Difference between revisions of "Tucano"

From Chessprogramming wiki
Jump to: navigation, search
Line 4: Line 4:
  
 
'''Tucano''', (TucanoChess)<br/>
 
'''Tucano''', (TucanoChess)<br/>
a [[Chess Engine Communication Protocol]], and since version '''8.0''', [[UCI]] compliant [[:Category:Open Source|open source engine]] under the [[Free Software Foundation#GPL|GPL]] written by [[Alcides Schulz]] in [[C]]. Tucano was first released in September 2012 <ref>[http://www.talkchess.com/forum/viewtopic.php?t=45066 Tucano chess engine 1.00] by [[Alcides Schulz]], [[CCC]], September 09, 2012</ref>.
+
a [[Chess Engine Communication Protocol]], and since version '''8.0''', [[UCI]] compliant [[:Category:Open Source|open source engine]] under the [[Free Software Foundation#GPL|GPL]] written by [[Alcides Schulz]] in [[C]]. Tucano was first released in September 2012 <ref>[http://www.talkchess.com/forum/viewtopic.php?t=45066 Tucano chess engine 1.00] by [[Alcides Schulz]], [[CCC]], September 09, 2012</ref>. As of version '''10''', released in January 2022, Tucano uses [[NNUE]] [[Evaluation|evaluation]] <ref>[https://www.talkchess.com/forum3/viewtopic.php?f=2&t=79172 ucano 10.00 release - Neural network evaluation] by [[Alcides Schulz]], [[CCC]], January 20, 2022</ref>.
  
 
=Description=
 
=Description=
Line 30: Line 30:
 
* [[Static Exchange Evaluation]]
 
* [[Static Exchange Evaluation]]
  
==Evaluation==
+
==[[Evaluation]]==
The [[Evaluation|evaluation]] features most common terms <ref>[http://www.talkchess.com/forum/viewtopic.php?t=47532 Eval development: is it better to tune or add new terms?] by [[Alcides Schulz]], [[CCC]], March 17, 2013</ref> with speculative calculation and aggregation of [[Opening|opening]] and [[Endgame|endgame]] [[Score|scores]]. The final score is [https://en.wikipedia.org/wiki/Interpolation interpolated] by [[Game Phases|game phase]] within a [[Tapered Eval|tapered evaluation]], with an additional [[Draw|draw]] adjustment in pawnless endgames.
+
 
 +
Starting with the version '''10''' release in Janauary 2022, Tucano applies [[NNUE]] [[Evaluation|evaluation]]. The network was trained on about 1.5 billion positions generated by Tucano with handcrafted evaluation at [[Depth|depth]] 8 using [[Hisayori Noda|Nodchip's]] [[Stockfish NNUE]] training code, along with probing code from [[Daniel Shawul|Daniel Shawul's]] [[Scorpio#ScorpioNNUE|Scorpio NNUE]] [[Scorpio#Bitbases|egbbdll]] library <ref>[https://github.com/alcides-schulz/Tucano/blob/master/README.md Tucano/README.md at master · alcides-schulz/Tucano · GitHub]</ref>. 
 +
 
 +
* [[NNUE]] since '''10.00'''
 +
 
 +
Prior to Tucano 10, the handcrafted [[Evaluation|evaluation]] featured most common terms <ref>[http://www.talkchess.com/forum/viewtopic.php?t=47532 Eval development: is it better to tune or add new terms?] by [[Alcides Schulz]], [[CCC]], March 17, 2013</ref> with speculative calculation and aggregation of [[Opening|opening]] and [[Endgame|endgame]] [[Score|scores]]. The final score is [https://en.wikipedia.org/wiki/Interpolation interpolated] by [[Game Phases|game phase]] within a [[Tapered Eval|tapered evaluation]], with an additional [[Draw|draw]] adjustment in pawnless endgames.
  
 
* [[Tapered Eval]]
 
* [[Tapered Eval]]
Line 70: Line 75:
 
==2020 ...==
 
==2020 ...==
 
* [http://www.talkchess.com/forum3/viewtopic.php?f=2&t=75317 Tucano 9.00] by [[Alcides Schulz]], [[CCC]], October 07, 2020
 
* [http://www.talkchess.com/forum3/viewtopic.php?f=2&t=75317 Tucano 9.00] by [[Alcides Schulz]], [[CCC]], October 07, 2020
 +
* [https://www.talkchess.com/forum3/viewtopic.php?f=2&t=79172 ucano 10.00 release - Neural network evaluation] by [[Alcides Schulz]], [[CCC]], January 20, 2022
  
 
=External Links=
 
=External Links=
Line 75: Line 81:
 
* [https://github.com/alcides-schulz/Tucano GitHub - alcides-schulz/Tucano: Tucano Chess Engine]
 
* [https://github.com/alcides-schulz/Tucano GitHub - alcides-schulz/Tucano: Tucano Chess Engine]
 
* [https://sites.google.com/site/tucanochess/ TucanoChess]
 
* [https://sites.google.com/site/tucanochess/ TucanoChess]
* [http://ccrl.chessdom.com/ccrl/4040/cgi/compare_engines.cgi?family=Tucano&print=Rating+list&print=Results+table&print=LOS+table&print=Ponder+hit+table&print=Eval+difference+table&print=Comopp+gamenum+table&print=Overlap+table&print=Score+with+common+opponents Tucano] in [[CCRL|CCRL 40/40]]
+
* [http://ccrl.chessdom.com/ccrl/4040/cgi/compare_engines.cgi?family=Tucano&print=Rating+list&print=Results+table&print=LOS+table&print=Ponder+hit+table&print=Eval+difference+table&print=Comopp+gamenum+table&print=Overlap+table&print=Score+with+common+opponents Tucano] in [[CCRL|CCRL 40/15]]
 
* [http://www.chessdom.com/tucano-the-brazilian-chess-engine/ Tucano, the Brazilian chess engine], [[Chessdom]], April 21, 2018 » [[TCEC Season 12]]
 
* [http://www.chessdom.com/tucano-the-brazilian-chess-engine/ Tucano, the Brazilian chess engine], [[Chessdom]], April 21, 2018 » [[TCEC Season 12]]
 
==Misc==
 
==Misc==
Line 90: Line 96:
 
[[Category:WinBoard]]
 
[[Category:WinBoard]]
 
[[Category:XBoard]]
 
[[Category:XBoard]]
 +
[[Category:UCI]]
 +
[[Category:Linux]]
 +
[[Category:Windows]]
 +
[[Category:PC]]
 +
[[Category:X64]]
 +
[[Category:NNUE]]
 
[[Category:Bird]]
 
[[Category:Bird]]
 
[[Category:Tania Maria]]
 
[[Category:Tania Maria]]

Revision as of 11:36, 18 March 2022

Home * Engines * Tucano

Keel-billed Toucan [1]

Tucano, (TucanoChess)
a Chess Engine Communication Protocol, and since version 8.0, UCI compliant open source engine under the GPL written by Alcides Schulz in C. Tucano was first released in September 2012 [2]. As of version 10, released in January 2022, Tucano uses NNUE evaluation [3].

Description

Bitboard Infrastructure

Tucano's bitboard infrastructure favors memory versus computation, with the option to gain some linear speedup if that becomes a bottleneck, i.e. using specific processor instructions. However, on recent Intel processors with huge caches, this might be rather insignificant concerning the strength of the engine. Tucano uses magic bitboards to determine sliding piece attacks.

Search

Tucano's search applies PVS alpha-beta with transposition table inside an iterative deepening framework without aspiration windows. Since version 7.00, Tucano performs Lazy SMP using a shared hash table [4].

Evaluation

Starting with the version 10 release in Janauary 2022, Tucano applies NNUE evaluation. The network was trained on about 1.5 billion positions generated by Tucano with handcrafted evaluation at depth 8 using Nodchip's Stockfish NNUE training code, along with probing code from Daniel Shawul's Scorpio NNUE egbbdll library [5].

Prior to Tucano 10, the handcrafted evaluation featured most common terms [6] with speculative calculation and aggregation of opening and endgame scores. The final score is interpolated by game phase within a tapered evaluation, with an additional draw adjustment in pawnless endgames.

Misc

See also

Forum Posts

2012 ...

2015 ...

2020 ...

External Links

Chess Engine

Misc

References

Up one Level