Let's Talk Tanks

[Shurrikhan]

Tenacity should make -tanking better-. Not "damage dealt", not "damage taken", but -tanking- overall. It should be the stat you work around breakpoints for before dipping into others. So how can we do that?

Idfk. Ya'll fickle. So lets try some things.

That's fair. My $.02:

What about a sort of revenge mechanic? Yes, it'd apply only to when taking/mitigating damage, but, so long as OT mutually exclusive choices (e.g. 10% mitigation and insignificant healing to co-tank vs. 20% mitigation on self) remain inferior to the personals they share CDs with, we're pretty well obliged to swap in and MT whenever there's damage to actively mitigate (and our mitigation sources ready are presently stronger than what our co-tank has available). Mitigating more = more counter-attack damage = damage balancing in a "tanking" way.

You can also apply buffs that last for a certain amount of damage to be taken, with a bit of server delay on their removal that doubles as a hidden buff. For instance, imagine if Shelltron were to apply a 12% HP shield which causes you to block all incoming attacks so long as it remains. Even if you only have 1 Shield HP left, you still auto-block the whole effect. Well, if your initial mitigation is stronger, that shield is also going to take longer to break, which means more blocking-based mitigation, which means... You get the idea. Likewise, with your Whirling Step effect above, if you had a temporary evasion/expertise buff that could be used to maximize either damage or mitigation, but rarely both in that the first strike it mitigates will consume most or all of the buff, stronger mitigation allows you to better double up the effect.

Edit: That being said, I'd honestly rather get rid of Tenacity, Direct Hit, and Piety, and have each role gain additional tailored effects from Determination, Critical Hit, and Speed... (Tanks, for instance, would recharge their means of mitigation more quickly via Speed, their effects would be dynamically increased via Determination (effectively, consuming a hidden resource in order to ensure that X mitigation type is "just enough" to reach some breakpoint, usually survival or debuff prevention), and Critical Hit would unlock bonus effects (so we don't rely on it for last-ditch RNG mitigation so much as it generates interesting counter-offenses). So, instead of 1 weird stat that's supposed to be more role-purposed, but is either obligatory or ignored, you have three different fitting stats to choose between.

[Kabooa]

That's fair. My $.02:.

With a robust enough sub-stat rework, I'd be pretty okay with relegating "Role specific" attributes folding into the base three.

While "X Shield is active" effects can be tricky, because then it brings about nesting concerns. What shields get hit first, should shields apply by priority or by order of application? We got to see some of the concerns with TBN's implementation, when I believe it was either lower priority or shields were removed in LIFO order.

If we fold role specific bonuses into the 3 universal substats of haste, critical, and determination, I agree that putting some valuable auxiliary effects works best. I do think assigning retributive bonuses to Critical would just serve to drive everyone to that stat anyways.

Perhaps Determination increasing duration of effects, Critical increasing Potency of Effects, and Haste being Cooldown Reduction?

For example, if we implemented that critical rework you post every now and then, and using Inner Release as an example.

Inner Release granting +100% critical chance and 100% resource reduction for 10 seconds at 90 second recast.
Haste reduces its Cooldown
Determination increases its duration
Critical would (I guess triple dip) and increase the Critical Chance and Resource Reduction bonuses.

Likewise, current Blood Weapon, Critical would increase blood and MP generation.

Adding a secondary tuning lever to each substat. The "potency increase" might scale from 0-20% over an expansion while Duration and CDR might scale 0-30%.

[Shurrikhan]

With a robust enough sub-stat rework, I'd be pretty okay with relegating "Role specific" attributes folding into the base three.

While "X Shield is active" effects can be tricky, because then it brings about nesting concerns. What shields get hit first, should shields apply by priority or by order of application?

I imagine we'd just go with weighted treatment: additional effects while active > on-consumption effects (so, avoid stacking them, rather than always stack them) > rest by duration.

Perhaps Determination increasing duration of effects, Critical increasing Potency of Effects, and Haste being Cooldown Reduction?

Personally, I prefer more or less general uses, but with the way a tank would use that affect inherently being fit to what they do. For instance, allow for "critical" mitigation. The actual mitigation bonus may be small, if even present, but it could have crit procs, just as many dps skills could. Determination would always be about using a hidden resource to meet a particular threshold (and granting pure bonus potency only when that hidden resource remains maxed/untapped). Speed will always be about... Attack Speed, with MP (which I'd like to make a thing again for those who currently just leave it as a useless blue bar graphic or generally a non-mechanic) ticks ticking per GCD.

Just as a point of comparison, Determination increasing durations acts exactly as a balanced form of Speed would, but without offering any control over gameplay pace. Critical increasing potency would play exactly as Determination does, ignoring any potential for gameplay adjustments except where it precisely makes or breaks a breakpoint (which then would largely overlap with your Determination, at -say- X additional gauge per Blood Weapon). Haste being solely a cooldown reduction doubles down on the balancing issues of Speed without, again, offering as much player-chosen gameplay variation. I'd want to double down instead on player choice, possibly even having players choose which, or what combination, of the two secondary stats present by default on their gear they want to use. That requires balance, but also some gameplay variation consequent to their decisions, and perhaps even some auto-adjustment (e.g. as a player I say I want to prioritize X GCD speed, Y Crit chance (and/or effect), and then dump the rest into Z, and my secondary stats will be chosen accordingly, similar to setting weights and minimums on ffxiv.ariyala).

[Kabooa]

Personally, I prefer more or less general uses, but with the way a tank would use that affect inherently being fit to what they do.

Generally speaking, sub-stats should be passive boosts that, at most, encourage angles of play.

Gameplay adjustments itself should primarily be the focus of kit interaction. Gear enhances the kit through bolstering passive effects. Unless i'm misunderstanding, giving specific interactions greater interaction through the stat itself (For Example - The auto-block when shield is present, with Determination granting the shield) seems the sort of thing stepping out of its lane.

Passive benefits might not directly change your gameplay, but it does adjust the relative value and weight of the decisions you can make, or enable some scenarios that you otherwise didn't have access to.

The offered attribute adjustments aimed for that, though admittedly as a spitball it likely misses the mark.

[Shurrikhan]

Generally speaking, sub-stats should be passive boosts that, at most, encourage angles of play.

Gameplay adjustments itself should primarily be the focus of kit interaction. Gear enhances the kit through bolstering passive effects. Unless i'm misunderstanding, giving specific interactions greater interaction through the stat itself (For Example - The auto-block when shield is present, with Determination granting the shield) seems the sort of thing stepping out of its lane.

That seems a fine opinion, but hardly any more than that. Mine is the opposite: if you can't use it for customization over gameplay, why should it even exist? Is it there only to ensure that one reward meant for your job should be better than another of equal item level, thus creating intentionally uneven or graduated gear progression targets? Is it to ensure that you have to meet priority checks like Accuracy, thus having to keep your ambitions checked against arbitrary requirements? Is any of that consistently fun, as compared to not only maintaining such checks (via gameplay breakpoints), but also being able to pursue new ones, with greater agency over those stat's use? I think stats should be simple, but gameplay-affecting, with intuitive player agency.

Passive benefits might not directly change your gameplay, but it does adjust the relative value and weight of the decisions you can make, or enable some scenarios that you otherwise didn't have access to.

Then it still changes gameplay, just in a way that feels less impactful or player-chosen. My idea is simply that there's some added flexibility and intuitive control: gear hold multiple options between which you have freedom to allot secondary stat points. You set, essentially, what you want to happen, and your gear allots its points accordingly. Where your desired breakpoints are not possible, it attempts your second choice system of priorities, or third, or so forth.

[Deceptus]

Tanks need to loose Determination and basically replace it with Tenacity. Tenacity should be the go to stat when you don't know where to meld next.

[Kabooa]

Then it still changes gameplay, just in a way that feels less impactful or player-chosen. My idea is simply that there's some added flexibility and intuitive control: gear hold multiple options between which you have freedom to allot secondary stat points. You set, essentially, what you want to happen, and your gear allots its points accordingly. Where your desired breakpoints are not possible, it attempts your second choice system of priorities, or third, or so forth.

It seems the sort of thing that would work better with a sub-stat expansion instead of a contraction, which I'm not opposed to. When the stats themselves play more direct roles into kit interaction, a greater variety of sub-stats also lets those stats have more meaningful progression since they do not cover more general cases.

In this particular example, have it be a defensive sub-stat (Maybe have gear with One Offense, One Defense stat as its budget, materia not included) with significantly higher gains, so said shield would scale significantly higher, rather than getting the same lethargic snail build over an expansion, not to mention that maybe you want to run a build without determination but still make use of a good shield.

Perhaps a retuning over the stat budget and reorganization is required. If gear is outfitted with one Offense, one Defense divided among its primary and secondary, and then materia can either fill the secondary or seek to bolster a third stat, there's a few more levers to tune in this regard.

[Shurrikhan]

It seems the sort of thing that would work better with a sub-stat expansion instead of a contraction, which I'm not opposed to.

I'm again not sure why you would assume that. Short of customization in the likeness of a sphere grid, a PoE skill map, or any other particularly dense talent tree, how are you divorcing specific gains (like "will my Shadow Wall be enough for this" or "can I have my Shadowskin counterstrike for me on all three hits of this tankbuster?") from general ones (like the Determination and Critical Hit examples I gave just before)? To separate general gains you would need tailored progression options in order to balance the weights of specific allotments, and even then those options would likely offer only a few competitive choices despite making the game far more esoteric and thereby divisive between players "in the know" and those outside. (That's not to say that I find such options to inherently boil down to "illusion of choice" -- no more so than job choice, at the least -- but that's not just a sub-stat expansion at that point; you'd be adding an entire Junction system of sorts.)

So, again, why not just take a running theme instead and apply it generally? The actual results of each can still be quite deliberate and inherently tailored to each role, since the rotational windows, procs, and numeric breakpoints of tanks and healers will inherently be linked also to tanking and healing just as those of damage-dealers will double down on dealing damage. I didn't pick those core three sub-stats arbitrarily; they specifically allow for control over rotational frames, and control over proc-based gameplay, and control over breakpoints outside of rotation or general proc chance, covering almost any and all gameplay alteration between them so you can tailor your job to your preferred macrorotations and windows for mechanical checks.

A slight sub-stat expansion, especially when allowing the choice between the offensive and defensive, only faces the same issues we already face: rather than, say, offering tanks the choice of how they want to tank, you offer the choice of whether to be more, or less, like tanks at all -- and the damage intake of any new tier, specific fight, or as faced by their toolkit in the context of their co-tank, will usually make that choice for them.

[Kabooa]

I'm again not sure why you would assume that.

Because what happens if I want to boost my shield via Determination, but I'd prefer to have Critical Strike and Haste as my focused stats? There's a specific part of my offense I want to focus on, but it's separate from what I want for my defense.

How would decoupling the defensive properties, give them their own categories, and then rebalancing the stat budget where gear comes with 1 Offense, 1 Defense, and then materia serves as your choice for where to further specialize, not serve your model better?

[Shurrikhan]

Edit: Confusion on my part.

How would decoupling the defensive properties, give them their own categories, and then rebalancing the stat budget where gear comes with 1 Offense, 1 Defense, and then materia serves as your choice for where to further specialize, not serve your model better?

Because we play under developers who prefer to make damage almost always the objectively best solution? Because when players have the choice of the unfun but numerically far superior option (as damage has pretty much always been in this game), most take it regardless of its relative cost to their gameplay? Because when you give options each from category Superior and category Inferior, the latter may as well not exist? Because it only really offers choices in answer to "How much can I be less of a tank" as the best progression path for tanks?

You'd effectively just give us gear with a single secondary stat each, completely contrary to the player choice my model is aimed at.