Komodo Dragon 2.6.1 - Review
This time we will take a look at one of the TOP chess engines.
According to the Creators: a lot of improvements, some new features, development of a "human" style of play.
Is indeed the new Dragon By Komodo Chess worth interest, is the MCTS algorithm a useful alternative to engines using standard search functions?
I will try to answer these and other questions in this review.
Source image by ChessBase
By way of introduction, I'm pasting information from one of the Komodo Dragon GM Larry Kaufman, who unveiled the latest version of this engine in December 2021.
[December 20, 2021] We have released Dragon 2.6 at komodochess.com. 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 is in its infancy, but is already showing great promise.
Dragon 2.6 is a huge strength improvement over Dragon 2, roughly 70 elo at blitz on four threads and 83 elo on one, more than 100 elo in MCTS mode and for playing Fischerandom (chess960). It is a significant improvement over Dragon 2.5, about 18 elo on one thread blitz, about 25 elo in MCTS mode or playing chess960. The gains over Komodo 14.1, the last pre-dragon release (Nov 2020), are in the 250 to 350 Elo range depending on threads, mode, and game type, all at CCRL blitz time control (2' + 1"). The improvement over Dragon 2 is due to a larger, "smarter", and better-trained net, to deeper search due to search enhancements, and to 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 can not see. Our testing also shows that Dragon is especially strong in Fischerandom (960) chess, which was confirmed when Dragon won the TCEC FRC 2021 championship, and again in our latest testing that shows Dragon 2.6 convincingly defeated Stockfish 14 in a 2000 game blitz Fischerandom match. 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 ratings in Rapid chess, so for example an 1800 FIDE rated player should be closely matched with UCI Elo 1800 at 15' + 10". 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 more than a few seconds to consider, and is also stronger than standard mode when looking at several lines of play at once (MultiPV).
It sounds very interesting, doesn't it?
Table of Komodo Dragon engine specifications.
Before I get into specifics, a brief history of this chess engine.
Komodo Dragon (Dragon for short) is a commercial chess engine that is a direct successor to the Komodo engine.
Dragon has a long history and many different successes. It is a multiple-time Computer World Champion. Its development over the years is regular. Every few months the Developers release new updates and new versions of this engine.
Let's start with...
Power of play.
Dragon is a very powerful chess engine with an established reputation, which usually occupies the top spots in chess engine ranking lists.
Source: CCRL
Attached for download is a match consisting of 100 games of Dragon against the Rebel 14.2 engine. Each engine had 1 CPU at its service and 1 minute per game; Fritz 18 GUI, Perfect 2021 opening book.
The result of the match shows the colossal difference in playing strength between top engine Dragon (over 3500 Elo) and very strong engine Rebel (about 3200 Elo).
Out of 100 games:
Dragon won 81 and tied 19.
Rebel won 0 games and drew 19.
The image below shows one of the games in progress.
Style of play.
Dragon plays positional style chess by default. Of course, an engine that plays with a strength of about 3500 Elo will excel in positions that require a tactical approach. Undoubtedly, Dragon's style of play has not changed dramatically compared to its predecessor, and it still favors a more maneuverable game leading to an advantage rather than a frantic attack paid for by the sacrifice of pawns and pieces.
Invariably, the ability to select predefined personalities that allow the user to easily customize Dragon's playing style for their own needs is a delight.
Using three examples from my practice, I will show how Dragon can play.
For comparison, I'll contrast Dragon's suggestions with those of the super-powerful Stockfish engine.
The following examples are from my cor