
Originally Posted by
Valkyrie_Lenneth
How would you balance talent trees for tanks? Its tank, tankier, and tankiest? or tank tank, tank dps, tank healer?
How would you do df if you got tank dps, instead of tank tank? Unless its actually varying, talent trees are pointless.
In short, I'd argue that tactics (healer-tank conventional timings) and strategies (primarily tank CD at different pull marks) already vary based on composition, slightly, but it may well be the lack of such specializations—if they were to offer an alternative to the trinity system in terms of titled-role proportioning—that limit that variance so much. That specialization could have occured on the class/job level itself, rather than in sub-specs, but either way, there's not that much differentiation. When it works, especially in raids, it works, (see Cover-M1, Clemency A3S, etc.) but for the majority of content it's hard to alter tactics enough to outweigh compositional preferences (e.g. for a SCH to make a PLD truly worthwhile in a high-AoE party and not enough dps to kill the enemy group before Broil Spam; no matter what he does, it'd be better for an AST to place a HoT and spam Gravity at that point, unless mana is a drastic concern, than depend on Bane while filling only with single-target dps or heals that the PLD can't really take advantage of to significantly boost his own dps anyways).
This kind of stuff is often why we see two divergent directions of suggestion for whatever classes don't fit certain conventions, e.g. PLD dungeon speedruns (since 1.2 or Wanderer's Palace NM), one asking to be able to perform more similarly to others (such as in PLD AoE dps), and others asking to be able to handle out-of-role functions or excel situationally in certain core functions so that they can make up for what they lack through party adjustment. But the latter type of suggestions, while well defended, may often seem less practical just because that isn't an aspect the community seems all that used to considering.
As it stands, the most variance I see is that when on PLD, I prefer to maintain threat via Clemency between Blinds on large AoE pulls than spam Flash. Upside is that I usually just need the initial shield or Regen for the whole trash pull. On the other hand, when playing ANY tank that can really AoE (or, a non-PLD), I'd prefer to go all out after AoE DoTs are up, unless an extra 1 in 5 offensive GCDs is worth more for the healer than an extra 25% dps is worth for me. That can be the case for a WHM or AST, but is only the case for a SCH if they're needing to refresh Bane DoTs, which I can easily enough delay or compensate for if I know the healer will act in a logical, predictable manner. Sadly, that's the end of variance as I've seen it. And a lot of that is because we're so dependent on the trinity as it's been. I can do small pulls as PLD without a healer, or with a couple more CDs as WAR or DRK, but there's no viable efficient strat that can forego a healer just because of me being an especially tanky tank, allowing for an extra dps (whose AoE damage, basically the way of the dungeon game, is typically close enough to a melee's most of the time anyways). Nor are there any efficient tank-less strats, even if every dungeon can be tanked just fine on a DRG or MNK, etc. with a decent healer, because 1) tanks and heals are close enough, while each supporting the other and group cohesion and ease of action for a greater raid dps contribution, to a DPS and either, and 2) a lot of the tools that could be used to adjust a particular situation simply aren't usable anymore (e.g. most forms of CC on most dungeon mobs).