• Total War: Three Kingdoms Review

    Total War: Three Kingdoms Review
    by zoner16

    Total War: Three Kingdoms brings the series to ancient China during events at the fall of the Later Han dynasty at the end of the second century AD. Taking control of one of the warlords, your task is to amass a legion of loyal and talented retainers, build a strong state through war, economics, and diplomacy, and reunite the land by declaring yourself emperor and destroying all pretenders.

    Three Kingdoms is the first mainline historical title since Total War: Attila, and the first full price historical title since Total War: Rome 2. The game features significant changes and additions, some ported over from Thrones of Britannia and the Warhammer games, some drawn up from Total Wars many years past, and others entirely new to the series. Given all of that, how does it hold up?

    The Setting

    The setting is the biggest addition this game brings to the table. There has never been a Total War title set in China, even as a theater or subregion. It’s been an oft requested destination, and the Fall of Han-Three Kingdoms era is far and away the most popular time period. By CA’s own admission, this was a game meant to break into the Chinese gaming market as well, which certainly helped it reach its current status as CA’s biggest launch ever.

    However, this era’s place as the setting of the next tentpole historical Total War is complicated by the near-legendary status of the events that take place within it. The Three Kingdoms era is much more than a historical conflict period. It’s also the setting of possibly the best-known work of Asian literature, the 14th century historical-fiction novel Romance of the Three Kingdoms (Sanguo Yanyi). The novel takes the events of the era recorded in the 3rd century historical account, Records of the Three Kingdoms (Sanguozhi), and dramatizes them, adding its own fictional stories, supernatural events, legendary feats of strength, and changes in characters to suit a narrative. It’s also spawned numerous adaptations and spinoffs throughout the years, including TV shows, movies, and video games.

    Within the game, it’s clear that the developers were first and foremost influenced by these later adaptations, with unit models being made in the image of the soldiers from the 2008 movie Red Cliff, various 2D artwork reminiscent of the 1994 Romance of the Three Kingdoms TV show, and much of the battle aesthetic is reminiscent of the 2010 TV show as well as the Dynasty Warriors game franchise. The foundation of the game is very much a Romance work seen through several centuries of reinterpretation. The Romance is ultimately a work of historical fiction that was evoking nostalgia for a well-known and celebrated historical era, so the game is still very historical in many ways.

    Romance vs Records Mode

    Three Kingdoms was billed as the next mainline historical Total War, so the presence of the “fantastical” Romance elements is likely to be offputting to some people who have grown tired of the dominance the Warhammer games have held over the Total War discourse for a couple years now. CA’s solution was to split the game into two modes, the “fantastical” Romance and the “historical” Records.

    It wouldn’t be correct to say that either directly follows the work that they’re named after. It would be more accurate to say that Records mode is mostly just curbing the biggest excesses of the Romance to make it into something that historical Total War players would feel more at home with. Generals go from Warhammer-esque single entity heroes to a heavy cavalry bodyguard unit, lose their special abilities, and can no longer duel. Additionally, stamina drains faster, recovers slower, and has more drastic effects. This usually causes battles to be somewhat longer, with equal, full-stack, open-field engagements going from 8-10 minutes to 12-15 minutes. Generally though, youÂ’ll find yourself using most of the same army compositions and tactics between the two modes.

    Beyond that, there’s very few differences between the two campaigns. Some changes to general mortality will keep them around for longer and allow you to be somewhat more aggressive with them. General equipment will also be somewhat more important due to the nature of duels. However, aside from a minor alteration to Dong Zhuo’s unique event chain, there’s no additions or changes to gameplay that will make Romance and Records feel that different at a campaign level.

    Faction Roster

    While the mode mostly feels like an aesthetic choice, faction assuredly isn’t. Despite only having Han Chinese culture factions, the game does an excellent job of differentiating them, even for people who aren’t familiar with the time period. Most factions have access to a unique mechanic in the vein of the Warhammer 2, and in most cases, it informs a good amount of that faction’s playstyle. Yuan Shu’s Legitimacy and Yuan Shao’s Lineage mechanics require that they keep a reliable network of diplomatic collaborators around even into late game, while Zheng Jiang’s Infamy and Dong Zhuo’s Intimidation mechanics force aggressive, high-risk militarism just to stay afloat.

    The faction select screen is about as detailed as you can ask for.

    Even for the factions whose mechanic is relatively lackluster, their start position, unique units, and specific dilemmas and events often make the strategy around them noticeably different. Sun Jian’s position in the quiet south allows him to prey on the passive Han Empire and colonize the unsettled wilderness while using the Yangtze as a natural defense. Meanwhile, Yuan Shu’s start in the center of the map ensures that he is constantly surrounded by rivals and lacks access to most commercial centers and major food production without taking at least a couple of them.

    Most of the unique units are just modified versions of existing units, so despite ostensibly having more unique units than Shogun 2, there’s nothing quite as unique as Otomo Donderbuss Cavalry or Hojo Hand Mortars. However, these units usually end up playing a much bigger role in differentiating faction tactics due to the way that they navigate around the restrictions on the recruitment system. Any general in a faction can recruit their faction specific unit after reaching a certain level, unlike most other units which require either technology, a specific general class, or both to recruit. Add on the fact that many of these units are much better than their generic analogues, and they can allow a faction to dominate a particular aspect of a battle for most of the campaign.

    Finally, there’s faction dilemmas and events. The dilemmas allow one to choose whether to follow the canon/historical events or deviate from it to change the pace of a campaign. Some involve choosing one alliance or war over another, but others may kill your faction leader to allow their heir to take control, change your diplomatic standing for a cost, or even confederate you with another faction. On one hand, these can sometimes feel arbitrary “but thou must…” sort of arrangements where a choice you were forced into but didn’t understand created a situation that doesn’t seem reasonable, but they can also allow you to throw a wrench into the campaign just when you feel that it’s getting a little too stale, bypassing certain restrictions to liven up your experience.

    Campaign Setup and Pacing

    The campaign follows much the same Total War structure as one would expect. You start with an army, usually some lands, and a couple of starting enemies to get you situated. The structure of the campaign after this is somewhat more flexible than it has been previously. Gone are specific province requirements, diplomatic goals, and wonder acquisition. Instead, a warlord progresses through the various faction ranks by acquiring and developing cities. While this progression unlocks more armies and the like as it did in Rome 2, it is also tied to the victory condition as it was in Shogun 2, as once one reaches the highest prestige rank, the realm divide begins.

    Unlike the previous Realm Divide, this does not cause the entire map to suddenly go to war with you. Instead, the three strongest factions claim the mantle of emperor, and their capital becomes their “imperial seat.” Winning the campaign is then tied to when a faction conquers all three imperial seats. While any faction who has claimed the three emperor titles is obviously on a collision course with each other, the remaining factions can still move around in between, either throwing their lot in with a specific emperor, sitting out of the conflict, or playing the dark horse, attempting to usurp one of the emperors by taking their seat and becoming a contender for the final victor.

    The combination of these two mechanics allows for a much freer experience from start to finish, more closely embracing the possibilities of the sandbox. There is an option to “Play Tall” with a limited number of well-developed commanderies earning you the prestige necessary to declare yourself emperor. Campaign strategy throughout early and midgame tends to be a lot more flexible, with wars and expansion being about more organic concerns than arbitrary victory provinces. Once endgame hits and the three-way brawl starts, you have a couple of very well-defined objectives in the emperor seats and its ultimately up to you on how you want to tackle them. Since said cities are the capitals of the next strongest factions at the time, they tend to vary from campaign to campaign, which helps the endgame remain interesting from campaign to campaign.

    The final blitz of the campaign can become frustrating if geography isn't kind to you. Thankfully, an AI emperor may just surrender if your power is convincing enough.

    Overall, the campaign pacing is probably the best its ever been in a Total War game. There’s still some slogging through uninteresting fights in certain scenarios, but it usually never gets to the point where you’re just taking territory just because it’s on a checklist, nor is the game insecure enough to just start throwing huge hordes of invaders at you from outside the map. Late game can still drag as imperial factions just begin throwing stacks of readily available conscript armies at each other.


    Diplomacy was perhaps the headlining feature for Total War fans prior to release. The system has never been particularly good in past titles, with it either being completely irrational or feeling incredibly pointless. This time, a host of new options and a Paradox-style numerical scoring system for deals help make it far more useful.

    Old favorites like Trade Territory and Issue Ultimatum make a return, but the real stars are the new Vassal interactions and the Coalition feature. The option to annex vassals down the line gives them much better long-term use, while Autonomy Guarantees can swing someone whose territory you may never need onto your side. Coalitions meanwhile, offer a way to create semi-stable multilateral alliances without being forced to join every war that the AI gets into, which has often been a problem of alliances. Refusing to come to a member’s aid or voting against membership decisions too much will likely end with your faction getting kicked out, but it does provide some much-needed flexibility. Military Alliances are now the evolution of a coalition, providing an ironclad guarantee of support, which also means that all members enter and peace out together. This prevents the awkward past situation where alliance members negotiated themselves out of a war and left their allies in without consequence or had no way of forcing an enemy to sign peace with their allies as well.

    The AI engages with all these new options too, mostly with success. As of this writing, the Yuan Shao vassal swarm has become something of a meme, as have the oddly shaped coalitions. By midgame, most of the map are connected in vast diplomatic networks that ensures that no war is against a single opponent. This, along with the expanded toolkit and greater willingness for the AI to go to peace, make midgame the heyday of diplomatic maneuvering as coalitions and alliances more fight for members than territory. It all has a tendency to change in an instant as well, a poor call to arms can cause vassals to rise up against their overlord, join coalitions or create their own, and then fight for the remainder, remaking the map in the span of a couple turns.

    Some diplomatic options are more balanced than others. Yuan Shu is willing to trade his daughter for a Clay Pig.

    Overall though, it’s the relative rationality of the diplomatic AI that makes it all work. The transparency of the numerical scoring system for deals means you can usually understand what’s working in your favor and what you need to work on. While it still has moments of baffling behavior, you usually won’t be stuck in situations where the AI completely refuses to make sense. It will mostly follow its opinion and threat modifiers, with a certain amount of its stated personality type as well. There are some factors which can be mystifying, such as the ever arcane “Balance of Power Shift” which can sometimes end up blocking even the most generous peace deals, and the AI’s valuing of “Strategic Threat” sometimes swings wildly, but the game’s telegraphing of these modifiers means that you’re usually aware of why particular things are happening, even if the AIs perception is debatable.

    Character Management

    If Diplomacy was the headliner for the Total War fans, the characters were the headliner for the Three Kingdoms fans. Both history and the novel are told in the perspective of the lives of the men and women who shaped the era, and their names and legacy are inseparable from the narrative. The marketing heavily focused on some of the most famous names, but the total number of named individuals in the historical records well exceeds a thousand. There are around 750 of these that are recruitable in the game, with procedurally generated generics buoying the gaps.

    Total War has always had a certain amount of character management and interaction. However, Three Kingdoms primarily focuses on its characters to far greater extent. Characters are the primary way that you perform the most important actions on campaign. Units are recruited into character retinues, diplomacy is informed by leader personality, and army and commandery buffs are controlled by the characters within them. Whether in Romance or Records mode, acquiring and developing good characters is perhaps the most important part of the game.

    In service of this new focus are the new satisfaction and relationship systems. Satisfaction is a measure of a character’s happiness in their faction. It’s controlled by several factors, including court rank, relationships with other characters, and their office. If a characterÂ’s satisfaction drops to 0, they’ll leave. If they’re a general or administrator, this will cause a rebellion, and if they’re the heir, prime minister, or on the council, this will cause a civil war. The issue with this is that the positions from where characters can cause the most damage are also the positions that give satisfaction buffs to the characters anyways. Having an administrator defect is so rare since just being an administrator will placate almost anyone. The relationship system is similarly a measure of who gets along with who. Characters form relationships by being around others who have traits they like or dislike, through events and battle, and through familial connections. Characters who like each other tend to perform better when fighting alongside each other, enjoy being in the faction more, and generally make your life easier, while the opposite is true of characters who dislike each other. While it’s not necessary to micromanage relationships, it can be a somewhat enjoyable metagame to foster friendships and watch cliques form around individuals. In my experience, the system isn’t as dramatic as CA probably hoped it would be, as relationships rarely trigger defections or rebellions, but this does mean that the system tends to be unintrusive. Most character drama can be resolved with a couple promotions or a new job.

    The family tree is a useful way of spawning new characters and roleplaying, but little else. Except for the leader and heir, most family members don't contribute anything extra to the state.

    The other new system that is much more present, though possibly not for the better, is the Wu Xing system. Nearly everything in the game is assigned and colored to one of the five classical elements from the Treatise of Five Powers and are supposed to interact with each other mechanically according to their elemental relationships. This means that every character is assigned to one of five classes and their strengths and weaknesses are supposed to follow on from this. The units they can, their skill tree, the assignments they can take, their equipment they can use, and their unit type are defined by their class.

    While this does make for some easy heuristics to follow, it loses much of the organic gameplay present elsewhere, feels restrictive and gamey, and makes little sense if you have any familiarity with the source material. One might ask why a general with decades of experience is unable to train any medium infantry, with the only answer the game can give you being that it arbitrarily decided that he was a Commander class and therefore it would be against his “element” to do such a thing. In reality, it’s because the game wants you to mix and match different general classes within armies, but the way its forced on you feels neither immersive nor congruent with the other features. General class matters more than relationships, stats, or basically any other factor aside from level, and its entirely artificial. The absolute nadir of it comes with the fact that only Strategist class generals can unlock unit formations without the use of ancillaries. This is both ahistorical and completely absurd, essentially meaning that an army full of professional troops led by the most experienced and prestigious generals can’t get into Loose Formation if none of those generals rolled the right class.

    Overall, the character management in this game is a great layer that helps expand the potential for organic gameplay and narrative. The satisfaction and relationship systems encourage clever staffing decisions without being too intrusive most of the time, despite some balancing issues. However, the restrictive class system is clearly trading a lot of this potential for either accessibility or a cheap gimmick.

    Administration and Infrastructure

    Where diplomacy and characters shine, administration, finance, and infrastructure just dully glisten. For a game set in an empire with perhaps the most complex bureaucracy in the ancient world, actually running your state turns out to be far less engaging than promised.

    Commanderies are laid out similarly to Thrones of Britannia’s provinces, with a central hub city joined with up to three resource settlements, though these mercifully have decent garrisons. The building tree for the cities is extensive, though not exactly complex, and the ones for the minor settlements are usually linear chains with at most one branch. There is thought that goes into figuring out what some of the more multi-faceted commanderies should look like, but ultimately there’s only two broad things to build up towards, production and defense. The latter only matters if you suspect the commandery capital might be in danger of attack. The former is just a series of flat income types upon which you can stack income modifiers. The main types of cash income (population, industry, and commerce) act essentially the same, with the caveat that population will increase peasantry income as well. It becomes very easy to optimize this past midgame, though picking and choosing what infrastructure to invest in with limited resources will still be important in the first half. However, because there aren’t any further considerations or endgame buildings to work towards, most administration just becomes matching the color-coded income types, and then deciding if you’d rather have extra defense or extra money.

    Food remains a big consideration, and operates similarly to Thrones of Britannia, though with one major expansion: you can now buy and sell it via diplomacy. This small change has a massive effect, as not only is food necessary to upgrade cities and keep replenishment going, but it can be used as diplomatic leverage and directly translated into money, as any faction running a food deficit will be desperate for it. Running a huge food surplus is an excellent way to buy yourself out of difficult situations and maintain stability. Some administrators and assignments can either cost or make food, but since there’s only one main settlement building that produces food and it will in no way be enough to offset the huge amount that the city chain itself takes up, the minor settlements that produce so much of it become valuable.

    Comparatively, the minor settlements that produce resources are mostly just extra income. While there are some building branches that unlock with certain resources, trade is now only influenced by what resources each side has, rather than how much they have. Having more tea settlements will therefore not increase trade money made from tea. There are some other settlement effects that are very useful, such as corruption and upkeep reduction, as well as the few that give you rare ancillaries.

    Ultimately, most settlement administration is just incrementally expanding the amount of cash, food, and resources you have to run your war machine. There’s very little further strategy that’s unlocked via making good infrastructure decisions, especially since recruitment has mostly been decoupled from infrastructure. While better than Throne’s baseline, it’s still not quite engaging enough past midgame to satisfy for long. The most important decision you’ll make with a commandery is who to make its Administrator, which ties back into the character focus.

    However, once you put an administrator in place, you’ll rarely engage with them. You can send people who aren’t doing anything else on assignments under them to gain some bonuses for the duration, but this is also a fire and forget task. Above your limited number of administrators, there are only a handful of gradually unlocking council positions and the prime minister seat, which is a far cry from the huge Han bureaucracy, or indeed, the number of people in your court who will clamor for some office. The council will form relationships with each other by proximity, but there’s no concept of superior/subordinate relationships between them, the administrators, the prime minister. There’s no bureaucratic hierarchy, and oddly enough, the council’s stats and traits don’t affect their performance in their job, unlike administrators and the prime minister. They merely exist as chairs to be filled to make people happy and occasionally milked for simple “council missions” to gain some easy rewards.

    The extent of the Han imperial bureaucracy is both writ small and makes poor use of UI space. Scrolling the character panels quickly becomes a pain.

    At the end of the day, the name of the administrative game outside of direct character management is to up production as high as possible and fill every desk as soon as it becomes available for more money and less subordinates bellyaching about not having enough to do. Not terrible but a missed opportunity.

    zoner16's review continues in Part 2
    Comments 1 Comment
    1. Welsh Dragon's Avatar
      Welsh Dragon -
      Just finished reading Part 1. A really good in depth review of Three Kingdoms, that's well balanced and has also taught me a few things about the game I didn't know. You also highlight several of the issues I have with the way the Total War series has developed in recent years (the emphasis on characters, at times at the expense of empire building being a major one.) Looking forward to reading Part 2.

      All the Best, Welsh Dragon.