Einander wrote:I think this class has some central design problems, and I'm not sure what can be done about it without substantially reworking the concept. I've played a character building Very Non-Optimally to get a grasp of how it works, and the results seem bad.
(I'm about to get very critical, so I feel like I should say that the concept is very cool and I like the ideas behind Animist a lot! It's one of the most interesting classes I've seen. It's just also kind of a mess.)
I had a big long point by point discussion/response to a lot of this, and then I forgot to save it and timed out of the forum login, losing it.
Anyways, I think a lot of your criticism is very valid, and I'm not at all discouraged by it. I really do appreciate the time you took to playtest and write all that out, and expect a lot of it to make that feedback to make it into later versions. My vision for the class doesn't always come through in the play of someone who isn't in my head, so apologies if some of my rebuttals seem like I'm dismissing your critique. Some brief points from what I remember of my response post:
• Animist's original design was "All the tools in the toolbox, but never all at the same time." The goal was for all the stances to bring something special to the table that isn't just your choice of sprite and damage type, so they have things like different utility skills and different ranges (a recurring sore point in your post was that ranges didn't quite line up the way one might expect). A lot of that differentiation is muddied now because there's almost no requirement to be in any particular stance, just max Spiritual Awakening. The big concern in everything about Animist has always been "What's the point in a class all about switching weapons if you never have to switch weapons?" hence all the ways to
force the player into switching stances, and some of the design problems are holdovers from when the stance switching was even more strictly regimented.
• As a follow up to that, the central idea of the class is difficult to realize. The stances have to have all the tools to function in their role and theme, but also can't really step on each others toes. For example, Salamander could get a shield because ToME mages love damage shields, but does that encroach on Gnome's role as "the defensive stance"? I had tossed around some ideas about another generalist tree that gives more stance-agnostic options, but couldn't settle on four talents that all four core stances might want. For example, Gnome and Sylph really want a Rush clone, but Salamander and Undine don't.
• Defensive option design is super tricky in ToME. You hit the nail on the head about its Path of Exile-tier burstiness, which necessitates huge amounts of effective HP. But in comparison to other classes, I don't think Animist is particularly low on defensive options. They're able to freely wear massive armor because Ether doesn't increase with Fatigue, and they have several heals built into their kit, plus Ether Barrier as free effective HP. They can almost always block because Gnome's Redoubt doesn't care as much about talent level. I compare here to things like Berserker or Cursed, which have similar life ratings and similar access to defensive talents (Juggernaut, Unstoppable, Gloom, Rampage), and far less access to ranged attacks than Animist. Most classes have to rely on Wilds to deal with status effects, and most classes have to rely on Movement or Teleportation infusions to get out of sticky situations.
Some clarifications about how the design was
intended, which is ultimately less important than how someone will figure out how to play it:
• Spiritual Awakening is always active, even if you don't have it ranked. It just reduces the talent level penalty for being in the wrong stance.
• Gnome used to have Block, when Redoubt was designed differently. When Redoubt became a pseudo-block, they lost it so as to not have Block and Block+. Block prodigies will interact with Redoubt like they would with Block. This should be listed somewhere, but the tooltips are already crazy long.
• Sharpened Spirit operates on a per-stat basis, so it should only ever serve as a net increase in your weapon damage (unless your stats radically change between equipping weapons). Animist was intended to be built multiple ways: Strength focused with Gnome and Sylph, Dexterity focused with Undine and Lux, Magic focused with Salamander and Lux. But since everything's rider effects tend to scale with Spellpower, Magic is the implied core stat, and Sharpened Spirit is meant to ease the gap of being a weapon class that may max key weapon stats second or third.
• Animist in general was meant as a class that would be comfortable bumping/shooting. They have typically above-curve weapons, especially once Egosmithing kicks in. Undine's intended pattern, for example, is to use Chill, shoot three or four times, and then Chill again, with Hailfalls whenever you pin a major target.
• On that note, the beams cast from Undine's Blessing can themselves generate stacks of Undine's Blessing. This is intended. Tuning them has been tricky because of a certain playtester's attraction to "EX Cirno Laser Disco". Grand Hailfall will also generate stacks when it stuns enemies.
• Egosmithing is what one addon developer has called a "honey tree talent." The "enemies in view" exists solely to prevent people from farming for a good Ego in a low-risk situation (e.g. in front of a honey tree).
• Salamander's T5 talent_on_spell is a weird bit of machinery, due to how it's considered by the game. Functionally, it means only Salamander talents will be followed up with Lash, with the edge case of Hailfall meteors landing while you hold Salamander, and
not whenever Dancing Flame hits a target while you're
not holding Salamander. It also checks per target hit (up to one cast per turn), so Sear and Dancing Flame can be very efficient triggers for it.
• On the flipside, you'll get Gnome's counterattack proc only when standing in Gnome stance. The T5 effects of Lux, Undine, and Sylph will all trigger when hotswapped to them, though.
• Gnome's "counterstrike payoff" is intended to be Earthbeat, which is tuned similar to Shield Slam.
• Ground Disruption is meant as a defensive talent, it takes a huge chunk out of opponent powers.
Some thoughts related to things you've brought up:
• Probably the biggest sore point is how tightly your offensive tempo (ether restoration) is tied to your defensive tempo (life restoration), and you don't always want both at the same time. It's a little bit of a shame because every mechanic I disconnect from the stance switching seems to drive another knife in the class concept. Considering ideas on how to disconnect the two.
• Ether costs need another tuning pass. This will probably include shifting Spiritual Armor and Grand Animism to full passives.
• Ether Reservoir's active needs reworking. In hindsight, the problem with it is really obvious and silly. When Spiritual Awakening was an active, it granted ether regeneration for several turns, so it'll probably pick up that effect so you can actually stay in the stance.
• Sylph could probably do with another tile of range on both Cutter and Cyclone. The "donut" of Cyclone then becomes "not adjacent" which will probably be easier to land.
• Sear should probably spark un-sparked targets.
• Dancing Flame may move to the capstone spot and become passive in some form to fit the pattern. In its place will probably be a defensive spark generator (e.g. "purge all negative status effects, take fire damage and generate sparks based on the amount.")
• Grand Hailfall will probably get an area increase to make it easier to use.
• Speeds will probably get normalized. ToME has all these speed classes that never get used and I wanted to use them, but I can see how speed unpredictability can lead to unpleasant gameplay hitches.
Thanks again for everything you've written. Real quality feedback.
