MZ CheSS ForuM
Would you like to react to this message? Create an account in a few clicks or log in to continue.
Top posting users this month
No user

Latest topics
» Judas PrO B1 all Test
Toga II 4.0 EmptySat Sep 26, 2015 12:10 am by klasman

»  SCCT – Tactical Test Suite's Results
Toga II 4.0 EmptyFri Sep 25, 2015 8:45 pm by echai

» Schooner 1.4.2
Toga II 4.0 EmptyThu Sep 24, 2015 6:33 pm by ful

» TEST Judas 1.02 vs Sugar pro 1.0
Toga II 4.0 EmptyThu Sep 24, 2015 5:56 pm by Trajkoski Goran 02

» Mars3.31_x64
Toga II 4.0 EmptyThu Sep 24, 2015 4:29 pm by ranomi2003

» Clubfoot
Toga II 4.0 EmptyThu Sep 24, 2015 2:21 am by ful

» V2-Samir.CTG
Toga II 4.0 EmptyThu Sep 24, 2015 2:18 am by ful

» Strong sicilian.7z
Toga II 4.0 EmptyThu Sep 24, 2015 2:17 am by ful

» GOALKEEPER.CTG
Toga II 4.0 EmptyThu Sep 24, 2015 2:16 am by ful

Last post


Toga II 4.0

Go down

Toga II 4.0 Empty Toga II 4.0

Post by MZ Fri Sep 04, 2015 9:22 am

This is latest development version of Toga here every once in a while for those who are interested. This is only a test version! It is not worth testing for CCRL/CEGT/IPON, as the release of TOGA II 4.0 is scheduled for June/July 2013.

The current version is about 30 elo stronger than Toga II 3.0 with equivalent compiles (scores ~56-57% in self-play ~6000 games at 15' + 0.05' and 60' + 0.05).
Only SLOW gcc compiles are provided in this package, and these are ~10% slower than the Intel compiles. If someone feels like making an Intel compile and/or compiles for other systems that would be great though Smile

If anyone would like to contribute to the Toga II project, any tests/bug reports/comments regarding this new Toga are appreciated. I hope you enjoy Toga II!

Jerry

Changes:

Improved Search
- move count based pruning
- improved null move reduction (3 + depth/4 from Stockfish used)
- better usage of hash in quiescent search
- bad captures reduced less
- history aging
- improved history increment
- removed useless verification search

Improved Evaluation
- removed queen mobility, which not only was not worth elo, but also reduced orthogonality as its scoring was based on attacking the opposition king
- rook/queen on 7th rank improved
- pawn support bonus added
- knight outpost bonus improved (also depends on no. of pawns)... D/L ... [You must be registered and logged in to see this link.]
MZ
MZ
 
 

Posts : 154
Join date : 2015-08-25

https://mzchessforum.forumattivo.it

Back to top Go down

Back to top

- Similar topics

 
Permissions in this forum:
You cannot reply to topics in this forum