JaksaH Engine

Questions and comments related to CCRL testing study
Post Reply
User avatar
ajaksic
Posts: 5
Joined: Mon Nov 08, 2010 6:21 pm
Sign-up code: 10159
Contact:

JaksaH Engine

Post by ajaksic »

Hello,

My name is Aleksandar Jaksic, and I am the author of JaksaH chess engine: https://sites.google.com/site/jaksahchess/

I've started developing the engine back in 2004, but it's been really last two months when I improved its stability and strength of play.

I would appreciate very much if you guys can include this engine in your test tourneys. According to my owns tests, JaksaH v1.06 x64 plays stronger chess than TSCP 1.81. So my best effort ELO estimate is somewhere between 1800-1900 for 40/40 and 40/20.

Please let me know if you have any questions.

Thank you!
Aleksandar
User avatar
Kirill Kryukov
Site Admin
Posts: 7399
Joined: Sun Dec 18, 2005 9:58 am
Sign-up code: 0
Location: Mishima, Japan
Contact:

Re: JaksaH Engine

Post by Kirill Kryukov »

Thanks Aleksandar, I added it to my testing queue. Which country of origin should be listed for this engine?
User avatar
ajaksic
Posts: 5
Joined: Mon Nov 08, 2010 6:21 pm
Sign-up code: 10159
Contact:

Re: JaksaH Engine

Post by ajaksic »

Thank you Kirill! Country is Serbia.
User avatar
Kirill Kryukov
Site Admin
Posts: 7399
Joined: Sun Dec 18, 2005 9:58 am
Sign-up code: 0
Location: Mishima, Japan
Contact:

Re: JaksaH Engine

Post by Kirill Kryukov »

Don't thank me yet, as my queue is notoriously long, and it may take months before I get to testing JaksaH. Hopefully someone else can provide faster results.
User avatar
Graham Banks
Posts: 26942
Joined: Sun Dec 18, 2005 5:47 pm
Sign-up code: 0
Location: Auckland, NZ

Re: JaksaH Engine

Post by Graham Banks »

Hi Aleksandar,

at present I'm restricting my 40/40 testing to engines that are 2000 ELO or above, otherwise I'd never give the engines I test a decent go.
Please keep developing your engine because I like nothing more than to encourage new authors by testing their engines. You're not far off 2000 ELO! :P

Cheers,
Graham.
User avatar
ajaksic
Posts: 5
Joined: Mon Nov 08, 2010 6:21 pm
Sign-up code: 10159
Contact:

Re: JaksaH Engine

Post by ajaksic »

Thank you Graham. I certainly have plans to continue improving the strength of my engine, and I am aware of many areas where I can make significant improvements. One of those areas is openings. I haven't implemented anything there yet.

What approach you suggest in this area? Should I go with my own book or to try to utilize one of the widely used non-proprietary opening book formats (such is Polyglot). Is it normally expected that all engines use the very same book during tournaments? What's the common practice among the leading engines?
User avatar
Graham Banks
Posts: 26942
Joined: Sun Dec 18, 2005 5:47 pm
Sign-up code: 0
Location: Auckland, NZ

Re: JaksaH Engine

Post by Graham Banks »

ajaksic wrote:Thank you Graham. I certainly have plans to continue improving the strength of my engine, and I am aware of many areas where I can make significant improvements. One of those areas is openings. I haven't implemented anything there yet.

What approach you suggest in this area? Should I go with my own book or to try to utilize one of the widely used non-proprietary opening book formats (such is Polyglot). Is it normally expected that all engines use the very same book during tournaments? What's the common practice among the leading engines?
Both CCRL and CEGT test with generic opening books, so engines aren't allowed to use their own book. However, other testers prefer to use the engine's own book. Therefore it's important either to give the option of disabling an engine's own book or allowing the engine to play quite happily without its opening book file.
From watching many engine v engine games over the years, the biggest weaknesses in the lower rated engines tend to be king safety, the recognition and handling of passed pawns, and allowing the exchanging off of a last piece only to go into a dead lost king and pawns ending.

Cheers,
Graham.
Post Reply