53 pm

Re: Arasan 20.3

Post by Damir » Sun Nov 05, 2017 6:41 pm

Yes I meant 20.2 version. :) Sorry was a mistake on my part. :(

Dann Corbit
Posts: 8512
Joined: Wed Mar 08, 2006 7:57 pm
Location: Redmond, WA USA
Contact:

Re: Arasan 20.3

Post by Dann Corbit » Sun Nov 05, 2017 10:58 pm

It's always a good day when there is a new Arasan.
Taking ideas is not a vice, it is a virtue. We have another word for this. It is called learning.
But sharing ideas is an even greater virtue. We have another word for this. It is called teaching.

User avatar
Guenther
Posts: 2220
Joined: Wed Oct 01, 2008 4:33 am
Location: Regensburg, Germany
Contact:

Re: Arasan 20.3

Post by Guenther » Mon Nov 06, 2017 4:31 pm

jdart wrote:Arasan 20.3 is available from https://www.arasanchess.org.

Changes in Arasan 20.3:

1) Change queen mobility scoring: square reach beyond 23 squares gets
same score.
2) Re-work scoring for material imbalance with pawns.
3) Bug fixes for EPD/FEN handling (mostly affects utility programs).
4) Bug fixes for bitbases. Mate scores from bitbases were not being
transmitted to the search. Also bitbase scores were being added to
other eval terms which are not relevant for KPK.
5) Syzygy code (Fathom) is imported as a git submodule now.
6) Code restructuring, move piece values into Param class.
7) Change eval scale so Pawn value = 128 (vs. 1000 in 20.2).
8) Implement UCI "searchmoves" option to "go" command.
9) Bug fixes in parameter tuner.
10) If special-case endgame detected, skip other endgame scoring and
just use the special-case code.
11) In UCI MultiPV mode, ensure bestmove that is eventually sent is the
highest-scoring move, and re-order PVs if necessary so that they are
in decreasing score order.
12) Implement BMI2 (PEXT/PDEP) attack computation.
13) Change evasion pruning in qsearch.
14) Pack hash entries into 2 64-bits words instead of 3.

This is close to, but not identical with, the version that is playing in TCEC 10.

In my testing it is measurably stronger than 20.2, but not by a whole lot.

--Jon
Thanks a lot Jon. It's incredible that ArasanX is still a living project after decades!
Guenther Simon
http://rwbc-chess.de/

Post Reply