Onimusha: Way of the Sword and Capcom's Twenty-Year Crack

In the first month of the launch window, one claim travelled further than every other number: the campaign of Onimusha: Way of the Sword is longer than Resident Evil Requiem and Pragmata combined. No hour figures were supplied for the other two titles. No methodology was cited. No timeframe was given. Only a comparative sentence, placed mid-article, repeated often enough to take on the shape of a verified fact. I sat with my notes on the third day of following the wave of information around this title. Twenty-one years of watching this industry has given me a reflex that makes me unpopular with press officers: when a publisher weighs its own product against the other products in its own slate for the same year, that is not a promotional campaign. That is a revenue civil war being presented as a press release. And in this case, the crack is not in the 36 boss fights. It is in the fact that nobody asked how long the other two games are. I have followed the Japanese games industry since before it became a topic for financial bulletins. In 2026, when Onimusha: Warlords launched on PlayStation 2, I was still organising small tournaments, and I remember the feeling: a Japanese historical action game, published by Capcom, selling over a million copies within months. That was a moment when a new brand could still define an entire product line. Onimusha 2: Samurai's Destiny followed in 2026. Onimusha 3: Demon Siege arrived in 2026. Onimusha: Dawn of Dreams closed the run in 2026. Then silence. Twenty years. Capcom did not disappear during those two decades — quite the opposite, they became one of the most disciplined publishers in the business when it came to portfolio management. But Onimusha lay dormant. In 2026, a remaster of Onimusha: Warlords was released, and the way it was handled said a great deal: a preservation product, not a revival product. No sequel was announced afterwards. No roadmap. Just a library being reintroduced to a new generation of players, most of whom had never held a controller through a single fight in this series. By 2026, Capcom placed three major titles in a single year: Onimusha: Way of the Sword, Resident Evil Requiem, and Pragmata. Three standalone products, three brands that differ in kind, competing for one wallet. In industry history, a publisher releasing three major titles in one year is not unprecedented — but it is always worth analysing, because it reveals the publisher's confidence in the market's carrying capacity, and at the same time reveals how dependent they are on their own slate. Onimusha: Way of the Sword is set in Edo-period Japan, with Musashi as its protagonist — Miyamoto Musashi, the legendary swordsman, a name familiar to anyone who has read anything about Japanese martial culture. Not a fictional warrior, not an anonymous hero, but a historical figure with a body of writing behind him, a school of swordsmanship, and a legend. That choice is not accidental. A historical figure already fixed in collective memory dramatically reduces the storytelling cost for a developer, and simultaneously raises expectations by a full tier. That is the context. But context is not where I want to stop. The content structure of this title, according to the information currently circulating, is built around four numbers. On the difficulty setting called Action — the baseline the source article uses for measurement — the main campaign runs thirty to forty hours. For players pursuing full completion, the figure exceeds fifty hours. The main story alone is said to sit around the thirty-hour mark. And the entire game contains 36 boss encounters. Four numbers, three of which do not agree with each other. This is the first point I want to pause on, because it is the kind of detail almost no commentator cares about, and that is precisely why it matters. When an article says the main story "could take thirty hours" and elsewhere says the campaign takes "thirty to forty hours on Action difficulty", there is no fatal contradiction — a midpoint does overlap a range. But there is methodological carelessness. That thirty-hour figure could be the lower bound of the range, an independent estimate from a different source, or simply a number written to fill a gap. There is no way to tell. In my profession, a number without a method is a number that does not exist. It is not wrong — it simply has not been born yet. And this is where the difficulty setting deserves proper treatment as a length multiplier, because it is the calculation the games press almost always skips. Same game, same content volume, but a player fluent in action games finishes in thirty to forty hours, while a player who is not needs more than fifty. The gap between those two endpoints lands around sixty-seven percent. No patch, no balance adjustment, no expansion pack produced that gap. Only player skill. This sounds obvious, but it has direct consequences for how length figures get used in marketing. When a publisher says "the campaign is thirty to forty hours", they are talking about good players. When they say "over fifty hours", they are talking about weaker players, or completionists. The two figures are not in the same unit. In headlines, they sit side by side as if measured with one ruler. I remember a comparable case: when Devil May Cry 5 launched in 2026, length estimates circulating in the community ranged from ten to fifteen hours for the main campaign, depending on whether the player replayed with different characters. People saying "twelve hours" and people saying "thirty hours" were not in conflict — they were playing two different games inside the same box. This is not a media failure. It is a structural feature of linear action design, where playtime is inversely proportional to skill. So why do I still mark it as a crack? Because of density. 36 boss fights across a roughly thirty-hour campaign means, on average, a boss encounter every fifty minutes. Put that number beside others in the genre. In most linear action titles with chapter structures, boss counts typically sit between eight and twenty for a campaign of similar length. Thirty-six sits far outside that band. It is not wrong — it simply points to an extreme design choice. There are two ways to read that choice, and each leads somewhere different. The first reading is positive: this is a title that makes combat the absolute centre, where rhythm is built from a chain of climaxes rather than conventional peaks and troughs. If every boss has its own moveset, its own visual language, its own approach, then this density creates a feeling of continuous feeding — something a segment of action players deeply enjoys. The second reading is a warning: 36 bosses can mean reused templates, reshaped silhouettes with unchanged attack structures, or encounters designed as obstacles rather than climaxes. In the industry this phenomenon has a name: filling space with density. Players detect it quickly, usually from the twelfth encounter onward, and once they detect it their language shifts from "long" to "tiring". I do not have the data to say which reading is correct. But I have an observation about how this information is being transmitted: nearly all of the surrounding coverage treats the number thirty-six as a pure value assertion — more is better. Meanwhile a boss fight is the most expensive design unit in this genre. It requires motion design, staging, balancing, testing. Multiply it by thirty-six and you have a production problem that anyone who has worked in the industry knows is extraordinarily hard to keep uniform in quality. This is the place to state a principle I have followed my whole career: do not ask what position a player occupies, ask what position he is disguised as. Applied here: do not ask what genre this game belongs to, ask what genre it is dressed as in order to sell. And it is dressed as an action game with narrative depth. But its internal structure — by its own description — is an upgrade economy. The protagonist's sword and clothing are upgraded with materials: iron and leather. Where do those materials come from? Side quests and a bow challenge called Ace Archer. In other words, the optional content layer is not purely decorative. It is a power gate. In action game design there is a large difference between two kinds of optional content. The first is reward content: you play out of curiosity, for story, for achievements, and the payoff is aesthetic or honorary. The second is gate content: you are compelled to play in order to be strong enough for what lies ahead. When upgrade materials for weapons and clothing are locked behind optional content, that optional layer has shifted from reward to gate. The consequence is not small. Players who go straight down the main story — the largest cohort, especially among newcomers — will reach the later boss fights under-geared. They will hit a difficulty wall that is not about skill but about economy. And when they go to forums to ask, the answer they receive will be the classic action-community response: "just go do the side quests". That answer is mechanically correct and a design failure. The source article estimates the optional layer at ten to fifteen hours. Combine that with 36 boss fights and a roughly thirty-hour main campaign. If a player completes both layers, total playtime lands between forty and fifty hours. That is a substantial volume, and in terms of value per hour it sits very well in the premium single-player segment. But one detail changes the whole picture: the platinum trophy requires finishing the game at least twice. This deserves close analysis, because it is the kind of detail that gets overlooked while carrying structural force. A two-playthrough requirement is not rare in Japanese action game tradition. It comes from an era when replaying — New Game Plus, higher difficulties, weapons unlocked on completion — was part of the design, not part of the achievement list. But when it is placed inside a platform achievement framework, its nature changes: from an invitation into a condition. For completionists, this doubles the time investment. For players without much time, it becomes a psychological barrier before they even start. And for the developer, it is a way to extend product longevity without adding new content — an entirely rational economic strategy, and an entirely fragile one in terms of perception. I have seen this reaction many times over twenty-one years. It does not explode. It seeps. It shows up in reviews after a month, in discussion threads after three, and in completion rate charts after six. And when it shows up, it always has the same shape: players are not angry about replaying. They are angry that they did not know in advance. That is why I have always argued that achievement structure information should sit alongside length information, not at the bottom of an article as a footnote. And this is where the largest crack appears. After all, the entire media story around this title is built on a single idea: content volume is value. Longer is better. More bosses is more impressive. More hours is more worth the money. And to prove that idea, the game is placed beside two other products from the same publisher, with an unquantified claim attached. I do not believe unquantified claims. Not because I distrust publishers. Because I have seen too many things measured by feeling collapse the moment people start counting. Recall the structure of one of the most famous media failures I ever analysed. In 2026, before the World Cup, I made a prediction that Germany would exit in the group stage. My basis was not a feeling. It was that four of their six defenders were over thirty, and that they averaged only 1.1 shots per match from runners in behind the defensive line. Germany lost 0-2 to South Korea in their final match, generating 0.4 xG from thirteen attempts, every one of them a long-range effort from outside the box. The number I published before the tournament and the number that appeared after it almost matched exactly. That is the difference between a claim and a measurement. In the Onimusha case, the claim "longer than the other two combined" cannot be verified by anyone at this moment, because there are no hour figures for the other two. It has the shape of a measurement but not the content of one. And the most interesting thing about claims of this type is that they usually do not need to be true. They only need to be repeated often enough during the launch window, when purchasing happens reflexively, and when verification takes at least thirty hours of play. That is a structural asymmetry between seller and buyer, and it has a name: the launch window. The crack always appears before the collapse, it is just that people prefer the sound of the collapse. Now place this title inside the publisher's larger picture. Capcom has three major titles in 2026. That is a strategic decision readable in two directions. The first is confidence: the publisher believes the market is wide enough to absorb three premium products in one year, that players will buy more than one, that each brand has a sufficiently large and non-overlapping audience. The second is portfolio defence: when a publisher releases multiple major titles at once, sometimes it is reacting to lengthening development cycles and rising production costs, which pile projects up and force them to launch close together because further delay is impossible. Both readings are plausible. But one variable cannot be ignored: audience overlap. Resident Evil Requiem and Onimusha are not fully separated brands demographically. Both target mature action-game players, both are Japanese brands, both are premium single-player products. For a meaningful share of players, these are not two different choices. They are one choice inside an annual entertainment budget. And Pragmata is the least defined variable. It is a new title, with no inherited audience, no brand history to lean on. Under normal conditions, a new title in a three-product slate is the highest-risk item. With two established brands standing beside it, it is the most likely to be overshadowed. This is why I argue the claim "longer than the other two combined" is not merely advertising. It is a defensive differentiation move, designed to pull Onimusha out of internal competition using an attribute the other two lack: content volume. When you cannot win on brand recognition — because your brand has been dormant for twenty years — you must find another axis on which to claim superiority. And that axis has been chosen: hours. I recognise this pattern because I have written about it many times in the transfer market. Behind every contract is a silent brain screaming. Nobody announces that, but every transfer release is written from some position of weakness. A club signs a player because it fears missing out on European qualification. A player chooses a new destination because he fears being overshadowed at his old club. Real motives always sit in what is left unsaid. And here, the real motive is not in the 36 boss fights. It is that this title needs a reason to exist in a year when its publisher has two other products needing the same wallet. There is another dimension I have been tracking, and it concerns how the industry observes itself. The source analysis I read carried a domain label of esports. That label is entirely wrong on first reading. There are no teams, no tournaments, no group stages, no rosters, no prize pools, no seasons, no competitive patch cycle. This is a single-player action game. The mislabelling is not harmless. It is symptomatic. Over the past decade, the content ecosystem around video games has developed a reflex: anything related to games has the potential to become esports, therefore anything deserves to be processed as esports. That reflex generated an enormous content industry, but it also generated classification colour-blindness. And when the classification system is colour-blind, the data gets contaminated. Articles about campaign length get mixed with articles about pick rates. Information about upgrade materials sits beside information about win rates. That is a crack in the information infrastructure, and it is worth recording, because it will continue. Back to the product itself. There is a question I kept asking while reading the information about this title: what will make a player feel satisfied after finishing? In a thirty-to-fifty-hour action game, the emotional reward structure does not live in killing the final boss. It lives in whether the memory of the journey stays vivid enough that the player wants to retell it. And memory of a journey is made of variety, not density. This is the point I want to make clear, because it runs against ordinary intuition. In sport, high density is usually taken as a sign of quality. A tournament with many great matches is a good tournament. A season with many title contenders is a compelling season. But in experience design, high density does not automatically create memory. It creates rhythm. And rhythm only becomes memory when there is contrast. Thirty-six bosses across thirty hours has no contrast. It has even rhythm. Even rhythm is wonderful for a short game. It is dangerous for a long one. I have seen this pattern across many fields. It shows up in seasons where every team is roughly equally strong and nobody remembers who won. It shows up in sporting events where every match finishes narrowly and spectators cannot recall a single one. Uniformity at high density produces uniformity in memory, and uniformity is the enemy of legacy. This does not mean the game will fail. It means the quality of each individual boss encounter will decide everything. If ten of the thirty-six are once-in-a-generation fights, players will forgive the other twenty-six. If none are, then the number thirty-six becomes prosecutorial evidence rather than exculpatory evidence. Now let me address what I consider the most underrated element in this entire story: cultural setting. The special story quests, called Curious Cases, are described as introducing players to the myths and legends surrounding Eastern Kyoto during the Edo period. This is a detail I want to dwell on far longer than mass coverage has. In the history of this series, Japanese cultural material was never mere backdrop. Onimusha: Warlords in 2026 built its world on the Sengoku period, placing historical figures inside a supernatural story. That formula was close enough to Japanese audiences to guarantee interest, and strange enough to international audiences to generate curiosity. It worked throughout the 2000s. The shift from Sengoku to Edo, and from military battles to local legends, is a deliberate move in scale. It narrows the geographic scope and widens the cultural scope. That is an intelligent artistic choice, and also a safe production choice: a tighter setting means fewer visual assets. But it also raises a question I have not seen anyone ask. When you build a game on a local cultural foundation, you are targeting two audiences with opposite expectations. Domestic audiences want accuracy and depth. International audiences want accessibility and surface impact. No design satisfies both perfectly. Choosing Miyamoto Musashi as protagonist is one way of handling this. Musashi is one of the few Japanese historical figures whose name is known internationally, through popular culture and literature. But to domestic audiences, Musashi is a figure who has been portrayed hundreds of times in hundreds of works. Expectations around him are so high they are nearly impossible to meet. That is a calculated gamble, and I think it is the right one. But it is also a potential crack: if the protagonist is not portrayed in a way that escapes the shadow of his own legend, no length figure will rescue the perception. Now I want to address what I consider most important in this whole analysis, and it is not about any specific title. The real match only begins when the whistle blows and the analysis room turns on the lights. In sport, people judge a player by what he does on the pitch, but they build that player's career out of what is said about him after the match ends. This is a truth the media understands better than anyone, and admits less often than anyone. The same performance, told two different ways, produces two different careers. In games, the pattern works identically, except the cycle is shorter and the intensity higher. The launch window for a premium title runs roughly two to four weeks. In that span, every piece of information about the product is produced and consumed at maximum speed. Then it sinks. And what remains — in community memory, in platform data, in long-tail reviews — is what was built during that window. That is why I always write during periods without matches. Not because I enjoy the quiet. Because quiet is the only condition in which structure becomes visible. And the structure here is a pattern I have seen many times before. A publisher owns a dormant brand. That brand carries emotional value for a generation of players who have now grown up and have disposable income. The publisher decides to revive it. But they do not revive it as a brand with a stable audience. They revive it as a new product that must rebuild its audience from zero, because the younger generation never knew it. In that situation, the marketing message cannot be "remember". It must be "notice". And to achieve that, the product needs an attribute that is measurable and comparable. Content volume is the perfect attribute for that purpose, because it is easy to understand, easy to compare, and requires no background knowledge from the buyer. You do not need to know what Onimusha is to understand that a game with 36 boss fights is a big game. That is the power of the number. And also its weakness. I hold a relatively strong belief about what will happen with this title, and I want to present it in a verifiable form rather than as a feeling. First, the basis for judging this game's success will not be initial reviews. Early reviews are written by skilled players, usually on the highest difficulty, who will finish in twenty-five to thirty-five hours. Their scores will be high. That is close to certain. Second, the real test will occur between the fourth and twelfth week after launch. In that window, the mainstream player layer will reach the later content, and they will face two things at once: a difficulty wall built out of upgrade economy, and a second-playthrough requirement for full completion. Third, the early warning signal I will be watching is not the score. It is the completion rate. If the share of players finishing the main campaign — not full completion, just the main story — drops below the genre average, that is a sign the content volume has exceeded tolerance, and the "more is better" message has betrayed itself. Fourth, the comparative claim against the other two titles will be automatically verified when figures for those titles appear. If the gap is not large enough to justify "combined", this will become one of the cleanest examples of volume marketing overshooting volume reality. And it will be remembered not as a game's fault, but as a fault in how the industry talks about games. Fifth, and this is the part that interests me most as a long-term observer: if this title succeeds, it will not succeed because it has 36 bosses. It will succeed because of Edo-period Kyoto, because of local legends, because of a historical swordsman placed at the centre of a supernatural story. Meaning it will succeed for reasons nobody used to advertise it. Every surprise on the pitch is an appointment we arrived late to. And in this case, the appointment was booked twenty years ago. What I want to leave behind is not a prediction of success or failure. I do not have enough data for that, and anyone who claims otherwise is selling you something else. What I want to leave behind is an observation about how this industry processes information about itself. We are in a period where content volume has become the common currency of every argument about value. More hours, more bosses, more maps, more seasons, more matches. And in the process, we stopped asking a very old question: more, in order to do what. There is nothing wrong with a thirty-hour game. There is nothing wrong with a ten-hour game. What is wrong is judging them with the same ruler and calling that measurement analysis. Onimusha: Way of the Sword may be a superb game. It may be an exhausting one. In either case, what is remembered will not be how many hours it lasted, but how long it stayed in the player's head after they put the controller down. And if I had to place a single bet, I would place it here: players will remember Edo-period Kyoto longer than they remember the number thirty-six. The twenty-year crack has closed. The question now is what is inside it — a revived legacy, or a void filled with hours. We will not learn the answer on launch day. We will learn it on the day nobody writes about it any more.