New engine releases 2020

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

Moderators: hgm, Dann Corbit, Harvey Williamson

Forum rules
This textbox is used to restore diagrams posted with the [d] tag before the upgrade.
carldaman
Posts: 2051
Joined: Sat Jun 02, 2012 12:13 am

Re: New engine releases 2020

Post by carldaman » Mon Nov 02, 2020 6:03 am

Graham Banks wrote:
Sun Nov 01, 2020 11:27 pm
Sophos keeps deleting Counter 3.6. :roll:

http://talkchess.com/forum3/viewtopic.p ... &start=400

It's been showing up as a positive in most scans. You could try Tony Mokonen' compilation:

http://tonyschess.x10host.com/CounterGo-1.36.0.7z

User avatar
Graham Banks
Posts: 34869
Joined: Sun Feb 26, 2006 9:52 am
Location: Auckland, NZ

Re: New engine releases 2020

Post by Graham Banks » Mon Nov 02, 2020 6:43 am

carldaman wrote:
Mon Nov 02, 2020 6:03 am
Graham Banks wrote:
Sun Nov 01, 2020 11:27 pm
Sophos keeps deleting Counter 3.6. :roll:

http://talkchess.com/forum3/viewtopic.p ... &start=400

It's been showing up as a positive in most scans. You could try Tony Mokonen' compilation:

http://tonyschess.x10host.com/CounterGo-1.36.0.7z
I've tried all of the ones posted so far.
gbanksnz at gmail.com

Guenther
Posts: 3694
Joined: Wed Oct 01, 2008 4:33 am
Location: Regensburg, Germany
Full name: Guenther Simon
Contact:

Re: New engine releases 2020

Post by Guenther » Mon Nov 02, 2020 8:58 am

OliverBr wrote:
Sun Nov 01, 2020 4:50 pm
Guenther wrote:
Thu Oct 29, 2020 6:27 pm
I noticed that OliThink release compilations since version 5.69 won't run anymore on my old hardware.
Therefore I compiled current 5.89 now for no popcount with SSSE3 instructions.
Thank you very much! I am interesting in the changes to made in order to let it run without hardware popcount?
No changes needed, except changes for the flags in the makefile according to the hardware, the compiler does the workaround.
Sureley this no popcount version will run slower then on modern hardware, but I cannot test the difference.

In case of Olithink I just replaced -mavx2 by -mssse3.


Guenther
Posts: 3694
Joined: Wed Oct 01, 2008 4:33 am
Location: Regensburg, Germany
Full name: Guenther Simon
Contact:

Re: New engine releases 2020

Post by Guenther » Tue Nov 03, 2020 6:00 pm

New engine Monchester 0.99 WIN compilations available at my site now.
https://rwbc-chess.de/download.htm

fabianVDW
Posts: 146
Joined: Fri Mar 15, 2019 7:46 pm
Location: Germany
Full name: Fabian von der Warth

Re: New engine releases 2020

Post by fabianVDW » Tue Nov 03, 2020 9:36 pm

I have released a new version of FabChess at https://github.com/fabianvdW/FabChess/releases

Version 1.16 comes with a significant strength increase over the last releasted version 1.15 and a lot less code.
Changes include:
-Better handling of endgames
-Retune parameters with Andrews data
-Late move reductions changes
-Add new evaluation features: King piece (shogi-like) incrementally updated tables.
Currently only for pawns.

Selfplay tests show:
1 thread:
ELO | 65.57 +- 4.74 (95%)
SPRT | 60.0+0.6s Threads=1 Hash=64MB
LLR | 0.00 (-2.94, 2.94) [0.00, 0.00]
Games | N: 12320 W: 4777 L: 2479 D: 5064

Special thanks to
-Andrew Grant for his public dataset
-Terje (author of Weiss) for submitting several cleanup patches
-Connor (author of Seer) for the idea of using king piece tables
-Noob for really existing
This release wouldn't have been possible without them

If someone could test out the Linux compiles and check if they are working and that BMI2 faster POPCNT faster old on a modern cpu that would be nice :)
Author of FabChess: https://github.com/fabianvdW/FabChess
A UCI compliant chess engine written in Rust.
FabChessWiki: https://github.com/fabianvdW/FabChess/wiki
fabianvonderwarth@gmail.com

Gabor Szots
Posts: 704
Joined: Sat Jul 21, 2018 5:43 am
Location: Szentendre, Hungary
Full name: Gabor Szots

Re: New engine releases 2020

Post by Gabor Szots » Tue Nov 03, 2020 9:52 pm

Great news Fabian!
Gabor Szots
CCRL testing group

Guenther
Posts: 3694
Joined: Wed Oct 01, 2008 4:33 am
Location: Regensburg, Germany
Full name: Guenther Simon
Contact:

Re: New engine releases 2020

Post by Guenther » Wed Nov 04, 2020 1:03 pm

not yet mentioned:

Scorpio 3.10
https://github.com/dshawul/Scorpio/releases

Code: Select all

v3.0.10-Oct 30 NNUE support via egbbdll. It comes with a DarkHorse net (dh-0.2) if you want to test it out.
With a Stockfish net is is about +40 elo stronger than Xiphos but with DarkHorse it is -70 elo below.
Main NN net is updated. Bug fixes with nps display etc.
FoxSEE 5.2.1dev (src - reverting a part of 5.2.0 changes)
https://github.com/redsalmon91/FoxSEE/r ... tag/v5.2.1

If someone is interested (PM please) I will upload K2 0.96dev compilations.
K2 has a lot of changes since 0.95, but still no release since 1.5 years now.
https://github.com/serg-meus/k2/commits/master

User avatar
Graham Banks
Posts: 34869
Joined: Sun Feb 26, 2006 9:52 am
Location: Auckland, NZ

Re: New engine releases 2020

Post by Graham Banks » Sun Nov 08, 2020 10:33 pm

Does Supernova only work on bmi2?
gbanksnz at gmail.com

User avatar
Graham Banks
Posts: 34869
Joined: Sun Feb 26, 2006 9:52 am
Location: Auckland, NZ

Re: New engine releases 2020

Post by Graham Banks » Mon Nov 09, 2020 12:26 am

Graham Banks wrote:
Sun Nov 08, 2020 10:33 pm
Does Supernova only work on bmi2?
I know what the issue is, as the Raven author had the same happen with one of his releasesw:
[Graham] - Steve - what causes the 0xc00000b7 error with an engine?
[Steve - Raven] - from what i read, it's caused by compiling with 32-bit dlls
[Steve - Raven] - you can get around it by compiling with -static
gbanksnz at gmail.com

Gabor Szots
Posts: 704
Joined: Sat Jul 21, 2018 5:43 am
Location: Szentendre, Hungary
Full name: Gabor Szots

Re: New engine releases 2020

Post by Gabor Szots » Mon Nov 09, 2020 7:24 am

Graham Banks wrote:
Mon Nov 09, 2020 12:26 am
I know what the issue is, as the Raven author had the same happen with one of his releasesw:
[Graham] - Steve - what causes the 0xc00000b7 error with an engine?
[Steve - Raven] - from what i read, it's caused by compiling with 32-bit dlls
[Steve - Raven] - you can get around it by compiling with -static
I could not compile it with GCC (lots of warnings) but I could with clang.
Gabor Szots
CCRL testing group

Post Reply