Texel 1.05

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

Moderator: Ras

petero2
Posts: 726
Joined: Mon Apr 19, 2010 7:07 pm
Location: Sweden
Full name: Peter Osterlund

Texel 1.05

Post by petero2 »

Version 1.05 of my chess engine Texel is now available for download from http://dl.dropboxusercontent.com/u/89684995/texel105.7z

The most notable changes are support for Gaviota and Syzygy tablebases and support for multi-threaded search on Android devices.

There are 7 executables in the download package:

texel-arm : For the armv7-a architecture. Should work on most modern Android devices.
texel32.exe : For 32-bit windows systems with SSE42 and POPCOUNT.
texel32old.exe : For 32-bit windows systems without SSE42 and POPCOUNT.
texel64 : For 64-bit linux intel systems with SSE42 and POPCOUNT.
texel64.exe : For 64-bit windows intel systems with SSE42 and POPCOUNT.
texel64amd.exe : For 64-bit windows systems with SSE42 and POPCOUNT.
texel64old.exe : For 64-bit windows systems without SSE42 and POPCOUNT.

For more Android executables, see the "Texel Chess Engine" app in Google play.

The linux executable require a fairly recent 64-bit linux distribution to run. To compile your own linux version, see the readme.txt file.

Texel 1.05 is about 57 elo stronger than Texel 1.04 on my linux computer at very fast time controls.

Changes:
- Added support for Gaviota and Syzygy tablebases.
- Improved the built in opening book.
- Implemented UCI option "Clear Hash".
- Optimization for NUMA hardware.
- Earlier transposition table prefetch.
- Added evaluation of "semi-trapped" rooks.
- Removed the guarded passed pawn evaluation.
- Don't rely on the transposition table to communicate search results from helper to master thread.
- Use more time early in sudden death and increment games.
- Smaller delta in aspiration window loop.
- Optimized history heuristic update formula.
- Improved fortress detection for bishop and pawns vs pawns.
- Detection of "16 pawn fortress positions".
- Added bonus for pieces protected by pawns.
- Fixed incorrect draw scores in multipv mode.
- Made piece square tables have left/right asymmetry.
- Detection of KQKRP fortress positions.
- Disabled LMP in PV nodes.
- Only do check extension for negative SEE moves if the remaining depth is small.
- Implemented singular extensions.
- More aggressive LMR.
- Fixed makefile to make multi-threaded search work for Android.
User avatar
Dr.Wael Deeb
Posts: 9773
Joined: Wed Mar 08, 2006 8:44 pm
Location: Amman,Jordan

Re: Texel 1.05

Post by Dr.Wael Deeb »

Thanks Peter....
Dr.D
_No one can hit as hard as life.But it ain’t about how hard you can hit.It’s about how hard you can get hit and keep moving forward.How much you can take and keep moving forward….
User avatar
Graham Banks
Posts: 44584
Joined: Sun Feb 26, 2006 10:52 am
Location: Auckland, NZ

Re: Texel 1.05

Post by Graham Banks »

petero2 wrote:Version 1.05 of my chess engine Texel is now available for download from http://dl.dropboxusercontent.com/u/89684995/texel105.7z
Excellent news.
I'll put it into my Division 2 tourney later. :)
gbanksnz at gmail.com
Dann Corbit
Posts: 12792
Joined: Wed Mar 08, 2006 8:57 pm
Location: Redmond, WA USA

Re: Texel 1.05

Post by Dann Corbit »

On a day like this, it's good to be alive.

Of course, it's good to be alive on pretty much any day, but even better when snazzy, updated chess engines are coming out of the woodwork.
:wink:
carldaman
Posts: 2287
Joined: Sat Jun 02, 2012 2:13 am

Re: Texel 1.05

Post by carldaman »

Thanks Peter!

So great to see another strong original engine crack the 3000 barrier

:D

Regards,
CL
gerold
Posts: 10121
Joined: Thu Mar 09, 2006 12:57 am
Location: van buren,missouri

Re: Texel 1.05

Post by gerold »

Thanks Peter.

Best,
Gerold.
Gurcan Uckardes
Posts: 196
Joined: Wed Oct 29, 2014 12:42 am

Re: Texel 1.05

Post by Gurcan Uckardes »

Very nice to hear we have now a multi core Texel in the pocket. Will be fun to fully load the Exynos babies. Thanks Peter.

Welcome back to Dann Corbit by the way. After Jim Ablett, it's another good news to hear.

A good year for chess is on the way.
My blog for Android users: http://chesstroid.blogspot.com
lucasart
Posts: 3241
Joined: Mon May 31, 2010 1:29 pm
Full name: lucasart

Re: Texel 1.05

Post by lucasart »

petero2 wrote:Version 1.05 of my chess engine Texel is now available for download from http://dl.dropboxusercontent.com/u/89684995/texel105.7z

The most notable changes are support for Gaviota and Syzygy tablebases and support for multi-threaded search on Android devices.

There are 7 executables in the download package:

texel-arm : For the armv7-a architecture. Should work on most modern Android devices.
texel32.exe : For 32-bit windows systems with SSE42 and POPCOUNT.
texel32old.exe : For 32-bit windows systems without SSE42 and POPCOUNT.
texel64 : For 64-bit linux intel systems with SSE42 and POPCOUNT.
texel64.exe : For 64-bit windows intel systems with SSE42 and POPCOUNT.
texel64amd.exe : For 64-bit windows systems with SSE42 and POPCOUNT.
texel64old.exe : For 64-bit windows systems without SSE42 and POPCOUNT.

For more Android executables, see the "Texel Chess Engine" app in Google play.

The linux executable require a fairly recent 64-bit linux distribution to run. To compile your own linux version, see the readme.txt file.

Texel 1.05 is about 57 elo stronger than Texel 1.04 on my linux computer at very fast time controls.

Changes:
- Added support for Gaviota and Syzygy tablebases.
- Improved the built in opening book.
- Implemented UCI option "Clear Hash".
- Optimization for NUMA hardware.
- Earlier transposition table prefetch.
- Added evaluation of "semi-trapped" rooks.
- Removed the guarded passed pawn evaluation.
- Don't rely on the transposition table to communicate search results from helper to master thread.
- Use more time early in sudden death and increment games.
- Smaller delta in aspiration window loop.
- Optimized history heuristic update formula.
- Improved fortress detection for bishop and pawns vs pawns.
- Detection of "16 pawn fortress positions".
- Added bonus for pieces protected by pawns.
- Fixed incorrect draw scores in multipv mode.
- Made piece square tables have left/right asymmetry.
- Detection of KQKRP fortress positions.
- Disabled LMP in PV nodes.
- Only do check extension for negative SEE moves if the remaining depth is small.
- Implemented singular extensions.
- More aggressive LMR.
- Fixed makefile to make multi-threaded search work for Android.
Congratulations for all your hard work. I'm testing now against Critter 1.6a (tc=15+0.15 threads=7), and it seems to be holding its ground better than I expected (remember Critter is an absolute killer at these short tc).

It's rare to see honest engine developers break the 3100 elo barrier. I hope more engine testers were capable of understanding the difference of an engine like Texel and a cheap clone (Amitis, Elektro, Shark, Taipan, etc.)
Theory and practice sometimes clash. And when that happens, theory loses. Every single time.
Frank Quisinsky
Posts: 7045
Joined: Wed Nov 18, 2009 7:16 pm
Location: Gutweiler, Germany
Full name: Frank Quisinsky

Re: Texel 1.05

Post by Frank Quisinsky »

Hi Peter,

the next good news for 2015!

Of course I will test too but at the moment I have a lot to do with my still running tourney and 7 updates I added.

ToDo:
http://www.amateurschach.de/main/_todo.htm

It's my feeling that computer chess is on the good way in the fight vs. clones. More and more important persons are back ... Jim, Dann and much programmer.

Your engine is "animation pure" for each other programmer!

Good luck with your great engine Texel and regards to Sweden!

Best
Frank

My wish is a new official version from Octochess, perhaps compiled by Jim or Dann. Perhaps time to write about wishes if all others come true.
petero2
Posts: 726
Joined: Mon Apr 19, 2010 7:07 pm
Location: Sweden
Full name: Peter Osterlund

Re: Texel 1.05

Post by petero2 »

Gurcan Uckardes wrote:Very nice to hear we have now a multi core Texel in the pocket. Will be fun to fully load the Exynos babies. Thanks Peter.
One warning though. Texel's SMP implementation is a bit unusual, which has the consequence that it does not kick in until the search depth exceeds 10. Normally this is not a problem on desktop computers, but on android hardware with short time controls, this could become a problem.

Therefore SMP for android is expected to work best at longer time controls and for analysis.