If You Were A Class/Job Designer...

[Shurrikhan]

Replies in bold.

PLD
2. I completely forgot to reply to 6, sorry, the only issue I had for which being that still 360 less TP per 3 minutes, 100 down from 460, when stuck as OT. I just don't see why TP changes should be stuck to either of these positions. The 'differentiation' really isn't worth it, imo.
3. As far as overall average output, it's definitely a buff, but much in the sense that Shield Swipe was. When you'd actually want it, it's crippled. I'm sure many PLDs would prefer the CDed oGCD version, but personally when I do use it, I usually need both casts to actually make the difference. Short of that it just becomes a better, support-capable (half-)Equilibrium, that can't be paired with (the now equal-CD, in case you were ever thinking of actually holding onto Equilibrium for burst saves before) Berserk. There are merits and demerits, but I guess it's just something I'm more personally against, based on my needs, than anything else.
4. 10s is cool, given the Defiance change. (Edit: now that you've buffed Defiance, it could actually even be lower, due to the MP cost, if there was anything else to really spend MP on)

WAR
2. It does though. In the OT role, where any WAR will be 90% of the time, the enmity is wasted, making it so there's at least some reason to use BB over, or with enough SS + Arrow/FW, between, the Storm choices.
3/4. You're at Butcher's Block with 5 stacks as Steel Cyclone/Decimate comes up. You can A) break Butcher's Block's combo, or B) waste Butcher's Block's stack (already capped). And this would be the conflict; my reference before was to the fact that WAR was the only tank that had no conflicts (or niche dependencies), only decisions.
5. It gave you something else to look for than either a 10s oGCD Grit or Sw/Sh Oath could -- falling under 75% hp before swapping from Defiance, and knowing how long it's going to be before a heal hits when swapping to it. It tied in a lot more with the healer, rather than just being something that, when at low enmity, you swap depending on whether you're in Butcher's or Storm combo. At 1 stack, Defiance HS, SS, BB, HS, Deliverance Maim, SE, FC, Infuriate, FC, etc. That's honestly where WAR enmity/potency output gets really scary, since the Defiance buff, more so than it's ability to tank in Deliverance outright (which doing this gets you to the benefits of sooner, without needing to hold off on any utility to maintain enmity lead).

DRK:
1. I just don't like the idea of a unbuffed version of an ability quite possibly granting more mitigation than its buffed version. At that point you're racing to spend it on SE in time just so you can get that shitty DA off your status bar. The other issue, though I could go either way on it, was simply that you now have two conflicting options. You either pop it early, wasting the 20% on AAs before the TB, while fishing for Reprisal, or you hold it to have the 20% on the TB itself, in which case it's basically a one-time ToB. I'd certainly like a buff for DD; I'm just not sure what it most ought to have.
2. Sorry, I was running a low estimate in case you needed 1 per minute on Dark Mind, etc. And ahh, I see then. Rates would indeed be very, very similar to WAR's stack moves, then. Still seems a bit much, though, given that your average TP costs elsewhere are basically Monk-level. Half would be really nice. Might as well attach the same to Power Slash too though in that case, right?
3. Sorry, I always multiply by any permanent damage modifiers when comparing tanks, but forgot to write that I had. 500x1.15 = 575. And... I still think it's going to be about the most overpowered move in the game. A NIN can losslessly Trick Attack after one GCD. Any DRK can reach its full raid damage (apart from cards, FW, BL, TA, HC, etc.) in a single GCD. A DRK becomes solidly the best 4-man tank, for whatever that matters. IN WAR/DRK comps, I'm fairly sure it actually would change something; since the Defiance buff and AP nerf, that 10 potency loss on BB is going to be far from noticeable in terms of dps. The new rotation it allows that would otherwise be limited to enmity-desperate 4-manning (not that a WAR would generally be enmity-short) would be noticeable, though. An MT WAR can now BB, BB, SP, where they would have lost 2 GCDs of SE uptime in a two-enmity combos per storm rotation. Far more enmity, with zero raid contribution loss. They may as well forget about SE, and just stick to Deliverance MTing even sooner, riding on a permanent 10% raid miti. Alternatively, had you left BB as is, it would also be a DPS buff depending on Infuriate timing, if running at a 2.38 GCD or less (iirc). Which makes it weird to me that you haven't really improved anything for OT DRKs here; DRK MT has been buffed, and now they buff WAR MT as well.

[Yuni_Queen]

However buffed version of dark dance is what makes DRK actually quite lower in terms of tanking than WAR or PLD.
20% evasion? Can anyone confirm, that lvl 60 dungeons or some kind sephirot trial has evade chance on at least 1%? In dungeons you may evade 2 or 3 hits from big pulls, but i don't know, if it's worth exchaning for DA + Abyssal Drain, since you will just self-heal yourself for sure, not a probability chance.
What's more, buffing evade by 20% doesnt mean we have 20% chance to evade attacks. We buff evansion stat by 20%.
And since DD increases parry by 30% not 50%, math is telling us to use something stable, that won't make healer's cry "use cooldown!".

"B-but i did... Dark Dance..."

[Violette]

However buffed version of dark dance is what makes DRK actually quite lower in terms of tanking than WAR or PLD.
20% evasion? Can anyone confirm, that lvl 60 dungeons or some kind sephirot trial has evade chance on at least 1%? In dungeons you may evade 2 or 3 hits from big pulls, but i don't know, if it's worth exchaning for DA + Abyssal Drain, since you will just self-heal yourself for sure, not a probability chance.
What's more, buffing evade by 20% doesnt mean we have 20% chance to evade attacks. We buff evansion stat by 20%.
And since DD increases parry by 30% not 50%, math is telling us to use something stable, that won't make healer's cry "use cooldown!".

"B-but i did... Dark Dance..."

Actually 20% increase in evasion in this game IS a 20% increase to evasion (flat), not multiplicative (as far as testing has shown). (Same with parry/block increases. Testing this at the moment in case I've momentary insanity).

Just as 30% parry increase is 30%+base+other sources.

Also, theoretically a 30% increase in parry is a 6% increase in physical defence, which can swing high/low. As we approach n=infinity, where n is number of attacks incoming, this will come closer to the true average.
Not to mention 20% evasion is (again theoretically) a 20% increase in defence against all avoidable attacks. This puts in on par with shadowskin (on a lesser cooldown), with the added parry bonus which adds to low blows/reprisal.

Now while we must recognize the lack of ability to normalize these % chances so they always happen X no of times out of Y tests, we can still see that these abilities approach their true affectiveness under 1 of 2 situations:

A) Multiple mobs are dealing physical damage to the dark knight

B) Mobs (needs not be they have many), are using rapid attack moves that hit quickly. These could be things like Bahamuts claw, Uplander Doom, etc.

In both cases, DA-DD will approach it's true effectiveness.

So while people may rag on DD all they want, it's still the most effective tool Dark Knight has for multiple mob situations.
Am I advocating the use of this for busters such as double rocket fist?
Of course not, but as you've so eloquently put it....."math is telling us", DA-DD can be more than useful as a replacement for shadowskin against multiple mobs.

[Tanathya]

I love DA+DD on trash, but other than that, it lacks of a real use on bosses. In fact, who ever uses DA+DD on bosses? It's DD without DA every single time, and just because it's there and not using it feels weird, but it doesn't do much in terms of mitigation, and I personally use it with the procs in mind, not the incoming damage. It seriously needs a change, or something added to what we have.

[Yuni_Queen]

I am not gonna argue, I had rare pleasure of avoiding hits, actually i rarely bother with using this against big pulls, since SS and/or SW was enough to take on larger packs of enemies.
Thou i will check it out, maybe its a waste of SW.

The problem are those swings. Old good RNG saying states: "If it's not 100%, it may be as well 5." However it cannot be considered as static cooldown to handle certain situations.
I would call it lesser cooldown, but i understand now how DD should be considered as.

Yeah, i wanted to put it in next post, but you made it faster. DRK is AoE adapted.
That's another aspect i find a bit burdensome. Boss fights forces me to keep in mind i have 2 omni-CD and one magic CD. when i use DD its mostly to proc low blows or reprisal.
I am not saying 3 cooldowns aren't enough.

My point is that DRK cannot give 100% of itself in every situation. Hi traits (and i speak mostly about traits) are AoE aspected. enchanced unmend and LB proc on bosses is rather "MEH".

[winsock]

Okay I admit it, I'm confused. How do the changes I outlined in any way affect WAR's stack management?

It more or less eliminates it. The only major consideration for holding stacks is IB. If you get the mitigation for free on change there's no reason to 'manage', you'd just spend it as you get it (small exception for triples).

can be annoying to stance-dance at times on the other two tanks, since they avoid their enmity combo. If you have zerked, BBing WAR while you're in SwO or Gritless, it can be a pain.

While I don't think we will agree on this being good/bad, I think we can both agree that the occasional saved dps isn't worth all the dmg this would do to the WAR playstyle:
* base rotation becomes stale, mostly 123 esp for OT
* no more maim considerations when spending stacks or using fracture
* raid utility would require next to no preplamning / anticipation
Many of the interesting mechanics WAR has just die or become very stale and uninteresting.

I'll give you the Clemency thing. Maybe a 600 potency heal with a 50 potency HoT over 30s? Although honestly the MP cost and the fact that its a lower potency than Equilibrium (with the changes), combined with the fact that none of PLD's CDs buff it like Berserk can, I feel makes it pretty balanced.

Ideally just spreading the potency over say 5 ticks/15 seconds or something similar. Maybe with a slightly increased potency if it takes HoT form.
Clemency can't be properly balanced with a sole focus on the tanking role as it impacts healers as well.

Since healing in this game is more timing based than HPS based, a HoT would likely be the least intrusive instant heal option, while still providing PLDs a skill they can conveniently use.