FRITZ 20: The Engine Sitting in Three Chairs — Coach, Opponent and Ally of the Chess World

One June evening, in a chess club tucked deep in a small alley in Jayanagar, Bangalore, I sat beside a fifteen-year-old boy. In front of him was a chessboard; beside it, an old laptop whose screen glowed with lines of notation that an outsider would not understand. He placed his finger on a knight, hesitated, then swallowed — that tiny sound is something I still remember. On the screen, the engine answered with a move he had never considered. He sat still for three seconds, then laughed. It was not the laugh of a loser. It was the laugh of someone who had just seen a door open. That night I went home and kept thinking about an old question: when did the machine become the teacher? And what happens to children who learn chess without a coach, without an academy, without a grandstand? The home board has no grandstand, but every time a piece is pushed, a whole sky of memory rushes back. I have written that sentence in many documentary drafts, and each time it feels a little truer. The news in Bangalore this season is not a player transfer but a release from ChessBase: FRITZ 20 has arrived, introducing itself in three short lines — your personal chess trainer, your toughest opponent, your strongest ally. One engine standing in all three roles at once. FRITZ is not a new name. In 2026, Dutch engineer Frans Morsch and Mathias Feist brought the first Fritz to market through the publisher ChessBase. More than three decades later, that product line is still standing in the world of chess engines. In an era when open-source Stockfish is free, Leela Chess Zero runs on neural networks, and Komodo Dragon is optimized for personal computers, a commercial engine surviving at all is a more interesting story than any Elo number. The reason is not computational strength. At the top tier, the gap between leading engines has almost no meaning for human players. What FRITZ sells is something else: a training environment designed for people, not for machines. The history of the Fritz line is bound to the “man versus machine” era. In November 2026, in Bonn, Vladimir Kramnik — then world champion — lost to Deep Fritz 2-4 over six games. That same year, in New York, Garry Kasparov also fell to Deep Fritz by the same score. Those two results closed a decade in which the public believed the end of human chess had arrived. What happened next was the opposite. The number of young players worldwide soared, middlegame quality among teenagers reached unprecedented levels, and names such as Arjun Erigaisi, Rameshbabu Praggnanandhaa and Dommaraju Gukesh emerged in India — a country that two decades earlier had almost no professional training system. In India, this context is especially striking. In December 2026, Dommaraju Gukesh became the youngest world champion in history at eighteen, beating Ding Liren in Singapore. Before that, Rameshbabu Praggnanandhaa reached the 2026 World Cup final and Arjun Erigaisi climbed into the top group of the world ranking. A whole Indian generation grew up alongside the spread of chess engines at home, and based on my experience following matches, that is no coincidence. FRITZ 20 pushes that logic one step further. The difference is not how much stronger the engine is, but how it breaks the training process into pieces. A club player can set a level that matches their own strength, letting the machine make human-like errors, letting it think more slowly and leaving room for human judgment. It is a small technical change but a large psychological one. FRITZ 20 takes on three roles at once, and each demands a different way of using it. As a trainer, it dissects every move after each game, ranks the severity of mistakes and points out missed moves. For a beginner, that is the fastest way to learn reflexes that has ever existed. For a tournament player, it is a tool for checking what the coach said in the morning session. As an opponent, the engine at its strongest level is not a good sparring partner — it is a wall. What the Fritz line manages is simulating skill levels, playing fast, playing slow, giving odds, and more importantly: playing in a human style, with moves that are imperfect but intentional. As an ally, the technical side that is rarely discussed comes into view. A modern engine ships with seven-piece Syzygy endgame tablebases, continuously updated opening books, and neural networks trained on millions of top-level games. NNUE — neural network evaluation — became standard several years ago, and the Fritz line is not outside that game. It is worth recalling a noisy chapter to see how this industry operates. In February 2026, ChessBase released Fat Fritz 2 and the open-source community pointed out that it was based on Stockfish code without fully complying with the GPL license. The dispute lasted months, and it showed one simple thing: in modern chess, computational strength is no longer anyone's private asset. What creates a difference is user experience, packaging, and teaching. In terms of features, FRITZ 20 inherits the toolkit this line has built over generations: tactical exercise libraries, opening training along variation trees, endgame practice with databases, cloud analysis, and playing modes at many levels. The new part is how these pieces are connected into a path. Users no longer have to choose their own exercises; the software suggests based on their own repeated mistakes. That is where algorithm and pedagogy meet. And that is exactly where I want to pause a little longer. I once thought a stronger engine would produce better players. I believed that for years, until I sat through enough training sessions in India and in China to see the opposite. Engines do not produce better players. Engines produce more similar players. A boy in Chennai and a boy in Hangzhou, four thousand kilometers apart, speaking different languages, reading different books, now open the same variation, choose the same system, and reach move fifteen in the same position. That uniformity has no precedent in chess history. It is effective. It is also worrying. The blind spot lies elsewhere, and it is harder to see. When every player has a strong engine in their machine, the implicit assumption of the coaching world becomes: the student already has the tool, our job is to teach theory. But in places without grandstands — the villages of Manipur, the roadside clubs of Kolkata, the classes opened in cultural halls in Kerala — what is missing is not opening theory. What is missing is the ability to sit still for four hours without getting bored. An engine cannot teach that. It may even undermine it, because it is always ready to give an answer within half a second, and there is always a new game to start. I remember an older coach in Bangalore telling me, between two rounds of a junior event: “Kids today calculate faster than I did at twenty. But they don't know how to sit still.” He said it and laughed, the laugh of a man who had taught chess for thirty years and had no intention of arguing with the times. In India, and I believe in China too, there is a paradox the Western chess world rarely mentions: the engine arrived before the coach. In many towns, a twelve-year-old learns the first move from software before meeting any teacher at all. In Europe the process runs the other way — the teacher comes first, the machine later. Those two paths produce two kinds of players, and I am not sure which is better. The spread of engines brings another consequence few want to discuss: cheating. In online chess, the same tool that makes a player better can also let them win games they do not deserve to win. That is why the training story and the integrity story in chess can no longer be separated, and a good teaching program in the future will have to answer both. What I am sure of is this: a dense tournament calendar is the real enemy of a young player, not a lack of training tools. A child plays a weekend event, competes online every night, adds a few rapid and blitz tournaments during the month, then goes home and sits another three hours in front of an engine screen — no software can save that kind of exhaustion. No update can compensate for a sleep-deprived brain. So is FRITZ 20 the revolution its press release claims? I do not think “revolution” is the right word. The revolution happened long ago, around the time Kramnik and Kasparov fell to Deep Fritz in 2026. What has been happening since is a gradual refinement, and FRITZ 20 sits inside that current: a better tool, more cleverly packaged, friendlier to ordinary users. What is notable is who is being served. For years, chess software was designed for strong players. Beginners opened the program and were frightened away. FRITZ 20 reverses that focus — its three introductory lines mention no computational strength, but training, personalization and companionship. Technically, this is the right direction. Culturally, it raises a question I do not yet have an answer to. The home board has no grandstand, but every time a piece is pushed, a whole sky of memory rushes back. I have written this sentence for the third time in my life, and this time it means something different. Before, it spoke of scarcity. Now it speaks of something else: when every board has a machine standing behind it, what remains to tell one person from another? Perhaps it is the capacity to endure silence. Perhaps it is the one thing no engine can evaluate: daring to play a move you know is wrong, just to see what happens. The boy in Jayanagar laughed that night after the engine answered. I do not know what kind of player he will become, or whether he will become one at all. But I know what I want to see in ten years: not children who play more like the machine, but children who know when to switch it off.