Open-source improvements released

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

Moderators: bob, hgm, Harvey Williamson

Forum rules
This textbox is used to restore diagrams posted with the [d] tag before the upgrade.
Lyudmil Tsvetkov
Posts: 6052
Joined: Tue Jun 12, 2012 10:41 am

Re: Open-source improvements released

Post by Lyudmil Tsvetkov » Wed Jun 28, 2017 7:11 am

kranium wrote:Hi all-

I've been playing quite a bit with 3 open-source programs, made some progress and have put it all up on Github:

Fridolin 2.00
- code optimizations, eval tuning, etc.
published on GitHub as Jinx 1.0 (~ +146 Elo)
https://github.com/FireFather/jinx

Bobcat 8.0
- code optimizations, eval tuning, etc.
published on GitHub as Tomcat 1.0 (~ +45 Elo)
https://github.com/FireFather/tomcat

Gull 3
published on GitHub as Seagull 1.0 (~ +5 Elo)
https://github.com/FireFather/seagull

Seagull changes:
- source code cleaned up, simplified, and broken up into various source and header files
- a couple of VS code analysis fixes (ex: gen_kpk() was causing stack exceed error)
- compiler warnings resolved up to level 4
- benchmark and perft utilities added (type 'bench' and engine will write a date-stamped text file with results)
- support for syzygy tablebases

All include Visual Studio 2015 project files and x64 binaries.

Best Regards,
Norm
thanks Norman!

+5 elo to Fridolin and +146 elo to Gull would have been much more appreciated. :)

Frank Quisinsky
Posts: 4852
Joined: Wed Nov 18, 2009 6:16 pm
Location: Trier, Germany
Contact:

Re: Open-source improvements released

Post by Frank Quisinsky » Wed Jun 28, 2017 7:20 am

Hi Norm,

good work (nice to see all the important changes for Gull). Bobcat with 45 Elo more is also great.

To have the same with Phalanx could be interesting. I like Phalanx a lot and have the most fun with Phalanx in winboard times.

Best
Frank
I like computer chess!

User avatar
Kotlov
Posts: 209
Joined: Fri Jul 10, 2015 7:23 pm
Location: Russia

Re: Open-source improvements released

Post by Kotlov » Wed Jun 28, 2017 8:14 am

Personally, I'm against making different engines equally strong.

Frank Quisinsky
Posts: 4852
Joined: Wed Nov 18, 2009 6:16 pm
Location: Trier, Germany
Contact:

Re: Open-source improvements released

Post by Frank Quisinsky » Wed Jun 28, 2017 10:12 am

Eugene,

Open Source!!
We have the original version or older versions of the original!

Often I am playing vs. very old engines like SSEChess or ETChess a lot of others, like DOS Tjes.

Enough for each of us is available!

Best
Frank
I like computer chess!

carldaman
Posts: 1708
Joined: Sat Jun 02, 2012 12:13 am

Re: Open-source improvements released

Post by carldaman » Thu Jul 06, 2017 8:33 pm

kranium wrote:Hi all-

I've been playing quite a bit with 3 open-source programs, made some progress and have put it all up on Github:

Fridolin 2.00
- code optimizations, eval tuning, etc.
published on GitHub as Jinx 1.0 (~ +146 Elo)
https://github.com/FireFather/jinx

Bobcat 8.0
- code optimizations, eval tuning, etc.
published on GitHub as Tomcat 1.0 (~ +45 Elo)
https://github.com/FireFather/tomcat

Gull 3
published on GitHub as Seagull 1.0 (~ +5 Elo)
https://github.com/FireFather/seagull

Seagull changes:
- source code cleaned up, simplified, and broken up into various source and header files
- a couple of VS code analysis fixes (ex: gen_kpk() was causing stack exceed error)
- compiler warnings resolved up to level 4
- benchmark and perft utilities added (type 'bench' and engine will write a date-stamped text file with results)
- support for syzygy tablebases

All include Visual Studio 2015 project files and x64 binaries.

Best Regards,
Norm
Thanks a lot, Norman. :)

I have to say that it is quite surprising that

http://computer-chess.org/doku.php?id=c ... ngine_list

has not listed your releases, while including some other derivatives (Sugar, for example), especially since you obviously had to put in some real effort into producing actual Elo gains over the originals.

Regards,
CL

Ron Murawski
Posts: 397
Joined: Sun Oct 29, 2006 3:38 am
Location: Schenectady, NY
Contact:

Re: Open-source improvements released

Post by Ron Murawski » Thu Jul 13, 2017 7:06 pm

carldaman wrote:
kranium wrote:Hi all-

I've been playing quite a bit with 3 open-source programs, made some progress and have put it all up on Github:

Fridolin 2.00
- code optimizations, eval tuning, etc.
published on GitHub as Jinx 1.0 (~ +146 Elo)
https://github.com/FireFather/jinx

Bobcat 8.0
- code optimizations, eval tuning, etc.
published on GitHub as Tomcat 1.0 (~ +45 Elo)
https://github.com/FireFather/tomcat

Gull 3
published on GitHub as Seagull 1.0 (~ +5 Elo)
https://github.com/FireFather/seagull

Seagull changes:
- source code cleaned up, simplified, and broken up into various source and header files
- a couple of VS code analysis fixes (ex: gen_kpk() was causing stack exceed error)
- compiler warnings resolved up to level 4
- benchmark and perft utilities added (type 'bench' and engine will write a date-stamped text file with results)
- support for syzygy tablebases

All include Visual Studio 2015 project files and x64 binaries.

Best Regards,
Norm
Thanks a lot, Norman. :)

I have to say that it is quite surprising that

http://computer-chess.org/doku.php?id=c ... ngine_list

has not listed your releases, while including some other derivatives (Sugar, for example), especially since you obviously had to put in some real effort into producing actual Elo gains over the originals.

Regards,
CL
Hi Carl,

Please send me a complete list of derivatives that you feel deserve or don't deserve listing and I will consider them. I have little time for computer chess these days, so my decisions may seem arbitrary to a knowledgeable insider such as yourself. It is the amount of coding changes that interests me here: What percentage of the original code remains unchanged compared to new code? Are the changes substantive, or merely a restating, a re-arrangement of original code, or an implementation of well-known techniques? Is the fixing of several bugs sufficient to announce another derivative engine with a different name? These are the questions that I do not have time to research and answer.

Keep in mind that Elo gains have nothing to do with an engine's authorship.

I see Norman's valuable contributions as an appeal to the authors of these engines to implement his bugfixes and improvements. If they do, I will consider that as a new version and it is possible I will add Norman's name as a co-author of that engine.

Best regards,
Ron

Engin
Posts: 832
Joined: Mon Jan 05, 2009 6:40 pm
Location: Germany
Full name: Engin Üstün
Contact:

Re: Open-source improvements released

Post by Engin » Sat Jul 15, 2017 11:48 am

A BIG GOOD MOVE from you Norman to helping improve weaker engines like Fridolin i am sure Sommerfeld will be very thanks to you.

User avatar
Guenther
Posts: 3014
Joined: Wed Oct 01, 2008 4:33 am
Location: Regensburg, Germany
Full name: Guenther Simon
Contact:

Re: Open-source improvements released

Post by Guenther » Sat Jul 15, 2017 12:15 pm

Ron Murawski wrote:
carldaman wrote:
kranium wrote:Hi all-

I've been playing quite a bit with 3 open-source programs, made some progress and have put it all up on Github:

Fridolin 2.00
- code optimizations, eval tuning, etc.
published on GitHub as Jinx 1.0 (~ +146 Elo)
https://github.com/FireFather/jinx

Bobcat 8.0
- code optimizations, eval tuning, etc.
published on GitHub as Tomcat 1.0 (~ +45 Elo)
https://github.com/FireFather/tomcat

Gull 3
published on GitHub as Seagull 1.0 (~ +5 Elo)
https://github.com/FireFather/seagull

Seagull changes:
- source code cleaned up, simplified, and broken up into various source and header files
- a couple of VS code analysis fixes (ex: gen_kpk() was causing stack exceed error)
- compiler warnings resolved up to level 4
- benchmark and perft utilities added (type 'bench' and engine will write a date-stamped text file with results)
- support for syzygy tablebases

All include Visual Studio 2015 project files and x64 binaries.

Best Regards,
Norm
Thanks a lot, Norman. :)

I have to say that it is quite surprising that

http://computer-chess.org/doku.php?id=c ... ngine_list

has not listed your releases, while including some other derivatives (Sugar, for example), especially since you obviously had to put in some real effort into producing actual Elo gains over the originals.

Regards,
CL
Hi Carl,

Please send me a complete list of derivatives that you feel deserve or don't deserve listing and I will consider them. I have little time for computer chess these days, so my decisions may seem arbitrary to a knowledgeable insider such as yourself. It is the amount of coding changes that interests me here: What percentage of the original code remains unchanged compared to new code? Are the changes substantive, or merely a restating, a re-arrangement of original code, or an implementation of well-known techniques? Is the fixing of several bugs sufficient to announce another derivative engine with a different name? These are the questions that I do not have time to research and answer.

Keep in mind that Elo gains have nothing to do with an engine's authorship.

I see Norman's valuable contributions as an appeal to the authors of these engines to implement his bugfixes and improvements. If they do, I will consider that as a new version and it is possible I will add Norman's name as a co-author of that engine.

Best regards,
Ron
Because no one so far mentioned it, I want to add that all the numbers (elo increase) given are w/o any objective reference so far.
Current foe list count : [93]
http://rwbc-chess.de/chronology.htm

flok

Re: Open-source improvements released

Post by flok » Sat Aug 26, 2017 8:02 pm

I accidentally stumbled upon some open source chess program that clearly needs a bit of love.

https://github.com/flok99/Embla[/i]

User avatar
JVMerlino
Posts: 1003
Joined: Wed Mar 08, 2006 9:15 pm
Location: San Francisco, California

Re: Open-source improvements released

Post by JVMerlino » Sat Aug 26, 2017 8:26 pm

flok wrote:I accidentally stumbled upon some open source chess program that clearly needs a bit of love.

https://github.com/flok99/Embla
:lol:

For years I have considered just throwing my (embarrasingly bad) source for Myrddin up on Github and hoping somebody will take pity on it.

Post Reply