Class Review & Balance Suggestions (spoilers obviously)

[Ryan Bonerius]

First off, thank you for the recent balance patch. It fixed a number of issues and is a huge step in the right direction. I completed most of the endgame bonus bosses after the balance patch, and have done a fair amount of testing of all the class options and builds I could come up with, and discussed them with friends who have done the same in the writing of this review. This is a wonderful game and I think with a bit of ongoing polish it could easily be an all-time great.

First off, a general issue I see: class combinations on physical character tend to be painfully restrictive due to weapon limitations on all or almost all of their skills. I suggest that a couple lesser skills on most physical classes lose their weapon requirements, depending on how common those weapon proficiencies are and how many non-weapon restricted abilities they have.

Second, while crit resistance has helped with late game tank issues, it’s still not widely available, and should probably be added in smaller amounts to most heavy/medium armor (maybe scaling up to 10/5% on medium armor/helms, and twice that on heavy?), rather than high amounts on certain very specific non-shield pieces.

---Warrior: All-around excellent basic toolbox that’s widely applicable to any STR based builds, either as a main or more frequently subclass. Passives are powerful and widely used. My potential only suggestion is that it gain access to medium armor, for use with non-tank builds where the enhanced threat generation on most of the armors is a detriment, but it’s more than fine as is.

---Monk: An offensive powerhouse with surprisingly effective tanking ability. With the buff to Focus Energy and other forms of AP generation, they can use their powerful skills very consistently. Their passives aren’t amazing but they fulfill useful niches, though HP Boost is perhaps slightly overcosted. Their only problem is Chi Burst, which outside of its status as an Ability rather than Skill, has little to nothing going for it; its damage is extremely poor for how expensive it is.. I suggest reworking it away from damage; monk has rather enough of that from its other skills. Maybe a 30 AP, no damage Silence?

---Rogue: What more is there to say about rogue? One of the most effective classes in the game, with blind, sleep, and high damage in its toolbox, to say nothing of the utility of escaping or stealing. Backstab is incredibly powerful on both Rogue and Hunter, Pocket Sand is wild, Reflex Stance is incredibly powerful in certain fights. B. The only problem with rogue is that stealing and escaping are too limited, making rogue at times feel necessary; adding them to accessories wouldn’t be amiss.

---Cleric: A very underwhelming class. Outclassed very early on by more versatile classes that can also heal well, the class’ selling points pretty quickly just become Spotlight, Mend, and Blackout; the latter two only being particularly relevant in specific contexts. Spirit scaling offensive spells are helpful but not really a draw, and Inner Warmth is numerically insufficient to ever really be worthwhile. The class needs work: party-wide status cleanse or lingering regens on their heals might give it a role as a premier healer.

---Wizard: Almost a perfectly effective bread-and-butter magic dps, wizard is hampered mostly by overcosted spells and a lack of utility as the game goes on, though Flare/Thunder remain very powerful throughout. Storm Stance is interesting but the class would probably be better served by a stance improving its offensive power and/or efficiency rather than a counter effect for some sort of mage-tank; and its tier 2/3 spells could stand to be something like 20% cheaper, given their lack of the secondary effects or sustain the competition enjoy.

---Warlock: A great utility caster with exceptional passives; even after the nerf it remains useful as a subclass, and as a class it is a unique chassis for interesting tanky mage builds. I do think it remains a bit poorly served by its stats; despite its equipment access its skill set remains pure mage, and shifting a bit of physical into caster stats could be helpful (though the higher Vit is useful for said tanky mage builds).

---Fencer: A very solid if basic physical dps. My only complaint is Checkmate; I find a 10% chance at massively boosted damage to be a poor tactical choice, really only useful for cheesing fights in a bad, random way. I suggest changing it to some other effect. From the name, and Fencer’s access to two turn based damage effects, I suggest an attack that consumes all poison, bleed, and burn on the target, immediately inflicting the total amount of damage those statuses would have inflicted, boosted by some percentage. Alternatively, Checkmate could inflict higher damage when enemy hp is below 50/25%.

---Shaman: A fantastic utility caster, every single ability on this class is useful, and spell lifesteal makes it an amazing chassis too. No changes needed.

---Scholar: A wonderful toolbox class, probably the best implementation of a blue mage I’ve ever seen. The only major sticking point is Atmoshear: player inflicted Confuse is useful in a vanishingly small number of circumstances; verging on none, and at all other times it's actively detrimental. I suggest that Confuse have a different effect on enemies than it does on players, or that it be replaced with a different monster magic.

---Aegis: Effective at what it does in itself (protecting a single powerful but squishy dps manly, but other defensive and some offensive utility), the class’ problems come from its passives. Natural Tank is overcosted for an effect that is both so small and so necessary; it ends up feeling like a tax, rather than a real choice. Honestly it would be perfectly reasonable at cost 0, but 1 would be fine as well. Tanks are already badly pressed for PP and having to give up 3 just to effectively tank in random encounters isn’t great. Stance Tank, on the other hand, is just difficult to use. Only a few classes encourage stance swapping in general, and those that do encourage doing so tactically, not to trigger a heal. I honestly think it should just be deleted and replaced with another passive, perhaps something that decreases AoE damage suffered by allies while the tank maintains threat?

---Hunter: A solid, self-sufficient damage class best paired with a support or utility class. Plays its role perfectly well and has widely useful passives, and particularly doesn’t want any change, other than maybe more diverse offensive options to pair it with.

---Chemist: Mostly just the ability to use items in battle, with some modest utility. Has some interesting tricks like AP Juice and Invisigel but many of its cooldowns are slightly on the long side and the class doesn’t quite bring to bear the toolkit other support options do. Fig and Sap Mixture are also dubious at best; restricting them to a pool of stronger effects might be called for, or replacing them with Protect/Shell mixtures perhaps. Selfless Cure is also so unreliable that even given the lack of good healing passives I just don’t see a use for it, even at 1 PP. It may just need to be replaced with something else: perhaps on-battle-victory HP/MP recovery for sustain while exploring?

---Reaper: An interesting and effective dual-role tank and dps. It performs both admirably well, and is actually surprisingly functional as a subclass on a strong physical dps with lifesteal equipment. Initial Oomph is harder to use than Initial Focus, but it still has a role, and Bloodspiller is perfectly solid. Doesn’t particularly need anything.

---Ninja: Ninja is frustrating, because it is simultaneously one of the most powerful jobs in the game, and one of the most annoying to use. The very low scroll cap means that it can’t be used for extended amounts of time; even just a standard dungeon will drain them empty before the end, and seals are frustrating to line up without specific options that manipulate the turn order. Used well, it trivializes many difficult fights and enables pretty ridiculous burst. It’s very easy to accidentally render it non-functional, however. I suggest that seals either last longer or not be removed by non-activating damage; reduce their power if need be. Also once again higher scroll cap I’m begging you.

---Dervish: The closest thing the game has to a straight upgrade, Dervish feels a lot like Wizard+, bringing stronger, more efficient spells and, as a chassis, better sustain and better equipment options, at the cost of somewhat longer cast times; the only thing Wizard offers over Dervish is Flare/Thunder, and with the right equipment even this advantage shrinks, though doesn’t disappear. This is mostly a Wizard problem though, and Dervish is perfectly fine. Focus Shield is niche but surprisingly useful.

---Beatsmith: Theoretically a rather strong damage/support hybrid, beatsmith remains rather difficult to use. locking in actions for three turns requires some specific planning, and when used as a subclass, it tends to be slightly underwhelming without its class trait to help. That said, it does remarkable sustained damage in a party that can play to its gimmick. I wouldn’t mind seeing its class trait weakened slightly in exchange for more base power on its abilities to open it up to other uses. Excellent passives.

---Assassin: With a solid and pretty expansive toolkit limited by varying stat scaling and weapon requirements, Assassin finds its own role defined by the class it’s coupled with. Rapier assassin does suffer a bit from overlap between Bloody Slice and fencer’s own bleed effect, and bow hunter is pretty sad now that its major selling point has been reduced to Harm Memory, a powerful but rather niche ability. Assassin is pretty packed but a new high-cost, high-cooldown, high-power Bow-only skill is probably worth considering. Crippling Coat is also a bit frustrating; while the effects are powerful, they’re unreliable enough that I find them hard to justify taking, especially given that both Assassin itself and other classes can inflict both statuses far more reliably.

---Samurai: Samurai only has one job: converting AP into damage, though it also provide very useful passives. Samurai suffers from being largely useless as a subclass and limited in its own effective subclass choices, due to its exotic proficiency and AP-thirsty nature, it ends up doing solid damage work both as a tank or physical dps. The class still doesn’t offer a ton compared to other Strength damage options, however, and probably needs either utility added to token builders, or doubling down on being the damage king of Str classes by increasing power.

---Valkyrie: A pretty premier tanking class, able to hold threat and survive well. It does suffer from high mp costs and very poor mp, however, and Steeling Strike isn’t sufficiently more powerful than Bracing Strike to justify the high AP cost. I suggest moving a fair amount of this class’ stats to MP, and to give Steeling Strike an additional effect, possibly Protect/Shell on self, as the name suggests? Commander remains badly overcosted; it is outperformed by Threatening except in the very longest fights.

---Beastmaster: I cannot find a real use for beastmaster, even after the CT buffs. It has weird jack-of-all-trades stats that don’t help with its variance gimmick and its damage output is somewhat mediocre compared to other strength classes, even committing to luck builds. Wild Stance is inferior to other options even discounting unreliability. It feels best served by something like a Valkyrie or maybe Beatsmith main class who can use the HP/MP drain to support themselves, but there are almost always better options. It’s too unreliable in a game where reliability is a huge asset, and doesn’t pay off big enough to justify the gamble. Maybe pull stats out of Mnd/Spr/MP to increase Luck and Str significantly?

---Summoner: One of the most powerful and versatile classes in the game with proper planning, summoner’s only real issue is Coyote: as noted in the scholar entry Confuse is an unhelpful status in player hands, and either needs to have a changed effect when inflicted on enemies, or changed to a different status. The randomness of Frostbite is also slightly annoying considering that it's largely only useful against random encounters and requires specific party comps, but isn’t a particularly big deal.

---Mimic: Mimic is hard to judge because it's comp dependent by nature, but from what I’ve tried and observed it has enough going for it to be useful in a variety of ways. Equip All is overcosted however; given that Equip Weapon passives exist it’s effectively just Equip All Armor and while that’s certainly a powerful effect, 10 PP is excessive and rarely if ever worthwhile, even with the most powerful gear.

---Weaver: Weaver is wild, one of the strongest classes in the game bar none, useful as a main and subclass and possessing some of the most useful passives in the game. Excellent, no changes wanted or needed.

---Nomad: A very interesting and effective hybrid class that can fulfill a ton of different roles, both as a main and sub class, helped by some widely applicable passives. The only problems with this class are En-Water and En-Ice, which simply don’t do enough compared to the other stances to justify themselves; the crit chance buff on En-Water is typically small, and while En-Ice is better, it’s rather situational.

[Reiker]

I hope you don't mind if I add some thoughts here as well.

Aegis

Before the last update it felt like the best general way to build tank characters was to set Aegis as the primary class and then either Warrior or Valkyrie as a secondary class depending on what you needed (single target vs. multi-target, power & armor debuffs vs. party support, etc). This felt pretty good and all of the game's primary tank classes were useful.

The recent Aegis changes kinda threw a wrench into that design and I don't really understand what the intention was. Now the Aegis is much worse at tanking itself, and is more of a "tank supporter" class, despite the rest of its design being of a tank and not of a support class.

It seems like we're now supposed to run an Aegis tank just for the Auto-Cover passive, or ignore Aegis on our tank character and give another character Chivalry just for the sole purpose of buffing the tank with 2 defensive abilities. This is very awkward since it's already difficult to choose between 8 ability sets across your party.

If this was meant to relieve tank characters from their turn economy issues then maybe Power Wall and Magic Wall should just be able to be cast on any target including self?

Cleric

I agree with you mostly. Without Blackout the Cleric could be completely forgotten. It feels like it's missing a few abilities from its kit, especially compared to Warlock which just does the Cleric's job better.

Warlock

Despite the problems you mentioned the Warlock is in my opinion still the best class in the game and it's not very close.

Their offensive spells don't do as much damage as a Wizard's but that doesn't matter much since they're cheaper and come with very relevant additional effects. Similarly Heal may not restore as much HP as Cure or Curen but I'd much rather have Heal for the debuff removal anyways. Warlock lacks multi-target abilities but Doublecast covers for that too. And that's still not mentioning Protect and Shell which are basically necessary throughout most of the game, and Refresher which is huge for sustaining MP through the early and mid game.

I don't really think that the Warlock should be nerfed (the Warlock Mail should though), but other classes need to be brought up to the Warlock's level. Clerics should probably have a big multi-target Curen + debuff removal (although this might still be worse than Doublecast Heal since Cleric spells also tend to have high CT, another advantage of Warlocks). There should be more options for setting Protect and (especially) Shell, etc.

Fencer

Checkmate is hard to evaluate. I watched someone who loved using Checkmate because of the randomness, and I watched someone else who hated the fight with the enemy that spams Checkmate since it was a random chance to just outright kill someone. These types of high-RNG skills tend to be pretty divisive in this way.

Fencer just feels weak compared to other melee DPS options and it definitely needs some stuff to set it apart.

Scholar

I'm okay with Atmoshear being useless since it's basically tradition to have a bunch of useless Blue Magic spells.

Chemist

"Additional Healing" bonuses affect MP healing too, so Selfless Cure is actually pretty useful.

Ninja

I don't know a whole lot about the class because scrolls are annoying and it initially felt underpowered to me. Scroll requirements should just be removed.

Other

I'd group Beatsmith, Nomad, Beastmaster, Samurai, and Ninja together as "classes that initially felt weak or without a strong niche so I didn't use them much." I've watched people use classes like Nomad and Beatsmith pretty effectively so they may not necessarily be underpowered, but I still think the design of these classes needs to be tweaked a bit further to make them more attractive.

[Arkhne]

Originally posted by Ryan Bonerius:

---Nomad: A very interesting and effective hybrid class that can fulfill a ton of different roles, both as a main and sub class, helped by some widely applicable passives. The only problems with this class are En-Water and En-Ice, which simply don’t do enough compared to the other stances to justify themselves; the crit chance buff on En-Water is typically small, and while En-Ice is better, it’s rather situational.

I agree with almost everything you said, but this is the only one I really wanted to comment on, since I not only wholeheartedly agree with those two stances being bad, but had a (probably bad) idea for changing them myself. I would like to see them changed to buffing the damage output of their element at the cost of the other. IE: En-Water would buff Water damage dealt by the Nomad, while lowering Ice Damage, and En-Ice would buff Ice Damage dealt by the Nomad, while lowering Water Damage. My only concern here is that it might actually make Nomad's offensive output too strong.

Originally posted by Reiker:

Fencer

Checkmate is hard to evaluate. I watched someone who loved using Checkmate because of the randomness, and I watched someone else who hated the fight with the enemy that spams Checkmate since it was a random chance to just outright kill someone. These types of high-RNG skills tend to be pretty divisive in this way.

Fencer just feels weak compared to other melee DPS options and it definitely needs some stuff to set it apart.

Fencer is tricky, it's arguably the first really good class you get, since early on, and in mid-game, Fencer is amazing, but once you start moving into end-game stuff, it really is weak, incredibly weak. It's really hard to make Fencer work in later fights, especially when something like Ninja is just outright better.

I think the reason for this is that in the mid-game, DoTs are really strong, but become pointless later on as raw damage scaling overshadows it, and that's where Fencer struggles. If Fencer's non-DoT skills had a stronger scaling, this would probably salvage Fencer's endgame (especially if Checkmate also got a chance buff).

[Ciaran]

Originally posted by Ryan Bonerius:

First off, a general issue I see: class combinations on physical character tend to be painfully restrictive due to weapon limitations on all or almost all of their skills. I suggest that a couple lesser skills on most physical classes lose their weapon requirements, depending on how common those weapon proficiencies are and how many non-weapon restricted abilities they have.

I've been working on a subclassing tier list, and this was by far the biggest issue with the game's class balance. The only skill limiting factors of magic classes are stat scaling between Mind and Spirit, with a caveat that elemental choice is a very minor limiter due to many equipment options only buffing one element. By comparison, physical classes have skills that scale on four stats (Strength, Vitality, Dexterity, and Agility) as well as limit to only one or two weapons out of a set of 12. It is massively restricting by comparison, with most physical classes having 3 or less subclass options that are compatible with both weapon and stats.

Brainstorming solutions also highlights a number of potential issues that are kind of baked into the UI, animations, and iconography of the game, if major fixes are ever going to be on the table.

Idea 1: Remove weapon restrictions altogether. This would bring weapon-based classes in line with the magic-based classes' freedom of matching based purely on stat scaling. Issues with this are that skill animations would have to be changed to work based on the weapon chosen, and even then likely have some sort of re-imagining of the visual effects in order to set it apart from other skills. It also would require the icons for many weapon skills to be changed, as right now they reflect what weapons can use them. All of this adds up to a lot of extra work for what ultimately may dilute the identity of each class.

Idea 2: Loosen weapon restrictions by enabling more weapons per skill. It could be reasonable for a lot of weapon skills to be compatible with more weapons without changing animations, which would significantly open up subclassing and retain class identity. This sounds like an ideal situation, but still comes with issues of its own. Some animations are much more compatible with multiple weapons than others. Take, for example, the Hunter's bow skills. There just aren't any weapons that would make sense firing an arrow out. So this would likely require the even more difficult skill icon updates (cramming 3-4 weapons into one of those small icons while keeping it readable is probably not going to happen, after all) and likely still a few animation updates to make it work, but with even an average of 3-4 weapon types per skill, it would create a massive increase in compatibility between physical classes.

Idea 3: Change physical skill scaling to work based on your dominant stat of a specific set. This would solve the disparity between scaling types, which could open up some class combinations to more efficiently work together. Fencer, Rogue, and Assassin, for example, have skillsets that could work together with a common ground of Rapier usage, but there are exactly zero classes that have high stats in both and also have access to a Rapier, making it just better to drop the Fencer from the equation and equip Daggers instead. Making their skills work off of the higher stat between Dexterity or Agility would be a relatively small change in the skill formulas for each. The problem that arises from this is how to display this new formula in the very limited UI space for it, requiring a trickling change in UI display for potentially all skills, which is a lot more work for what would be a relatively small change to actual class compatibility.

This isn't to say that some sort of solution to the limitations of physical classes shouldn't be pursued. I just wanted to bring up that a decent chunk of the game's design and presentation would have to be adjusted in order to pull it off ideally, which makes it a much larger obstacle than it may seem at a glance. I personally think the game has enough classes that it felt like there were at least some sort of synergistic subclass options available for any given class and build, so my actual play experience wasn't negatively impacted, but I would certainly enjoy it even more if there were more flexibility for effectively subclassing.

[Alilatias]

On the subject of Fencer, I think what would help is if one of their offensive skills was overhauled to basically trigger all remaining DoT ticks on an enemy at once, including burn. Then they would combo quite well with Warlocks, especially when Warlocks can wield rapiers by default. And it would give you a reason to use Rogue's Rupture as well, which applies 9 turns of bleed. Triggering all 9 turns of bleed at once is quite the amount of burst damage.

Right now, the class design feels very disjointed and doesn't really have synergy with much of anything at all.

If you intend to use Fencer as a reliable way to apply DoT as it currently is, a rogue running both blood spiller and venom coat passives will do just that with superior action economy, AND provide much more to the party.

[Illuminia]

It seems like a lot of similar thoughts to when i was building my Class Overview and ratings guide.

I haven't gone back into game to test each class again with the Balance Update, but in reading the patch notes it didn't seem like a lot of them got the attention they really needed.
Some classes feel like they are meant to be forgotten/replaced later in game. (example: Fencer, which I personally loved and used for maybe longer than the game meant for you to.... Fencer AoE + passives to apply bleed/poison were a huge source of my damage mid to mid/late game. Eventually was forced to switch to Ninja/Rogue because rogue felt too required, and Ninja main worked well with Rogue).

I understand in these games where you unlock more classes are the story progresses, you are usually meant to forget about earlier classes in favor of later "better" classes - Fencer feels like it's built to be a "great early/mid game but replaced later" class.

Then you have things like Rogue being practically required ..... with limited other class synergy (Req Daggers for many skills, so Ninja or Assassin are your options).


I don't think we're going to get much more in the way of changes to classes. Dev seems to want late/end game to be super-strict in what classes/strategies you are allowed to use.

[Wyrn]

Originally posted by Illuminia:

Then you have things like Rogue being practically required ..... with limited other class synergy (Req Daggers for many skills, so Ninja or Assassin are your options).

I don't think we're going to get much more in the way of changes to classes. Dev seems to want late/end game to be super-strict in what classes/strategies you are allowed to use.

Maybe I'm missing something about Ninja/Rogue, but I definitely did not feel like it was the only option in late game. I'm playing in hard, and just beat the two last bosses ( Ashley 2.0 and Spirit Cage ) without using those classes. Instead, my physical DPS was a Hunter for one of those fights, and a Fencer for the other - and in each case was one on my two main sources of damage.

In defense of Fencer :
- unless Ninja/Rogue/Assassin/Hunter, it can use multitarget attacks each turn, which is extremely useful (including against some bosses that spawn minions, like Spirit Cage ). Even more useful when combined with passives that apply debuffs like Crippling coat or Venom coat.
- sure, Fencer's normal attacks are quite weak compared to other physical DPS, but their critical hits are insane. Between their passive and their stance, it gives +85% damage to critical hits ; and with good late game equipment, you can reasonably crit 2/3 of the time, allowing you to do massive damage.

The main downside to me is that, for your Fencer to be effective, it needs to have BOTH good agility (because their abilities scale off it) and good dexterity (to have frequent enough critical hits). Fortunately, some equipments and accessories improve both those caracteristics at the same time, which helps a lot.

[Blank Syndrome]

I do think Beastmaster feels sort of like a 'flavor' class - it really doesn't offer anything unique that might compensate for its lack of reliability. Samurai is a lot of maintenance to output damage closer to what other classes do with less effort. Still perfectly good in its own right though; lacking in utility, perhaps.

Other than that? In my opinion, rogue is powerful enough standalone that the Mug ability lashed onto various daggers should just be a passive you get towards the end of its mastery, because there are a lot of party comps that really can't accommodate either a dagger user or sacrificing a sub. Not everyone wants the rogue skillset, good as it is.

It's tough for me to nail down any class I feel is too effective except perhaps Ninja. It is a little obnoxious to use with the scroll limit, but it's overall pretty overtuned for a reasonably accessible midgame class. Warlock is borderline, it's really powerful early and late, but it has a serious slump period in the middle chunk or so that takes it down a notch.

[Apostatis]

Mastering a class should grant you the ability to equip the classes passive in some way, when it's not the main job.

[Einwand]

Played on hard, overall game balance seems pretty good until somewhere around the time you get the owl at which point you mostly just have to not let the boss do anything or you die defensive options be damned, with occasional exceptions like the high -ele damage equipment pieces against exclusively elemental bosses. Enemies with multitarget spells were the most dangerous thing in the game by a huge margin, with most of them just outright oneshotting my entire party even with -magic damage debuffs in place if I ever let them cast
warrior - was a good solid phys toolbox for most of the game because it worked with every weapon type unlike almost every other job, and taunt was important until "tanking" stopped mattering
monk - I never tried it because of the high AP costs, when rogue was right there
rogue - is rogue, probably the best class in the game, tied with summoner and weaver
cleric - really is just worse at its job than warlock, with the exception of blackout being basically required for a few fights
wizard - fine early until you get better magic combat classes such as shaman and dervish
warlock - extremely good through the entire game, the only source of buff removal and the most consistent way to remove debuffs while contributing more to the fight than cleric does when support isn't required
fencer - great in the early/mid game but falls off hard as you get better ways to apply damage statuses/enemies become basically immune to % damage
shaman - extremely good, non-elemental damage means you don't randomly run into something you can't harm, has good damage with useful debuffs as bonuses at reasonable costs, as well as being the only other fast source of sleep though you probably have a rogue tossing a sleep bomb before anything can act if you need it
scholar - great healer/support in the midgame but most of its options kind of suck by the end, the only source of consistent phys up is nice, and sleep aura is useful on the occasional boss fight with multiple enemies you can't just immediately burn down
aegis - resist break is nice but warlock can already apply it and do other things at the same time, natural tank feels required for random encounters but the job itself just doesn't do anything you particularly need from a tank that warrior doesn't handle until later on when defense stops mattering
hunter - respectable but the -hitchance on most of their skills/their weapon makes them kind of annoying to use, good upfront burst which I made use of in the octopus fight in particular where rogue was just slightly too slow
chemist - literally never used this job other than for its stat growth, if a fight went long enough I ran out of MP my strategy was bad anyways, and the rest of its tools are done better by any other job or just plain useless
reaper - extremely good, its main passive was the only thing making "tanking" work at all in the late game, combined with the self healing while still doing respectable damage, since the -max hp debuffs don't apply to its class passive it becomes nearly invincible when stacking slime coat + undead ring + bone trophy
ninja - the scroll stuff is kind of annoying to use, if summoner didn't exist this class would feel pretty much required to be able to survive the amount of damage endgame boss fights deal, solid class chassis to put rogue subclass on
nomad - tried it in the anubis labyrinth, seemed alright but high AP cost stuff is mostly just worse than other options that don't have such heavy limitations
dervish - the premier midgame combat mage, better served with another job at the end and an annoying amount of enemies are immune to earth damage
beatsmith - didn't try this, looked weird, could probably be used as a tank on a difficulty where tanking would work without reaper's +hp cheese
samurai - so if I funnel this class 60AP it can do equivalent dps to a more consistent job, pass
assassin - another very good job base, most of its skills kind of suck but you mostly just want its class passive and it combines well with hunter or rogue
valkyrie - seems bad for its tank options but its support spells are a decent option to pair with warlock or scholar for a decent chunk of the game
summoner - absolutely horrendously broken once you get a couple of the good summons
beastmaster - didn't try it, even if I could somehow get a perfect variance thing going why would I want a phys job that's slower than the other phys options and more consistent utility options, Hew is the only other source of buff removal which would make it a little more reasonable of a choice if warlock wasn't one of the best classes in the game already, and CT on a physical job just sucks
weaver - it has haste and slow, and quick if you feel like it, unsurprisingly it is absolutely busted and if you really want to minmax you can do some nonsense with its turn manipulation passives/skills
mimic - there's probably more you could do with this than I did, but copying unique buffs like fierce stance and undead is busted enough on its own to make it a good choice

[Blank Syndrome]

Advancing into the very endgame, and against my better judgment as Red Mages are an all-time favorite class of mine, I've had to reevaluate Warlock somewhat - it's got stretches where it's mostly fine due to MP constraints, but Dispel needs to be moved onto Cleric. There's enough in the Warlock's kit (Protect/Shell/Doublecast/Remedy/Life, nukes, late game chain casting with the best class armor, excellent equipment pool, amazing class synergies) that it frankly doesn't deserve Dispel.

It muscles in on the value of the Beastmaster's buff removal, a class that's already considered dubious, and provides most of the healing you're likely to need for 95% of the encounters in the game while contributing offense and utility the Cleric couldn't dream of. Chain Dispelling trivializes a lot of pretty tough fights on a whole.

Dispel itself is, I think, completely fine. It's fairly expensive and costs a turn to do - if you're going to be chain casting them, it should be via the hefty cumulative price of Quick + Dispel, or the CT + CD of the Beastmaster skill. On Warlock, it's aggressively out of hand and shuts down many of what should be Crystal Project's most difficult bosses. If you have their class armor, this isn't even at the expense of their turn; a Warlock can wholly dispel a buffed boss and then full-party heal all at once.

I'm not really, like... A "nerf" type of person, especially in single-player titles. It's all about having fun, right? But the way Warlock treads on other classes is kind of the worst thing it could be - a no-brainer, something that's almost never the case in this game by design, and the one class I never found myself questioning the value of the entire run, while the rest of my crew was constantly iterated upon and constantly changing.

Keep its amazing flexibility, its deep equipment pool, its well-rounded stat gains, its late-game chain casting, etc etc. That's what's fun about Warlock. But Dispel gives it an answer to one too many situations, IMO.

[thehonorableryu]

I'm only 7 Crystals into the game, but the Aegis Stance Tank passive doesn't seem that bad. Warrior's Berserker/Defender stances don't consume a turn and have a very small 2 AP use cost. (Fencer's Hawk/Eagle stances also don't consume a turn and have zero use cost.) Getting 20% HP healing (especially on a unit with a large HP pool, especially on a unit tanking most of the hits) without consuming an item, MP, or a turn is a nice perk, and eases the pressure on your money and healing units in the early game.

The strongest argument against it, in the early game at least, is that Warlock's Regenerator passive that applies Regen (15% healing per turn for 6 turns) at the start of combat is better for the same 3 PP cost. But you may not necessarily want to train a tank unit as a Warlock to get enough LP to buy Regenerator when you'll be getting Stance Tank on Aegis anyway.

[Philippe Saner]

For the most part, I think the class balance is very good. But there are a few things that could use tweaking.

First, tanks in general are really bad against area attacks. Some way for them to prevent or mitigate damage to the squishes from nukes would be very helpful. Perhaps a debuff that turns area attacks into single-target ones, or a passive that reduces damage to allies as long as the user is top threat.

Second, rogues feel overly necessary. Not because they're overpowered, but because nobody else can do what they do. Would be nice if their schtick was a bit more widely available.

Third, the Cleric and Beastmaster aren't very good. They want some upgrades, I think.