Tucano

From Chessprogramming wiki
Jump to: navigation, search

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