weaker than fairymax 4.8Q but stronger than (deepbrute)pos

Discussion of chess software programming and technical issues.

Moderators: hgm, Rebel, chrisw

flok

weaker than fairymax 4.8Q but stronger than (deepbrute)pos

Post by flok »

Hi,

I've been googling the world but could not find it but are there any chess engines that run on linux that are weaker than fairymax 4.8q? (but stronger than (deepbrute)pos)
Because fairymax is soo much stronger than dbp that seeing if it improved is difficult as it'll always loose of it.
Henk
Posts: 7216
Joined: Mon May 27, 2013 10:31 am

Re: weaker than fairymax 4.8Q but stronger than (deepbrute)p

Post by Henk »

flok wrote:Hi,

I've been googling the world but could not find it but are there any chess engines that run on linux that are weaker than fairymax 4.8q? (but stronger than (deepbrute)pos)
Because fairymax is soo much stronger than dbp that seeing if it improved is difficult as it'll always loose of it.
Currently Skipper is playing worse than fairy max. For my last changes made it play worse but I don't want to go back to an older version. My engine is kept private.

ELO of Skipper is around 1800-1900 I guess, but I don't know.


You can play against shredder medium on internet. But you have to make the moves on the screen. There is also Lichess with different playing levels.

Shredder medium has ELO 1600.
Last edited by Henk on Wed Sep 17, 2014 3:35 pm, edited 1 time in total.
flok

Re: weaker than fairymax 4.8Q but stronger than (deepbrute)p

Post by flok »

Henk wrote:
flok wrote:Hi,

I've been googling the world but could not find it but are there any chess engines that run on linux that are weaker than fairymax 4.8q? (but stronger than (deepbrute)pos)
Because fairymax is soo much stronger than dbp that seeing if it improved is difficult as it'll always loose of it.
Currently Skipper is playing worse than fairy max. For my last changes made it play worse but I don't want to go back to an older version. My engine is kept private.

ELO of Skipper is around 1800-1900 I guess, but I don't know.


You can play against shredder medium on internet. But you have to make the move on the screen. There is also Lichess with different playing levels.

Shredder medium has ELO 1600.
Thanks but I'm searching for something doing xboard/uci so that I can use it as a reference point in automated testing.
Henk
Posts: 7216
Joined: Mon May 27, 2013 10:31 am

Re: weaker than fairymax 4.8Q but stronger than (deepbrute)p

Post by Henk »

flok wrote:
Henk wrote:
flok wrote:Hi,

I've been googling the world but could not find it but are there any chess engines that run on linux that are weaker than fairymax 4.8q? (but stronger than (deepbrute)pos)
Because fairymax is soo much stronger than dbp that seeing if it improved is difficult as it'll always loose of it.
Currently Skipper is playing worse than fairy max. For my last changes made it play worse but I don't want to go back to an older version. My engine is kept private.

ELO of Skipper is around 1800-1900 I guess, but I don't know.


You can play against shredder medium on internet. But you have to make the move on the screen. There is also Lichess with different playing levels.

Shredder medium has ELO 1600.
Thanks but I'm searching for something doing xboard/uci so that I can use it as a reference point in automated testing.
Yes ok. That's what I already thought. But playing many very short bullet games can be misleading too. For instance when you want to tune R (formulas).
matthewlai
Posts: 793
Joined: Sun Aug 03, 2014 4:48 am
Location: London, UK

Re: weaker than fairymax 4.8Q but stronger than (deepbrute)p

Post by matthewlai »

flok wrote:
Henk wrote:
flok wrote:Hi,

I've been googling the world but could not find it but are there any chess engines that run on linux that are weaker than fairymax 4.8q? (but stronger than (deepbrute)pos)
Because fairymax is soo much stronger than dbp that seeing if it improved is difficult as it'll always loose of it.
Currently Skipper is playing worse than fairy max. For my last changes made it play worse but I don't want to go back to an older version. My engine is kept private.

ELO of Skipper is around 1800-1900 I guess, but I don't know.


You can play against shredder medium on internet. But you have to make the move on the screen. There is also Lichess with different playing levels.

Shredder medium has ELO 1600.
Thanks but I'm searching for something doing xboard/uci so that I can use it as a reference point in automated testing.
I have a gauntlet of open source Linux/OSX engines if you are interested -
http://talkchess.com/forum/viewtopic.ph ... highlight=

I fixed them all to compile under GCC 4.8 and LLVM.

Most of them are stronger, but I tried including weaker ones, too. The problem with weaker ones is that they tend to be buggy as well. It's hard to find weaker ones that are high quality (don't crash or lose on time, etc).

All the engines in my gauntlet should play "0 0:01 0.1" (1 second total, 0.1 second increment) games fine.

I also have my own program for matching engines, since xboard doesn't work well at that kind of time controls.

Repository is here (go in to cmatch, and type make - requires LLVM or GCC 4.9, just run it without arguments to see help) -
http://bitbucket.matthewlai.ca/chessenginetools
Disclosure: I work for DeepMind on the AlphaZero project, but everything I say here is personal opinion and does not reflect the views of DeepMind / Alphabet.
flok

Re: weaker than fairymax 4.8Q but stronger than (deepbrute)p

Post by flok »

Matthew,

Great initiative!
Maybe you can add a list of their bugs to it? And maybe a list of their relative strengths?
matthewlai
Posts: 793
Joined: Sun Aug 03, 2014 4:48 am
Location: London, UK

Re: weaker than fairymax 4.8Q but stronger than (deepbrute)p

Post by matthewlai »

flok wrote:Matthew,

Great initiative!
Maybe you can add a list of their bugs to it? And maybe a list of their relative strengths?
When I noticed bugs I actually contacted the authors to fix them, so I only included engines that ran without bugs, or engines where the author fixed the bugs. I fixed bugs myself too when they were obvious and easy to fix, though I don't really have time to spend significant amount of time on one engine. So if an engine is buggy beyond quick fix, and the author doesn't reply or doesn't want to fix it, I just drop it from the gauntlet.

I actually contacted about 6-7 engine authors while compiling the gauntlet :).

I am planning to write a tournament management program, too. Then I can do a quick round robin or something and make a rating list. That's still at least a few weeks out, though. Don't really have time these days.
Disclosure: I work for DeepMind on the AlphaZero project, but everything I say here is personal opinion and does not reflect the views of DeepMind / Alphabet.
jdart
Posts: 4366
Joined: Fri Mar 10, 2006 5:23 am
Location: http://www.arasanchess.org

Re: weaker than fairymax 4.8Q but stronger than (deepbrute)p

Post by jdart »

Arasan (http://www.arasanchess.org) has a variable-strength feature, so you could probably find a strength level that is appropriate.

Also TSCP (http://www.tckerrigan.com/Chess/TSCP) is weak but functional. It has a rudimentary implementation of the xboard protocol.

--Jon
bob
Posts: 20943
Joined: Mon Feb 27, 2006 7:30 pm
Location: Birmingham, AL

Re: weaker than fairymax 4.8Q but stronger than (deepbrute)p

Post by bob »

flok wrote:Hi,

I've been googling the world but could not find it but are there any chess engines that run on linux that are weaker than fairymax 4.8q? (but stronger than (deepbrute)pos)
Because fairymax is soo much stronger than dbp that seeing if it improved is difficult as it'll always loose of it.
You might try Crafty, compiled with -DSKILL. then add skill=n (where skill is between 1 and 100 to the command line or the .craftyrc file. skill=1 plays VERY weakly which will give you something you can tune to play equally with your program if you want.
Richard Allbert
Posts: 792
Joined: Wed Jul 19, 2006 9:58 am

Re: weaker than fairymax 4.8Q but stronger than (deepbrute)p

Post by Richard Allbert »

There's Vice which is very basic, source should compile on linux ok.

http://bluefever.net/Downloads/ViceReadMe.html