top of page

Dragon 3.2 by Komodo Chess - released

Considered by many to be the most versatile chess engine.


Here's what GM Larry Kaufman wrote about Dragon 3.2 by Komodo Chess:


We have released Dragon 3.2 at, an upgrade of the engine that won the World Computer Chess Championship in July 2022 after a tiebreak match with Lc0. Dragon uses NNUE (Neural Network Updated Efficiently) technology, originally developed for the game of shogi. Komodo has a great deal of chess knowledge in its evaluation. Training an NNUE network based on this evaluation was both an advantage and a challenge, requiring experimentation with architectures and data generation of billions of positions. The reinforcement learning phase for Dragon has already made great strides.

Dragon 3.2 is a huge strength improvement over Dragon 2, roughly 100 elo at blitz using normal openings, much more in MCTS mode or when playing Fischerandom (chess960). It is a significant improvement over Dragon 3, about 20 elo in all modes in blitz with normal openings. The gains over Komodo 14.1, the last pre-dragon release (Nov 2020), are in the 300 to 400 Elo range on one thread, depending on mode and game type, all at CCRL blitz time control (2' + 1"). The improvement over Dragon 3 is due to a newer net architecture, newer net, to speedups, and to search improvements and parameter tuning. The improvement from Komodo to Dragon is due to the embedded neural network providing much more accurate evaluation and also in some sense gaining an extra ply or so of search by seeing some tactics that a normal eval would not see. Our testing also shows that Dragon is especially strong in Fischerandom (960) chess. Dragon also won the 2021 TCEC Swiss and Swiss2 championships (normal chess). We believe that Dragon will play in a more human-like style than standard Komodo since it relies on learning what actually works in games rather than just on pre-assigned values for eval terms.

The Skill levels have been replaced by Elo values, made much more human-like, and are now intended to correspond to human Fide Rapid ratings or Rapid ratings, so for example an 1800 FIDE rated player should be closely matched with UCI Elo 1800 at 15' + 10". In the grandmaster range this was confirmed by test games between Elo = 2600 setting and Grandmaster Alex Lenderman. Also the MultiPV play of standard mode is dramatically improved from the initial dragon release, on top of the actual strength gain. Other features added since Komodo 13 include Armageddon mode, MCTS Optimism, Personalities, and Auto-Skill. Standard mode is usually the stronger mode objectively, while MCTS mode is usually better at setting problems for human opponents as it does not assume perfect defense, especially with multiple threads, and is also stronger than standard mode when looking at several lines of play at once (MultiPV).

Dragon 3.2 is about ten elo stronger in all modes than 3.1, and has scaled down the evals so that an eval of 1.00 will mean that the position is thought to be on the edge between a win and a draw.


In addition, on the TalkChess forum page, we can get some additional information from the same Dragon chess engine Developer:


(...) It is likely that 3.2 will be the last of the Dragon 3 series, and we don't know when

Dragon 4 (or whatever it might be named) might be available.


On online forums dedicated to computer chess, users speculate whether this is the end of the development of the Dragon by Komodo Chess engine, whether the degree of difficulty in raising the strength of the engine is so large and time-consuming that the Developers ending the subscription model want to give themselves more time, or perhaps they are simply carrying the intention of selling their engine to another developer ? Or maybe there are other reasons for this ?

Dragon by Komodo Chess is an extremely powerful chess engine. Its closest competitors, i.e. Berserk, Koivisto, RubiChess or Revenge engines, although consistently increasing in strength, in my opinion it needs at least 1 year to match Dragon 3 by Komodo Chess in strength.

Dragon 3 by Komodo Chess is a sure No. 2 on the MCERL rating list.

Source: MCERL (ongoing)

Undoubtedly, today, the development of a chess engine at the highest level - which Dragon by Komodo Chess presents - requires the creative work of a team of dedicated Developers and many immeasurable hardware resources.

This is the path chosen by Developers of free open source engines Lc0, Stockfish and several others. Entire communities are focused according to specific engines - developing them as a "never-ending" project. And you can see the tangible results of this.

I have this feeling, although I could be wrong, that in the case of Dragon by Komodo Chess, in time it could be integrated into another chess and commercial project run by one of the biggest and recognizable brands dealing with the chess topic.

What do you think about it? The comments section is at your disposal.


466 views2 comments

Related Posts

bottom of page