Dark Knight Megathread

[Shurrikhan]

I know, I said that too but in a different way. No two similar things are the same, but red and blue are primary colors.

My point was that increasing max eHP (mitigation / ToB-likes) and recovering back to maximum eHP (healing) are opposite sides of the same interaction (surviving). (Red -> Purple <-Blue, if you will.) They work together, and may therefore amount to the same thing in many contexts, but if you treat them as literally the same, ignoring the unique advantages of max eHP increases, you end up with imbalances similar to what 2.0 Warrior (which died not due to type, but due to oversights) faced and end up seeing jobs fundamentally reworked instead of addressing a simple, very specific, lack of parity that bottlenecks the larger concept.

And my point is that I've done the same thing with HoC.

I gave that example because in as direct A|B comparison as I can make, TBN (and/or Oblation) did do something unique. That aforementioned dorito-run was with a GNB. They once tried to save an out-of-place DPS (a little over HP) with it. It barely wasn't enough. When I made the same attempt later on, TBN was enough. When both healers (both Sages) were about to take every possible AoE during snakes-fire-birds, he HoCed the other and I cast Oblation on both of them. They came into that damage with the same HP, but only the TBNed one came out (barely) alive (183 HP, iirc).

Now, if both heals had all of 200 more max HP, yeah, HoC probably would have done it, too, but that cases where, trying to save someone at ~60% HP from a single hit (not yet triggering Catharsis but not leaving them with enough HP, even at 1.43x [1/(1-.3)], to survive the next attack).

It's no huge point against HoC (though Intervention and Nascent, with enough precasting, could technically handle that situation better), but it is something that TBN does a bit better than its alternatives (in some cases reaching, yes, a unique result), and I really like that about it. And it is, again, just the difference between frontloading effect into entirely mitigation rather than partly into mitigation and partly into healing.

Glad that's cleared up. I was trying to say that yes your HP scales up which buffs your TBN value, but the damage you're taking also scales up with harder content. The older content gets the less important mitigation becomes since none of it is %hp based so it's not relevant for which class is better at what in any role.

Yeah, my bad there. I way over-assumed that your reply was much more narrowly aimed.

Let me try to be more clear. You know you're going to take damage so you use something that will heal you while you do.

Again, though, that only works if, without mitigation, you'd last for more than one hit. That is not always the case in Savage among tankbusters that can hit pretty damn hard. Any time we've seen reactive tank design ignore that need for single-hit max eHP, they get completely barred from certain fights, which sucks because, yeah, reactive tanks are fun, especially just as a matter of diversity. Single-hit max eHP (what is normally meant by "max eHP" anyways, but just to be clear) is not something that can be ignored past the casual level.

I was going off of all of this with last patches tomestone gear before getting Ex trial gear, which had me around 70k HP and I actually only melded once I already had accessories. Not sure if that's what you're getting at or not though.

No, no, nothing like that. I just mean that jobs can't be designed with only the current tier in mind so long as there's going to be notable gear-step going into the next. And, again, that you can't single-hit max eHP -- the difference between mitigation and ToB-likes... and healing. The two only become even figuratively interchangeable at the point ("non-reactive") tanks no longer need to use their skills.

If I hadn't seen whole, fun tank concepts (2.0 Warrior included) die over exactly that kind of conflation previously, I wouldn't harp on this. Sorry if that came off as anger. I'll admit I let myself get a little exasperated, hence (and worsened by) my error in comprehension, but I was never angry with you. I'd just seen that core idea that max-eHP increasing and simply HP-increasing effects are interchangeable kill tank classes/specializations/builds in the past and did not at all agree.

[Ryaduera]

I gave that example because in as direct A|B comparison as I can make, TBN (and/or Oblation) did do something unique. That aforementioned dorito-run was with a GNB. They once tried to save an out-of-place DPS (a little over HP) with it. It barely wasn't enough. When I made the same attempt later on, TBN was enough. When both healers (both Sages) were about to take every possible AoE during snakes-fire-birds, he HoCed the other and I cast Oblation on both of them. They came into that damage with the same HP, but only the TBNed one came out (barely) alive (183 HP, iirc).

Now, if both heals had all of 200 more max HP, yeah, HoC probably would have done it, too, but that cases where, trying to save someone at ~60% HP from a single hit (not yet triggering Catharsis but not leaving them with enough HP, even at 1.43x [1/(1-.3)], to survive the next attack).

It's no huge point against HoC (though Intervention and Nascent, with enough precasting, could technically handle that situation better), but it is something that TBN does a bit better than its alternatives (in some cases reaching, yes, a unique result), and I really like that about it. And it is, again, just the difference between frontloading effect into entirely mitigation rather than partly into mitigation and partly into healing.

Having done this several times for myself I find it hard to believe. they usually survive with around 40% hp after the heal takes effect when I do this on my GNB. Maybe it wasn't timed properly to take full advantage of the additional 4s of mitigation? Idk this seems incorrect. Not trying to call you a liar here, just saying my own experience is very different. Worth mentioning I only use HoC as an example here because it's the skill I'm most familiar with next to the other tanks.

Yeah, my bad there. I way over-assumed that your reply was much more narrowly aimed.

Aye o7

Again, though, that only works if, without mitigation, you'd last for more than one hit. That is not always the case in Savage among tankbusters that can hit pretty damn hard. Any time we've seen reactive tank design ignore that need for single-hit max eHP, they get completely barred from certain fights, which sucks because, yeah, reactive tanks are fun, especially just as a matter of diversity. Single-hit max eHP (what is normally meant by "max eHP" anyways, but just to be clear) is not something that can be ignored past the casual level.

If you're using heals without using mitigations in this game you're basically griefing, and GNB in particular has some insanely long uptime for damage mitigations with cooldown staggering. In this particular game there has never been a tank barred out of encounters because of these types of HP demands since each tank can deal with each encounter. A drain tank without mitigation works like this in, say, League of Legends where the drain tank can just be locked out with a single ability and killed in that time, but we're playing FFXIV, where damage reduction and heals are hand in hand and even tied to the same abilities now, and if you were to skip damage reduction while healing you're lowering the eHP that the heal even gives you, which is just inefficient.

No, no, nothing like that. I just mean that jobs can't be designed with only the current tier in mind so long as there's going to be notable gear-step going into the next. And, again, that you can't single-hit max eHP -- the difference between mitigation and ToB-likes... and healing. The two only become even figuratively interchangeable at the point ("non-reactive") tanks no longer need to use their skills.

I strongly disagree here. HoC is, again, my best example. Think of it like this, if you worded the skill like this...

Reduces damage taken by a party member or self by 15%.

Duration: 8s
Additional Effect: When targeting a party member while under the effect of Brutal Shell, that effect is also granted to the target
Duration: 30s
Additional Effect: Grants Clarity of Corundum to target
Clarity of Corundum Effect: Reduces damage taken by 15%
Duration: 4s
Additional Effect: Grants Catharsis of Corundum to target
Catharsis of Corundum Effect: When you take damage that would drop you below half HP but would not kill you, reduce that damage by an addition value equal to a 900 cure potency
Duration: 20s

...then it would be damage reduction, but that is in effect what is happening. The heal occurs the instant the damage is taken. While they are fundamentally different, the effect that happens is extremely comparable to the point that if you showed both effects without showing HP recovery being visible (those green numbers) then the person seeing the effects in game wouldn't know the difference at all. Being a heal just technically makes it better since you will always get the full 900 potency (think being at 51% and getting hit for 2% damage) and it can crit, but I absolute hate using chance in discussions like this and I always assume a fail-state for said chances. As long as the condition of survival is met, which it pretty much always is even before 6.0 release (always able to survive without a DRK) you get that value of eHP.

If I hadn't seen whole, fun tank concepts (2.0 Warrior included) die over exactly that kind of conflation previously, I wouldn't harp on this. Sorry if that came off as anger. I'll admit I let myself get a little exasperated, hence (and worsened by) my error in comprehension, but I was never angry with you. I'd just seen that core idea that max-eHP increasing and simply HP-increasing effects are interchangeable kill tank classes/specializations/builds in the past and did not at all agree.

So did I, apologies. That being said, usually the thing that kills it is the communities opinion on it. Because heal tanking is reacting instead of proactive it is a naturally more difficult style to optimize, so novice players of this style see it as weak and seasoned players of this style make it seem broken because "no damage can stick to them." Seeing a tank drop in HP and go back up on their own is going to be seen more than a tank that can reduce damage heavily because... well, on the latter, there is nothing to see in the first place. I do, however, think the real problem with DRK is that it doesn't really get either of those... it has downtime in damage reduction and TBN desperately needs to be taken off their damage resource.

Unreleated idea, if you really want TBN to cost mana, why not indirectly cost mana by making it just consume 30s of Darkside? Not exactly a great fix, but a thought I had, albeit far from a deep thought.

[Shurrikhan]

If you're using heals without using mitigations in this game you're basically griefing.

I'm just comparing the two (a heal vs. a shield). The reactive element can complement the proactive, replacing the need for increased HP value per point outside of potential one-shots (which, unless greatly overgeared or doing only casual content, are a thing), but you need to survive the initial hit first. That might not require any mitigation in content you've seen thus far, but unless you want to turn all use of defensives into simple bonus Glare casts, irrelevant to actual tank survival, that difference will matter at some point.

Don't get me wrong; I like reactive tanks. Such makes for, imo, much more enjoyable play as you have this 'zingier' sense of fighting for your life instead of just delaying the inevitable (even if, ultimately, you'd still die at the very same point, give or take self-healing ppgcd variance and (un)lucky timing, if wholly balanced).

I strongly disagree here. HoC is, again, my best example. Think of it like this, if you worded the skill like this...

But even that still changes a further aspect. That mitigation now increases max eHP for the next attack, whereas the cure will not. Will that be significant? Maybe, maybe not, but it is a difference -- the same one as always: max eHP increase and finite duration of eHP increase (since every mitigation skill is, yeah, generally tied to a maximum duration).

That is "in effect, what is happening" only so long as the latter max eHP increase is irrelevant, just as for any other tankbuster. Games oughtn't design tanks, though, around such casual content as to make skill use irrelevant to surviving tankbusters. Or at least, if they don't want to rob tanks of a significant portion of their excitement.

Seeing a tank drop in HP and go back up on their own is going to be seen [seem?] more than a tank that can reduce damage heavily because... well, on the latter, there is nothing to see in the first place.

Oh, I agree. That can sometimes maybe go a bit too far, as when trying to heal a Death Knight tank with a Holy Priest without decent enough cooldown trackers for said Death Knight, where some relevant waste is even guaranteed just due to possibly crit-healing at the same time, but having the moments where a tank's HP goes down but I can know "s/he's got this" and sure enough they heal themselves back up has a lot more... flair than just "Eh, 40% HP will still last them 8 more seconds. Glare. Glare. Glare." They 'should' be the same--and from the healer's end alone, except for some added constraints in large-heal timings, they are mostly the same--but they feel different.

Unreleated idea, if you really want TBN to cost mana, why not indirectly cost mana by making it just consume 30s of Darkside? Not exactly a great fix, but a thought I had, albeit far from a deep thought.

Nah, don't sell yourself short; there's actually some interesting interactions that it could put forward. I'm not sold on them, but I'll get to them as soon as I cover what I believe the existing state around TBN and its MP costs to be like, generally:

In practice, TBN is already seeing half to two-thirds more casts than its 25s equivalents even despite its mana cost, so I'm not sure why its mana cost would have to be removed. The MP cost and DA interaction is the reason we can have such a low CD, and I'd much rather hold onto that low CD than drop the MP cost.

Even if we could change the initial MP cost to a MP penalty on failure to break (allowing us to somehow go into outright MP-debt), I prefer the margining vs. all-in behavior of having that MP cost; if I push a final Edge into raid buffs, I like that there's that risk of not having access to TBN or a bit. It's a simple, decently intuitive, but worthwhile mechanic, imo (so long as content doesn't end up pervasively undertuned as to make it damn hard to ever pop).

Okay, now for Darkside. At present we can generate 11.4 to 12k MP per minute (roughly 4 Edge/Floods/mana-cost-TBNs), with E/Fs giving 30 seconds' Darkside duration a pop, leaving us with exactly just a minute's excess. That's our starting point.

This brings an important question, though: Should DA-Edge/Floods grant Darkside, essentially refunding their costs?
___________

Yes?

If so, you effectively increase TBN's ideal Edge/Floods per minute from 4 to 6-8 (8 if TBN's cooldown somehow isn't increased), since TBN itself costs no mana. Though, you'd have to hold at least 39s Darkside going into the TBN cast, which... might force some "clunky" bar-watching.
More importantly, gameplay-wise, rather than not getting one of your set number of Edge/Flood casts back, you're not getting a "bonus" Edge/Flood.
I have to admit, I am not a fan of that. When we account for the added Edge's impact on total dps and the kits gets balanced for it, not only do we get a wonky spike in throughput all at a single level (70), but we're not actually getting "bonus" casts in terms of actual throughput. We'd still be wasting throughput every time we fail to break a TBN, just in a less obvious and now more punishing way, all while having to constrain defensive casts around offensive casts and introducing new MP overcapping issues.

____

No?

In that case, you'd need to set the cost value at something that wouldn't put TBN at a lower frequency than the other tanks' 25s CDs. If TBN were to cost 30s, and you only have 60s spare to work with per minute, you'd actually have just under 2 TBNs available per minute AND you'd have to delay Edge/Flood casts from raid buffs just to avoid overcapping Darkside so that you could even use said TBNs up to twice per minute.

At a more reasonable value, such as would allow for an effectively ~20s CD on TBN, still flexible to 15 seconds if weaving an Edge/Flood between, down from 30 seconds, it'd still give free Edges and there might therefore be an impact on our other potencies to try to keep our average available ppm the same, but at least it'd give us a couple potential screw-ups per minute that... No, no, it'd still then be costing us edges. Hmm. Darn.

_____________

Next question then: Should TBN then grant any free potency at all?

If yes, you still have the above issue.

If no, you could use it, sustainably, n/60 times per minute (where n=the Darkside cost), and, for the most part, you really would be able to cast them as many 15s apart as that same "per minute" value allows. This I kind of like. Quite a bit, possibly, though I'd need to actually play around with it.

I like the current MP margining involved with TBN, since it makes playing with that margin feel more deliberately "all in" (even if once quickly learns whether it's an actual risk or not and therefore to do or not, but that's the flattening that all content-derived interaction/gameplay faces with experience), but... having to actually think about our Edge/Flood casts for future TBN allowance, if done right? As long as the constraint's aren't too tight, such that TBN feels convolutedly "unlocked" by Edge, I think I'd actually dig that. Again, I'd need to really imagine out the moment-to-moment decisions, though.

[Ryaduera]

I'm just comparing the two (a heal vs. a shield). The reactive element can complement the proactive, replacing the need for increased HP value per point outside of potential one-shots (which, unless greatly overgeared or doing only casual content, are a thing), but you need to survive the initial hit first. That might not require any mitigation in content you've seen thus far, but unless you want to turn all use of defensives into simple bonus Glare casts, irrelevant to actual tank survival, that difference will matter at some point.

That's awfully presumptions of you to say. Let's put it this way. I still have PTSD from Gordias Savage, so I don't want to hear "any content I've seen thus far". The server tick counting... Oh the horror... And Thordans eyes. Oh lord his eyes, no please god...

[EDIT] if you got that by looking at this characters job levels, this was a new character made to play with some friends on a different data center and is not indicative of my actual experience with this game, but I use this one because it is the one I will now exclusively use.

Don't get me wrong; I like reactive tanks. Such makes for, imo, much more enjoyable play as you have this 'zingier' sense of fighting for your life instead of just delaying the inevitable (even if, ultimately, you'd still die at the very same point, give or take self-healing ppgcd variance and (un)lucky timing, if wholly balanced).

But even that still changes a further aspect. That mitigation now increases max eHP for the next attack, whereas the cure will not. Will that be significant? Maybe, maybe not, but it is a difference -- the same one as always: max eHP increase and finite duration of eHP increase (since every mitigation skill is, yeah, generally tied to a maximum duration).

It's a fundamental difference but it yields identical results, since the cure that happens on HoC is instant, meaning it also instantly increases max eHP as part of the action. There is no functional difference, this is how it works in the games actual code.

That is "in effect, what is happening" only so long as the latter max eHP increase is irrelevant, just as for any other tankbuster. Games oughtn't design tanks, though, around such casual content as to make skill use irrelevant to surviving tankbusters. Or at least, if they don't want to rob tanks of a significant portion of their excitement.

I cannot begin to tell you how many times I've caught an auto after a tankbuster in difficult content... But it matters for more than just tank busters anyway. Anything that can be used to survive a buster will do regardless of what it leaves your HP at, but using it as an actual general mitigation tool as you should is where the measure becomes truly telling since every tank can indeed survive any tank buster because the balance team here is on point. Balance is not the issue with DRK.

Oh, I agree. That can sometimes maybe go a bit too far, as when trying to heal a Death Knight tank with a Holy Priest without decent enough cooldown trackers for said Death Knight, where some relevant waste is even guaranteed just due to possibly crit-healing at the same time, but having the moments where a tank's HP goes down but I can know "s/he's got this" and sure enough they heal themselves back up has a lot more... flair than just "Eh, 40% HP will still last them 8 more seconds. Glare. Glare. Glare." They 'should' be the same--and from the healer's end alone, except for some added constraints in large-heal timings, they are mostly the same--but they feel different.

No, not seem, seen, as in visually watching the healthbar jump.
This literally exists in Warrior. I don't think drain tanks necessarily belong in FFXIV. In terms of design, I like GNB healing wise because it's calculated and rhythmic healing, Aurora when consistent damage is coming and a charge is available, HoC between cooldowns unless a tankbuster or mechanic requiring it is expected. Not draining, but a general regeneration. PLD is similar but their self healing is built into their DPS rotation, so since it is naturally used less intentionally, it doesn't quite check that box of using it defensively and with defensive intent. For GNB, it's satisfying and rewarding, something that DRK is not.

Going to point out the accessibility of HoC is often not seen since most GNB players literally only use it for tank busters and also tend to not use it on the main tank. The same applies to TBN, but still worth pointing out, since I think I see GNB use HoC once every 60-90 seconds as opposed to rotating it as a regular part of their defensive cooldown rotation. I don't hate TBN because to be perfectly honest the dream is having both of these at the same time.

Nah, don't sell yourself short; there's actually some interesting interactions that it could put forward. I'm not sold on them, but I'll get to them as soon as I cover what I believe the existing state around TBN and its MP costs to be like, generally:

-snip-

Not really selling myself short, just if I were to sell this as an actual idea, I'd have to actually think about those interactions before claiming it as a proper idea. A good read though.

[Shurrikhan]

That's awfully presumptions of you to say.

Fair enough. I forgot that the forum accounts are separate from your general Square Enix account and you might therefore have started more than the above two months ago, and with one-shots (or bust-and-auto killers) having yet to get mentioned I did not check Lodestone first (though, looking at it now, I assume also that this is your alt?).

It's a fundamental difference but it yields identical results, since the cure that happens on HoC is instant, meaning it also instantly increases max eHP as part of the action. There is no functional difference.

That's a matter of context, not functionality. A 100,000-damage attack and 4,000 damage attack are "the same" if their target only has 4k HP. That doesn't mean their isn't a difference ready to go the moment you hit an enemy with 4,001 HP.

I cannot begin to tell you how many times I've caught an auto after a tankbuster in difficult content...

If there's not enough time to do anything between the two hits, they will act as one hit until you have something that can be woven between them (e.g., Catharsis, since it's not going to wait for human reaction times... though it still has a little delay of its own, sadly; I've certainly been AAed off before even it can go off).

No, not seem, seen, as in visually watching the healthbar jump.

Right. Sorry, my eyes have trouble catching small words; somehow I passed over the "be" both read-throughs.

This literally exists in Warrior. I don't think drain tanks necessarily belong in FFXIV. In terms of design, I like GNB healing wise because it's calculated and rhythmic healing, Aurora when damage is coming and a charge is available, HoC between cooldowns unless a tankbuster or mechanic requiring it is expected. Not draining, but a general regeneration. It's satisfying and rewarding, something that DRK is not.

Milage may vary. I find GNB more fun than Warrior in raids in every way but its defensive and healing setup, though I certainly enjoyed WAR's healing more when it better rewarded banking (in that its short-CD had a lower healing floor and slightly higher ST ceiling, all on a shorter duration).

...The only thing in DRK's kit, defensively, worth note to me are TBN, Oblation (kind of), and Abyssal Drain in AoE when my healer's... not great. And yet, for me, TBN and Oblation themselves faintly, faintly put it over GNB for me.

A good read though.

Sorry it got so ramble-y. I just did think it tapped a cool area, actual interaction design pending.

[Ryaduera]

-snip-

I have actually never enjoyed Warrior, even in 3.0 when it was... well, slightly less broken than it is now. I do realize that whichever tank you enjoy for whatever reason is going to be entirely opinion. I actually don't main GNB for that reason, but rather because I like noisy but calculated classes. As GNB everything has to fit within your 60s burst window and 30s window inbetween. It's fun and requires good awareness of both your abilities and the encounter. This is, of course, opinion, as similar things could be said about DRK ( why I want to love it, actually) because it is also noisy, it's just far less calculated and more of a "BURST WINDOW GO" kind of playstyle.DRK just feels left out. It doesn't have heals, it still suffers from the magic resistance gimmick (the only reason I dislike Dark Missionary is because it's literally Heart of Light, but you get it later). If they'd just fix Dark Mind I entirely believe it's defensive abilities would be fine, it just needs some love in its offensive abilities (not in damage, it is top right now) to make it overall more fun and satisfying and giving it a heal off of its offensive abilities would be great, something like a modernized Sole Survivor.

Also never forget they took Dark Dance away, reworked it to not be evasion but just damage reduction, kept the parry, renamed it Camouflage, and gave it to a different tank entirely.

[Shurrikhan]

I have actually never enjoyed Warrior, even in 3.0 when it was... well, slightly less broken than it is now. I do realize that whichever tank you enjoy for whatever reason is going to be entirely opinion. I actually don't main GNB for that reason, but rather because I like noisy but calculated classes. As GNB everything has to fit within your 60s burst window and 30s window inbetween.

At present, what with Double Down and whatnot, the only thing I'm really enjoying about GNB's lulls is that it helps to remember which are full lulls and which are Bloodfest lulls, and then either burn (apart from ensuring you'll have enough for GF combo when it comes up) or bank over the lull accordingly. (Because of that, I don't particularly hate Double Down, contrary to a lot of the flak it's been getting, as it doesn't really change what I'd be doing anyways.) The rest just feels so rigidly on-rails that it's not my cup of tea compared to, say, HW DRK, or any of various other MMOs' tanks. The 30s CDs, meanwhile, are just more click-on-CD filler to me. (For much the same reason, I never cared about Circle of Scorn outside its handy mob-grabbing while sprinting through dungeons, and cared even less about Spirit's Within once its HP interaction was removed.)

For me, DRK and WAR, rotationally, are on the opposite extreme; they're diarrhetic. Outside of banking for raid buffs, and not overcapping, nothing seems to matter. You just throw shit out there. (Of course, GNB is the same in that you just don't want to overcap and do want to maximize window, but at least there's enough desync between BF and NM to make it feel like there are distinct phases among and between bursts, whereas Enhanced Infuriate is a constant but negligible mechanic and, well, DRK has nothing that falls away from the 60s timers.)

I think that's the first thing that'd need to change about DRK: it needs something to look forward to over its lulls, and (imo) ideally that something shouldn't just be a simple hit-on-refresh (and not just because it'd be a combo-breaking GCD, either, as that just screws over downtime and most SkS tiers). After that, it ought to have more reason for varying the order of attacks in its bursts -- whatever reason that may be -- and more integral use of its resources, especially over its lulls.

Also never forget they took Dark Dance away, reworked it to not be evasion but just damage reduction, kept the parry, renamed it Camouflage, and gave it to a different tank entirely.

The saddest thing is that if ever I imagine some parry/dodge based counter-offensive mechanic, already I'll assume it should belong to GNB.