Tree of Savior Forum

Design Breakdown - Swordsman Tree


WARNING - This thread isn’t made for new players or those that seek guidance on how to use said class, please look for proper threads regarding it such as class overviews and build guides. Design Breakdown is a design discussion thread only and has no purpose to help players on practical aspects of the game.

This thread will be edited upon Re:Build release in order to correct possible mistakes about these classes design, it is a broader take as the main subject are the tree and not the class itself but experiencing the class can sometimes present a different look when considering some elements (along having all descriptions translated).

Before get into the discussion itself i want to defend IMC for being so bold as to move classes out of trees altogether, as the goal is to set classes into more path options without power scaling and decrease punishment for certain combinations it looks like a good decision. However no big change comes with no drawback and this is visible from class migration and builds being taken away, both on synergetic value and the fantasy achieved by combining them. That said it wasn’t the only way to reorganize trees and this will be one point to be explored here, additionally the new tree has to be discussed as a new element in the already settled balance of roles and how their basic characteristics have to be fleshed out.

The main topics to address here are the unique value of each tree, how advanced classes are connected to it, which classes were/could be taken out due identity and theme, how shifted classes work on their placement and what these classes may face on their new environment. It may be impossible to restrain some comments that are more appropriated to a full class analysis so they may come back depending on how things go. With these threads i’ll also point out some undiscussed natural synergy and identity overlap within the tree. Don’t forget that the development team is subject to limitations and some ideas may not work due engine, practicality and balance issues. Sometimes it isn’t IMC’s fault.

Since the game release Swordsman was the worst tree regarding class synergy as it deals with a multitude of arms, most of the classes don’t have enough flexibility on which unconventional weapons setups to use aside for the class default or the one/two handed counterpart. With the addition of Scout this tree shifted all the focus into brawl, endurance and crowd fighting, they have access to more stat reduction and damage amplifier debuffs both being powerful tools when dealing with multiple enemies. Taking the mechanical side block and defenses aren’t affected by crowds aside the durability consumption and this is where the tree excels, for where the new tree goes it has can’t deal with that many enemies without relying on special tricks and must end combat quickly. On the thematic side Swordsmen classes will deal with military association, martial discipline and bravery.

Military - To be considered into military one has to be part of an armed force trained for war, this also include tactics, formations and teamwork as soldiers have to work in units to achieve a common goal. Most of the classes under the military front fought in platoons of hundreds to thousands which isn’t possible to translate when parties are limited to five players. Stretching it out a bit we can consider any class based on a group that fought a historical battle or participated into a war as part of this category. As an interesting point to address all armies had ranged units on their ranks being them a different type of soldier or even the same group that had training with ranged weapons but this is ignored due the balance issues and to keep consistency on both trees.
Martial Discipline/Lineage - Refers to developing and cultivate a martial art through years, passing it down from a generation to the next one. It can be done through training as contemporaneous arts do or through old scripts which is the case with medieval sword fighting that were lost through time, either way this is the cue for any fighting based class to join the tree.
Bravery - The guts to get dive into the battle regardless of the odds and that includes a striving will to survive, it’s a common element into the thematic narrative of classes that deal with harsh environments and presents fight as their only mean to live. However this same bravery isn’t the kind that uses all means to achieve victory but rather the one that fights with open arms embracing death if it comes to be.

Those elements define the thematic requirements for classes to be in the tree but they aren’t good guidelines to discuss classes and synergy, for that the best approach is to split them based on the major dissonant element that is their weapon choice. However it’s also interesting to take each class individually as they offer, or should, unique value for the tree, unlike the full analysis of design breakdown only the synergetic value and identity within the tree will be taken into consideration.

Spear Group

Being a spear user is the easiest path in the tree as the majority of restricted classes lean to a variant of the weapon, only Retiarius and Lancer have absolute restriction on their wielding and the other three offer no resistance at all. For a single hand spear user there are only four classes that doesn’t present a viable route with all skills available to use, those that take two hand spear as their weapon still have half of the tree to pick from. As an important point to keep in mind is that being able to use other classes doesn’t correlate to them being taken often, it is more likely for spear group classes to be taken together as they can make use of each other in full potential.

Lancer - Everything on it shouts “charge” and this is a rather interesting way to design it, Lancer is a class to get right to the enemy it wants as quickly as possible to destroy it. One issue with the class is how it has powerful AoE skills when Cataphract is right at the corner, even with the majority of skills having small hitboxes Gigante Marcha is way out of the class scope and reach. It’s partially understandable as a relic from power escalation but it doesn’t fit Re:Build.
Cataphract - As a counterpart to Lancer this one is the formation breaker, it isn’t just about getting into the battle but rather to stay in the center of the enemies and get them out of place. All it does has AoE and disruption, except for Rush which has defensive components, and that is even supported by Trot as a mere movement speed buff. The only issue with the class is how Acrobatic Mount pulls Lancer to the build, or itself to it, as is exclusive to mounted skills, the concept was never healthy to the game and it isn’t needed in the new class system. As a side note there shouldn’t be any reason for not allow Cataphract skills to be used with two handed swords given the way the class skill set isn’t backed up by the source nor that the attacks use the properties of spears in any particular way.
Dragoon - There’s nothing in particular on the class design that makes it stand out as it does a little of everything, ironically part ot the theme implies the class it is meant to be a ‘goliath slayer’ but that is barely touched. Of all spear based classes it feels like the one that has the heaviest blows despite needing more identity value, for now it won’t matter as it is likely to be taken by proxy in any spear based build.
Hoplite - On all irony this is the true ‘goliath slayer’ of Swordsman with critical buffs, damage amplification and multihits that take the enemy size to deal more damage. Historically speaking Hoplites had swords to be used in when opposing army phalanx formations were too close to each other and for that it wouldn’t be as bad to have Synchro Thrusting and Sharp Spear (or Sharp Tip) compatible with one handed swords and rapiers. By sheer buff value Hoplite is likely to be the most popular class in the group being part of any single spear oriented build.
Retiarius - Easily the weakest connection with the tree allowing it to be taken to Scout (as Murmillo’s thematic rival) with easy by a simple shift into the dagger side (losing one spear skill), after all their tactics were focused on evade attacks and disarm enemies before finish them with a single blow from a safe distance. Historically they didn’t used two handed spears but knowing the fighting style and how it was developed it would be natural for them to adopt the weapon if they had it available, this kind of reinterpretation is acceptable as it is the driving force that made such warrior groups change and form through history. The closest thing we have as an identity is a burst damage pattern and a possible PvP direction (which also fits Lancer and it has a more defined role), it would be expected to be some sort of opportunistic or sadistic class that sets up enemies to deal massive damage (with long cooldowns) and that is almost achieved with the net bonuses. That said the default utility the class provides may be enough to keep it relevant especially when it presents itself as almost weaponless class when taking the dagger route, since the offhand has no effect on it’s own it can be taken only in offhand and force autoswap when using a dagger skill which is likely to be the only way it will be used as the equipment itself brings nothing to the table.

Shield Group

Shields best trait is to be an offhand equipment and that allows any one handed weapon to be taken with it, it’s the most versatile group composed of only 3 classes and, in theory, is only dissonant to Lancer.

Peltasta - Changed into a dedicated tank and provoker as it was already directed to, all straight defensive buffs are allocated here and that’s enough to promote itself. As the only defensive shield class it deals with the tank paradox both on skill generation provocation but also by having the shield defense to be factored into damage (via attribute).
Rodelero - In short it’s a mobile class that focus preventing enemy action safely in a sort of attack and offense hybrid, this was already seen in the old design but Re:Build took the channeling skills and made their defensive value ramp up with levels. Comparing to the previous version the class change was minimal but this small change it is perceived, it won’t be just a control monster anymore.
Murmillo - The major accomplishment of Murmillo is to be a heavy damage oriented shield class as nothing else stands for an identity, it’s fine in the local sphere but is rather bland when you put next to other classes in the tree. In the Rank-Circle system Murmillo major value was to empower shield classes and bring them to high rank value but on Re:Build it isn’t needed being likely the reason it feels out of place, on that regard the class invites itself to shield oriented builds and doesn’t appeal to anyone else. The next shield based class might shaken out Murmillo considerably since it lacks polish, from the options to pick from it is the most dismissable and that may lead to its end.

Two Handed Sword Group

The name isn’t fitting as only one class has such requirement but that can slide off as the other two support the weapon and enable its use. By default taking two handed weapon classes closes a lot of paths and in this case in particular it only works out with classes in the Other division, even then those are a small group and has one class as incompatible. What this group needs desperately is more synergy outside its own bubble and it doesn’t seem like that will happen in any near future.

Highlander - The base design of Highlander pointed out to be a conditional damage class which is the type that requires time to set up all skills and deal massive damage, this was kept and simplified with Re:Build but still needs some push to be uncontested. Some researchers point out that Highlanders would rely on unusual tactics to win battles which includes undressing their kilts for pure shock value while charging towards enemies, that’s more of a Scout like trick but there’s not enough information on that regard to take it as true. We know for sure that Highlanders fought to protect their land only, this can be enough of a reason to keep it in Swordsman but is open to debate.
Barbarian - Even before Re:Build Barbarian didn’t had enough of an identity, it was split between building up power and disabling enemies and that is even less present now. In terms of raw power the class puts too much stress on Cleave, Frenzy and Warcry leading to a less interesting skill set, it’s likely for it to stay relevant as it has no restrictions and can empower build with ease but that falls more into the gimmicky side.
Doppelsoeldner - As a starting point this class doesn’t have anything special that it goes for, the latests changes added fixed combo chains to its gameplay which is a new addition to the tree even if done as a topping. This new direction put it closer to Highlander gameplay but being more limited both in combo variation and class choice, picking Doppelsoeldner is a death sentence to build pathing as the player will be forced into two handed swords or multiple weapon commitment. There’s a lot to be done in order to make the class not be a burden in the tree and such changes are really needed, unfortunately it isn’t possible to lift the weapon restrictions further as the class is based on real two handed fencing and few moves are shared among different weapon fighting style. Due that Doppelsoeldner will be an unlikely class to be seen outside specific builds despite offering an weaponless point distribution path.

Rapier Group

Among all groups this was the most troublesome, before Re:Build it was a group that added a new one handed weapon but didn’t want shield to be taken along. From a thematic side it made sense to avoid them as no class would use them but that is dumb considering the benefits of shield and their availability. The real issue with the group is how it exclude all other weapons to be used with those classes, it is expected them to provide at least one possible route to taken without forcing rapier into players. As a preemptive commentary both classes are unlikely to be taken individually as they have few synergetic options, the shield group in particular benefits more from its own bubble and to take a rebel rapier class as the third choice.

Fencer - Conceived as a quick class with reactive combat it had its place in the tree and still stays close to it, however the class could be more into timed dodge than burning overheats as that is what it makes it unique and is less prominent as evasion is mostly a Scout trait. Excluding Epee Guard, Preparation and Flash/Fleche the other skills could be performed with one handed sword with ease, even if that comes with a small penalty such as not factoring attack speed.
Matador - In theory Matador is meant to be a reactive rapier class that has players to time their attacks to avoid damage, just like Fencer. Granted that doesn’t mean the class doesn’t have anything going on but rather it lacks focus on what it wants to do and steps inside a class that shouldn’t, it needs more clarity on the direction and rely less on gimmicks to be a stronger contender. On the rapier matter the only skill that does require it is Faena, which is basically cloned skill, as Muleta could be performed with a sword, as a curious fact Matadores are split into three categories and one being a mounted spearman known as Picador (represented as attribute).


Hackapell - Right now it has everything to go wrong, it is a mounted class with a single weapon option, even tho rapier is compatible with the skill concept (but locked as infantry), that isn’t seen anywhere in the tree. To help it even more the base design was of a skirmisher class that would damage enemies from afar and finish them up with powerful melee hits but this is irrelevant in a tree that is all about brawl, it’s like the class doesn’t try to be taken at all and is too generic to promote itself aside being the mounted swordsman. Synergy wise the best options it has are Peltasta, Rodelero, Highlander and Templar (despite horse incompatibility), it is a laughable cast as even the rapier group has more options to pick from. Considering the class playstyle, theme and mechanics it is a perfect match for the Scout tree despite being main hand oriented just because it’s completely out of place and that won’t change until restrictions are lifted or a new one handed sword class arrives. It is a small price to pay for having it free from all restrains, in fact it can even allow players to take support classes and keep the focus on the main haind for a low investment build.
Templar - Being reworked again the class was rebirth as a commander that boost troops power by being in the heat of the fight, it’s a Swordsman that deals with buffs and auras which is an identity with no close contenders. For now we can’t know how useful it will be as the skills are tailored to GvG content and the last iteration of Templar still had a hard time in PvE, even that being the most important game mode, but this will need to be reviewed once the patch kicks in. It seems like the decision of limiting all skills to sword types is purely thematic and they should be usable with spear regardless, the class does offer a path for them but it has no real reason to not have all available.
Nak Muay - Before Re:Build the class didn’t had that much value in itself aside a super power buff that had all skills cooldown greatly reduced, this was kept and expanded when taking as a full class and also came with lower base cooldowns on the pretext to turn it into a fighting game character. The only issue with this direction is how it doesn’t care about the way skills are used once the fighting mode is active, there’s no particular reason to pick another skill over the highest damage one since they don’t interact in any particular way but that’s a matter for a full analysis. As it lacks weapon restrictions it can easily strive in any kind of build but seems more likely to be a complement to heavy restricted groups, however it still needs some identity tuning to reach full potential.

Former Swordsmen

All classes in this category were moved to Scout, mostly due compatible theme and gameplay in the new tree, where they can strive easier, that said they weren’t as alien in Swordsmen.

Shinobi - Since the release it was always an issue in the tree due clones, classes that had their skills copied received a huge burst of damage and those that didn’t were left behind in crumbles. Ironically Shinobi was one of the few classes that had high value with two handed swords ( and rapier builds which is gone for good, it is a major hit on these classes despite the unhealthy playstyle. On the theme side it does fit the new tree more as recon and espionage are under a same branch.
Corsair - As the only offhand class in the tree it never worked out that well, it was interesting for rapier builds (as shield was shoved out) but that’s where it ends. With a considerable share of strong two handed weapon classes Corsair value decreased over time and, along the absent identity, it became more of a thematic class choice, on Scout this will be less of a deal as it has no dissonant options. Oddly the class theme is quite balanced between the two trees but it when taking on the common sense it will feel more at home in Scout.
Squire - The only class that didn’t had enough of good reasons to be taken out, it is a supportish shop class that works on the principle of ‘equipment care’ as the identity and was taken away of the tree with most weapons diversity. IMC already positioned itself about the ‘path to knighthood’ subject justifying the change due the class support nature it has but that doesn’t click as well in thematics, in fact this change only serves to remove one of the few pure support classes that can be done in Swordsman in favor of loading Scout with classes of similar kind. Squire may never participate in combat at all but were always at the side of military, they are all about supporting individual warriors in campaigns and get equipment to be ready for battle at any given moment. Along that this is one of the few classes with no restrictions the tree had which is quite a big deal when most of the options aren’t available at all, now restricted to dagger use for the only offensive skill it has.

It is unlikely for Swordsman to ever become a friendly tree to play with due the weapon diversity but this is by far the best state of the tree, there’s still a lot to be done such as integrating Hackapell and two handed swords into a state they can be picked and have options on how to proceed.

Warrior Tree

As pointed out in the weapon discussion swords weren’t that useful in warfare which is the reason why so many spear based classes are seen (historically and) in the tree, even classes like Highlander and Barbarian would have their source more likely to use spears than swords as range increase survivability. It’s known as a fact that there’s nothing in Swordsman itself that does tie it to swordsmanship and the few three classes that do use them are the exceptions, if the tree operated under spear and shield only the synergetic issues would be greatly reduced and it can work under the name of Warrior which accommodates all classes.

Re:Build could be way more radical on the new class direction to the point swords could be removed completely, or at least rapiers, as another solution to the problem, the sole fact Scout is based on offhand makes it a great gateway for these out of place classes. IMC stated that one of their goals with the new tree is to reduce the stress of investing in multiple weapons to be successful but dual wielding is more effective in a system with no circles. Having to invest in multiple weapons consumes a lot of time and money which is understandable to be avoided but that’s not the same issue has not having synergy due weapon incompatibility which prevents classes to be picked or force double investment regardless of the new distribution, with the current class roster it is impossible to have a completely well defined class arrangement that won’t generate any kind of dissonance and that’s the point where the discussion has to weight which side’s convenience is more important. It’s more friendly and manageable for players to have different classes taking different weapons and be able to use both at the same time than having a full hand of options to pick from if you want to play class X or Y.

Ironically Corsair does is a class that takes the other hand side already making it the perfect mid ground class for this transition if one does wish to pick both sides. Classes like Matador, Fencer and Hackapell would never synergistically be out of place with Thaumaturge, Enchanter, Linker and Squire as weaponless options despite any other classes being eligible due their offhand nature. It’s important to recall that neither the rapier classes have any ties to military and weren’t inspired by any martial group, in fact they’re more into the sport side than combat itself but it’s understandable to not have the image of the weapon tied to a less honorable tree (even tho Matadores use terrible tactics to ensure their victory).

For the next tree we will get into religion using a similar approach of groups and individual classes on Cleric, i’m not sure about which one to take after that but Scout will be the last one as is the newcomer. As usual join the discussion on the comments below or drop a like if you just want to support the series, i try to make those threads as informative as possible but the real goal is to have the debate going.


I always enjoy climbing these walls of text~

Kinda suprised this appeals more to Mumillo than Rodelero.

I don’t think there should be any debate since the most famous of all Highlander “William Wallace” wore a huge 2-handed sword and that should be the general orientation (no, not only saying so because of “Brave Heart”)

The Hacka-Section made me really laugh and it isn’t that suprising considering it was already seen as an oddball in the Archer Tree, rightfully so. I really hope it will find its path.

I still think they didn’t take Nak Muay over to Scout since Swordies would lack a hidden class in that scenario. While the theme fits okay into the Swordsman Tree just like Squire, its way of fighting would be easier to be translated to Scout, since Squire has 0 synergy with any off-hand weapon whatsoever.


I’ll have to recheck my sources later but if i recall correctly, when i was doing Highlander research for their own breakdown, he’s actually a noble from lowlands (which had an organized kingdom) which has some significant cultural difference.

You just gave me a fantasy i didn’t knew i needed, a true mercenary with CQC combat as Nak Muay-Rogue-Outlaw (later replaced by something else). It would look amazing…


Here’s something to think about. It seems IMC has a direction for swordies compared to scouts. They want swordies to have shorter cd’s but lower damage compared to scout.


Great job with this post. I learned a few things about the classes individually and more of what to expect from rebuild regarding swordsman. You more or less convinced me that retairius and nak muay would’ve been a good fit for scout, something I didn’t already comsider, specifically trading squire for nak at least.

1 criticism I do have is that I don’t see how a 2H sword would work for Impaler and Rush. Although I would definitely be interested in using 2h hand sword for Cataphracts other 3 attack skills. If they could wield a 2h sword for at least those skills, that would improve the tree as a whole as that would open another option for 2h sword builds and for roleplaying purposes as well, all the more reason to ride into battle riding your steed while cleaving down enemies.



I saw Crevox’s translation on that and ironically i ignored this factor without noticing, i’ll add it in Scout’s next week.


Don’t take my word as 100% accurate, design has a goal and presents solutions for problems but some things turn out different than expected (especially in games). I never discard the possibility i got the wrong interpretation of the intended class design nor that my predictions will turn out to opposite directions. Also i wasn’t the one that pointed Nak as a candidate for Scout and personally don’t believe it fits in the tree theme, yet it would be fun to have it there.

Not all skills have to be useable for 2h swords, having a minimal build path for it could be enough even tho the ideal scenario is to have at least one extra skill to present some choice. Personally i can see Impaler possible with it despite being odd, Rush however seems natural for me as the principle is to swing the weapon in circles and a sword would be fairly more effective to slash in helicopter style (even tho if you consider the weight distribution it would be impractical).