Velvet Resolver

On Monday, Celestalon kicked off the official Alpha Theorycrafting season by posting a Theorycrafting Discussion thread on the forums. And he was kind enough to toss a meaty chunk of information our way about Resolve, the replacement for Vengeance.

Resolve: Increases your healing and absorption done to yourself, based on Stamina and damage taken (before avoidance and mitigation) in the last 10 sec.

In today’s post, I want to go over the mathy details about how Resolve works, how it differs from Vengeance, and how it may (or may not) fix some of the problems we’ve discussed in previous blog posts.

Mathemagic

Celestalon broke the formula up into two components: one from stamina and one from damage taken. But for completeness, I’m going to bolt them together into one formula for resolve $R$:

$$ R =\frac{\rm Stamina}{250~\alpha} + 0.25\sum_i \frac{D_i}{\rm MaxHealth}\left ( \frac{2 ( 10-\Delta t_i )}{10} \right )$$

where $D_i$ is an individual damage event that occurred $\Delta t_i$ seconds ago, and $\alpha$ is a level-dependent constant, with $\alpha(100)=261$. The sum is carried out over all damaging events that have happened in the last 10 seconds.

The first term in the equation is the stamina-based contribution, which is always  active, even when out of combat. There’s a helpful buff in-game to alert you to this:

In-game tooltip for Resolve, out of combat.

In-game tooltip for Resolve, out of combat.

My premade character has 1294 character sheet stamina, which after dividing by 250 and $\alpha(90)=67$, gives me 0.07725, or about 7.725% Resolve. It’s not clear at this point whether the tooltip is misleadingly rounding down to 7% (i.e. using floor instead of round) or whether Resolve is only affected by the stamina from gear. The Alpha servers went down as I was attempting to test this, so we’ll have to revisit it later. We’ve already been told that this will update dynamically with stamina buffs, so having Power Word: Fortitude buffed on you mid-combat will raise your Resolve.

Once you’re in combat and taking damage, the second term makes a contribution:

resolve2

In-game tooltip for Resolve, during combat.

I’ve left this term in roughly the form Celestalon gave, even though it can obviously be simplified considerably by combining all of the constants, because this form does a better job of illustrating the behavior of the mechanic. Let’s ignore the sum for now, and just consider an isolated damage event that does $D$ damage:

$$0.25\times\frac{D}{\rm MaxHealth}\left ( \frac{2 ( 10-\Delta t )}{10} \right )$$

The 0.25 just moderates the amount of Resolve you get from damaging attacks. It’s a constant multiplicative factor that they will likely tweak to achieve the desired balance between baseline (stamina-based) Resolve and dynamic (damage-based) Resolve.

The factor of $D/{\rm MaxHealth}$ means that we’re normalizing the damage by our max health. So if we have 1000 health and take an attack that deals 1000 damage (remember, this is before mitigation), this term gives us a factor of 1. Avoided auto-attacks also count here, though instead of performing an actual damage roll the game just uses the mean value of the boss’s auto-attack damage. Again, nothing particularly complicated here, it just makes Resolve depend on the percentage of your health the attack would have removed rather than the raw damage amount. Also note that we’ve been told that dynamic health effects from temporary multipliers (e.g. Last Stand) aren’t included here, so we’re not punished for using temporary health-increasing cooldowns.

The term in parentheses is the most important part, though. In the instant the attack lands, $\Delta t=0$ and the term in parentheses evaluates to $2(10-0)/10 = 2.$ So that attack dealing 1000 damage to our 1000-health tank would give $0.25\times 1 \times 2 = 0.5,$ or 50% Resolve.

However, one second later, $\Delta t = 1$, so the term in parentheses is only $2(10-1)/10 = 1.8$, and the amount of resolve it grants is reduced to 45%. The amount of Resolve granted continues to linearly decrease as time passes, and by the time ten seconds have elapsed it’s reduced to zero.  Each attack is treated independently, so to get our total Resolve from all damage taken we just have to add up the Resolve granted by every attack we’ve taken, hence the sum in my equation.

You may note that the time-average of the term in parentheses is 1, which is how we get the advertised “averages to ~Damage/MaxHealth” that Celestalon mentioned in the post. In that regard, he’s specifically referring to just the part within the sum, not the constant factor of 0.25 outside of it. So in total, your average Resolve contribution from damage is 25% of Damage/MaxHealth.

Comparing to Vengeance

Mathematically speaking, there’s a world of difference between Resolve and Vengeance. First and foremost is the part we already knew: Resolve doesn’t grant any offensive benefit. We’ve talked about that a lot before, though, so it’s not territory worth re-treading.

Even in the defensive component though, there are major differences. Vengeance’s difference equation, if solved analytically, gives solutions that are exponentials. In other words, provided you were continuously taking damage (such that it didn’t fall off entirely), Vengeance would decay and adjust to your new damage intake rather smoothly. It also meant that damage taken at the very beginning of an encounter was still contributing some amount of Vengeance at the very end, again, assuming there was no interruption. And since it was only recalculated on a damage event, you could play some tricks with it, like taking a giant attack that gave you millions of Vengeance and then riding that wave for 20 seconds while your co-tank takes the boss.

Resolve does away with all of that. It flat-out says “look, the only thing that matters is the last 10 seconds.” The calculation doesn’t rely on a difference equation, meaning that when recalculating, it doesn’t care what your Resolve was at any time previously. And it forces a recalculation at fixed intervals, not just when you take damage. As a result, it’s much harder to game than Vengeance was.

Celestalon’s post also outlines a few other significant differences:

  • No more ramp-up mechanism
  • No taunt-transfer mechanism
  • Resolve persists through shapeshifts
  • Resolve only affects self-healing and self-absorbs

The lack of ramp-up and taunt-transfer mechanisms may at first seem like a problem. But in practice, I don’t think we’ll miss either of them. Both of these effects served offensive (i.e. threat) and defensive purposes, and it’s pretty clear that the offensive purposes are made irrelevant by definition here since Resolve won’t affect DPS/threat. The defensive purpose they served was to make sure you had some Vengeance to counter the boss’s first few hits, since Vengeance had a relatively slow ramp-up time but the boss’s attacks did not.

However, Resolve ramps up a lot faster than Vengeance does. Again, this is in part thanks to the fact that it isn’t governed by a difference equation. The other part is because it only cares about the last ten seconds.

To give you a visual representation of that, here’s a plot showing both Vengeance and Resolve for a player being attacked by a boss. The tank has 100 health and the boss swings for 30 raw damage every 1.5 seconds. Vengeance is shown in arbitrary units here since we’re not interested in the exact magnitude of the effect, just in its dynamic properties. I’ve also ignored the baseline (stamina-based) contribution to Resolve for the same reason.

As a final note, while the blog post says that Resolve is recalculated every second, it seemed like it was updating closer to every half-second when I fooled with it on alpha, so these plots use 0.5-second update intervals. Changing to 1-second intervals doesn’t significantly change the results (they just look a little more fragmented).

Vengeance and Resolve timelines. Boss hits for 30% of tank health every 1.5 seconds, no variation.

Vengeance and Resolve timelines. Boss hits for 30% of tank health every 1.5 seconds, no variation.

The plot very clearly shows the 50% ramp-up mechanism and slow decay-like behavior of Vengeance. Note that while the ramp-up mechanism gets you to 50% of Vengeance’s overall value at the first hit (at t=2.5 seconds), Resolve hits this mark as soon as the second hit lands (at 4.0 seconds) despite not having any ramp-up mechanism.

Resolve also hits its steady-state value much more quickly than Vengeance does. By definition, Resolve gets there after about 10 seconds of combat (t=12.5 seconds). But with Vengeance, it takes upwards of 30-40 seconds to even approach the steady-state value thanks to the decay effect (again, a result of the difference equation used to calculate Vengeance). Since most fights involve tank swaps more frequently than this, it meant that you were consistently getting stronger the longer you tanked a boss. This in turn helped encourage the sort of “solo-tank things that should not be solo-tanked” behavior we saw in Mists.

This plot assumes a boss who does exactly 30 damage per swing, but in real encounters the boss’s damage varies. Both Vengeance and Resolve adapt to mimic that change in the tank’s damage intake, but as you could guess, Resolve adapts much more quickly. If we allow the boss to hit for a random amount between 20 and 40 damage:

Vengeance and Resolve

Vengeance and Resolve timelines. Boss hits for 20%-40% of the tank’s hit points every 1.5 seconds.

You can certainly see the similar changes in both curves, but Resolve reacts quickly to each change while Vengeance changes rather slowly.

One thing you’ve probably noticed by  now is that the Resolve plot looks very jagged (in physics, we might call this a “sawtooth wave”). This happens because of the linear decay built into the formula. It peaks in the instant you take the attack – or more accurately, in the instant that Resolve is recalculated after that attack. But then every time it’s recalculated it linearly decreases by a fixed percent. If the boss swings in 1.5-second intervals, then Resolve will zig-zag between its max value and 85% of its max value in the manner shown.

The more frequently the boss attacks, the smoother that zig-zag becomes; conversely, a boss with a long swing timer will cause a larger variation in Resolve. This is apparent if we adjust the boss’s swing timer in either direction:

Vengeance and Resolve timelines. Boss hits for 20-40 damage every 1.0 seconds.

Vengeance and Resolve timelines. Boss hits for 20-40 damage every 1.0 seconds.

Vengeance and Resolve timelines. Boss hits for 20-40 damage every 2.0 seconds.

Vengeance and Resolve timelines. Boss hits for 20-40 damage every 2.0 seconds.

It’s worth noting that every plot here has a new randomly-generated sequence of attacks, so don’t be surprised that the plots don’t have the same profile as the original. The key difference is the size of the zig-zag on the Resolve curve.

I’ve also run simulations where the boss’ base damage is 50 rather than 30, but apart from the y-axis having large numbers there’s no real difference:

Vengeance and Resolve timelines. Boss hits for 40-60 damage every 1.5 seconds (20% variation).

Vengeance and Resolve timelines. Boss hits for 40-60 damage every 1.5 seconds.

Note that even a raw damage of 50% is pretty conservative for a boss – heroic bosses in Siege have frequently had raw damages that were larger than the player’s health. But it’s not clear if that will still be the case with the new tanking and healing paradigm that’s been unveiled for Warlords.

If we make the assumption that raw damage will be lower, then these rough estimates give us an idea of how large an effect Resolve will be. If we guess at a 5%-10% baseline value from stamina, these plots suggest that Resolve will end up being anywhere from a 50% to 200% modifier on our healing. In other words, it has the potential to double or triple our healing output with the current tuning numbers. Of course, it’s anyone’s guess as to whether those numbers are even remotely close to what they’ll end up being by the end of beta.

Is It Fixed Yet?

If you look back over our old blog posts, the vast majority of our criticisms of Vengeance had to do with its tie-in to damage output. Those have obviously been addressed, which leaves me worrying that I’ll have nothing to rant about for the next two or three years.

But regarding everything else, I think Resolve stands a fair chance of addressing our concerns. One of the major issues with Vengeance was the sheer magnitude of the effect – you could go from having 50k AP to 600k AP on certain bosses, which meant your abilities got up to 10x more effective. Even though that’s an extreme case, I regularly noted having over 300k AP during progression bosses, a factor of around 6x improvement. Resolve looks like it’ll tamp down on that some. Reasonable bosses are unlikely to grant a multiplier larger than 2x, which will be easier to balance around.

It hasn’t been mentioned specifically in Celestalon’s post, but I think it’s a reasonable guess that they will continue to disable Resolve gains from damage that could be avoided through better play (i.e. intentionally “standing in the bad”). If so, there will be little (if any) incentive to take excess damage to get more Resolve. Our sheer AP scaling on certain effects created situations where this was a net survivability gain with Vengeance, but the lower multiplier should make that impossible with Resolve.

While I still don’t think it needs to affect anything other than active mitigation abilities, the fact that it’s a multiplier affecting everything equally rather than a flat AP boost should make it easier to keep talents with different AP coefficients balanced (Eternal Flame and Sacred Shield, specifically). And we already know that Eternal Flame is losing its Bastion of Glory interaction, another change which will facilitate making both talents acceptable choices.

All in all, I think it’s a really good system, if slightly less transparent. It’s too soon to tell whether we’ll see any unexpected problems, of course, but the mechanic doesn’t have any glaring issues that stand out upon first examination (unlike Vengeance). I still have a few lingering concerns about steady-state threat stability between tanks (ironically, due to the removal of Vengeance), but that is the sort of thing which will become apparent fairly quickly during beta testing, and at any rate shouldn’t reflect on the performance of Resolve.

This entry was posted in Tanking, Theck's Pounding Headaches, Theorycrafting, Uncategorized and tagged , , , , , , , , , , , , , . Bookmark the permalink.

74 Responses to Velvet Resolver

  1. Thels says:

    Nice write-up. I personally am very glad to see Vengeance go sleep with the fishes, even though it was funny on occasion to top both DPS and Healing meters while tanking. :P

    I’m just a little confused about the Damage/MaxHealth formula. Doesn’t that mean that the more health we gain, the less self-sufficient we’ll be?

    Say if I go kill a group of mobs, and then flash heal myself, it’ll heal me for X. A year later in the expansion, I go back to kill the same group of mobs and then flash heal myself again, it’ll heal me for less than X? Don’t think it’ll actually be a problem, but it feels weird.

    • Talarian says:

      Ostensibly if you have more health you’ll also have more attack power/defensive stats to make up the difference in lost resolve.

    • Theck says:

      It’s a bit of a double-edged sword. More stamina increases your baseline Resolve, but reduces the rate at which your damage-based Resolve increases. So there’s a little positive scaling and a little negative scaling involved. I’m not actually sure which is the larger effect, since I don’t know offhand what stamina/health values will look at level 100.

      • Airowird says:

        If you want to compare values, I made a spreadsheet with gear & boss comparison you can find here: http://www.tankspot.com/attachment.php?attachmentid=3896&d=1402605717

        It’s originally intended to show how Resolve skews Shield Barrier vs Shield Block over the course of the expansion (as an example for all recovery vs mitigation abilities), but you can put in certain HP & stamina values and look up the Resolve a bit lower if you like.

        If you wonna read my entire rant about Resolve, it’s on Tankspot here: http://www.tankspot.com/showthread.php?82134-WoD-Versatility-Resolve-a-Damage-Scaling-Concern
        There is also a link to the official forums for more coverage ;)

        • Theck says:

          Regarding the excel spreadsheet: very nice demonstration of how using post-mitigation damage instead of pre-mitigation damage keeps them balanced more effectively. It’s a topic I touched on in this post:

          http://www.sacredduty.net/2014/03/03/vengeance-with-a-vengeance/

          “There are some ways to ensure this doesn’t happen, or at least to prevent the need to change k several times per expansion. The issue here is that the ratio of k/(1−A)(1−M) grows as A and M grow. So the logical solution is to let k vary the same way. The simplest way to do that is to use actual damage taken to determine Vengeance rather than raw damage. That introduces a factor of (1−A)(1−M) in the numerator, which automatically corrects for variations.”

          I know they’re hesitant to do so, because it feels like a punishment to have your active mitigation and avoidance reduce your Resolve. Though as I said in the post, that’s really not an issue from a survivability standpoint – it can be tuned such that adding survivability stats increase your survivability more than Resolve reduces it. Especially if they continue to give Resolve on avoided attacks. It’s more a psychology limitation than a numbers one.

  2. Tengenstein says:

    I really enjoy the blogs, so rather hope you can find subjects to continue talking about. But I do have one topic I’d like to see you explore; Versatility and how it affects active mitigation choices.

    I’m concerned that with one part of it multiplicatively increasing our WoG (or in my case Shield Barrier), and the other multiplicatively decreasing the damage we take and thus the raw amount of damage stopped by your ShotR (or my Shield Block), that we have a stat that will push us into just using WoG/SBar.

    • Theck says:

      It depends on the magnitude of the effect. Note that if we limit our analysis to physical damage, armor functions essentially the same way on live. But the numbers are tuned around the two options being competitive when your armor is mitigating 50-60% of the damage.

      It’ll be likewise with Versatility. If the options are balanced around 0%, then it may very well become unbalanced at 30%-40%. On the other hand, maybe we won’t be able to stack that much Versatility, or maybe they’ll be balanced around having about 20% Versatility (so that they don’t get too out-of-whack).

      In addition, keep in mind we’re simplifying a lot here. WoG has a natural restriction since you can’t WoG yourself at full health, so I’m not really concerned about it for paladins. Shield Barrier doesn’t have the same restriction, so the discrepancy may be larger for warriors.

      And that’s all assuming Versatility makes it live at all – it may get dumpstered between now and release just like Readiness and Amplify did.

      • Airowird says:

        The problem I have here is that Versatility is pretty much the statification of gear progression. It’s pretty much a passive and more boring Amplification, but more importantly, it shows the impact gear has on Resolve and the balance between percentual damage reduction effects (ShotR & co) and ‘recovery’ abilities (WoG, SBar and all other self-heals and/or absorbs)

        • Theck says:

          Yeah, it’s definitely not an exciting stat. With the way it’s being designed, my suspicion (or perhaps hope?) is that it will be everyone’s dump stat.

          I’d rather see them dumpster it and find new ways to make more item combinations. Like, for example, having multiple split ratios of the two secondaries (50/50, 65/35, 80/20) or allowing three secondaries on an item.

          • Airowird says:

            I actually thought the Amplification part of increasing crit damage (perhaps also Multistrike damage?) is more interesting as a design, as it slightly shifts the balance between stats as you gear up. Versatility is just the ICC debuff poured into a stat. Maybe if the stat increased crit/MS damage, armor & major stats by a small amount it could become interesting? DPS & healers would get the extra output as they do now (+ armor as reduction), while tanks would get mostly armor + str/sta from it, with a bit of damage as ‘bonus’.

            The problem with a skewed & triple stat items is that due to the way ilvl & stat allocation works, an even spread of 3 stats simply gives more overall secondary stats to a point where it is bis regardless of what stats are actually good for you. You can already see now that 40/60 secondaries on items have a lower total than an evenly split one.

          • Theck says:

            Amp had its own problems, though. A “crit damage” stat (ala D3, other games) would work too, but then you get a strong synergy between crit chance and crit damage which basically means stacking both is always the best situation for everyone (not interesting).

            The issue with item stat allocation is easily fixed by changing one number in the item stat allocation formula so that it’s less punishing. Or removing it altogether and making the process linear (iirc it’s an exponential function atm).

          • Airowird says:

            Right now Crit, Haste and Mastery are all multiplicative already. And at some point in D3, 1% Attack speed actually becomes better than 1% crit, because CD &CDC improve its value so much, while they don’t improve themselves (and D3 crit damage is balanced on being 300+%). As long as Amplification works a similar way, you will want a balance of all stats. As reference, you’re nothing with 70% of the current Amp if you sacrificed all your crit/haste/mastery for it. Someone might want to settle for a 20%crit/50% amp balance, while others go for 55%/15%, depending on procs & effects, but that just makes gear interesting!

            I had some random thoughts on this here:
            http://www.tankspot.com/showthread.php?82134-WoD-Versatility-Resolve-a-Damage-Scaling-Concern&p=572182#post572182
            More concepts than implementation-ready mechanics, but they might trigger an idea in someone :)

            Stat allocation could indeed be improved upon, as the old formula was to encourage you from going for a single stat (e.g. 100 crit rating vs 80 crit & 80 haste is an obvious choice, even when stat balance is out of whack), but should they succeed in their efforts to bring stats closer together in values, this might be worth looking into.

            Not convinced about the different stat balances though, wouldn’t that basicly be some sort of Reforge, where you basicly say “Here is an item with 80 crit & 80 mastery, you can add another 40 to whichever of those 2 you want, so you end up with 200 total” ?

          • Theck says:

            Even though they’re all multiplicative, it takes a lot of CD+CC to reach the point where IAS is valuable enough to be desirable in D3, if my understanding is correct. I realize it’s also balanced that way, but the concept is that the feedback loop between those two is much more sensitive than the feedback loop between e.g. crit and haste.

            I think a lot of that can be solved by balancing numbers, or starting with a lower crit damage multiplier (i.e. crits do 125% dmg instead of 150% or 200%, but then are increased by crit damage). It’s just a question of whether it’s worth the work and complexity it creates in what players should be ideally gearing for. In some ways, the simple “haste/crit/mastery” system is a little nicer in its simplicity.

            Regarding the different stat balances (and to be fair, this isn’t my idea, it was something Landsoul mentioned in a conversation, so he should get credit), the idea is to provide more itemization options. One of the reasons given for adding Versatility is that you only have 6 permutations of items with “4 choose 2″ secondary stats. Going up to “5 choose 2″ gives you 10 permutations, but at the expense of adding a stat that could be boring/problematic (Versatility).

            But another way to get more permutations is to be more discrete about the split between the 2 chosen secondary stats. For example, a haste/mastery item nowadays might have a 50/50 split of the two stats. It varies somewhat from item to item (one item might be 55/45, another might be 51/49, etc.) for variety, but there’s generally only one haste/mastery option in any given slot.

            The idea here is to codify that more discretely. For example, within a tier, maybe there’s a 70/30 haste/mastery option and a 40/60 haste/mastery option (again, assume that the formula is relaxed so that neither of these is significantly below the 50/50 item in terms of total rating). A class that likes haste > mastery would prefer the 70/30 item, but the 40/60 item might still be a solid upgrade. Likewise a player that prefers mastery would prefer the 40/60 item.

            This doubles the number of possible itemizations the item team has to work with. Instead of one haste/mastery item, they can now easily put in two that are distinct in such a way that each is more attractive to certain classes. As long as e.g. ret paladins want haste>mastery, fury warriors want crit>haste, and so on, each class would prefer different items but would still use the “non-ideally itemized” ones.

          • Airowird says:

            The D3 crit issue is also because the stats can hit far higher values then we’ld expect to see in WoW. I doubt one would expect to be raiding with 60% crit and +350% crit damage at some point (imagine the Chaos Bolts!!).
            Consdering you could probably balance Amp value relatively well up to 20-25%, the slight double-dip effect from crit would be countered by the fact other effects inherently part of that same stat. e.g. a prim stat increase doesn’t amplifiy weapon damage (or SP from weapon) like crit,haste & mastery do right now. So it could be made to be fairly linear within the expected range (and a bit beyond, because expectations work out so well).

            Like I argued for the Readiness effect (see Tankspot), if a newer Amplification is a collection of small increases in mechanics overall, it is far more stable, because the weight of the stat is not carried by a single granular effect, nor an exponentially scaling effect. Surely I don’t need to tell you that a distribution of different scalings is far less likely to shift towards either side of the stat value spectrum than a single stat, as that’s just how distributing effects work.
            I know that the current Amplification is nothing but a collection of exponential increases and as such it would need to changed around on some parts to make it more balanced before it can truly become a secondary stat.

            I guess what we need is a table of blues & T17/18/19 expected stats. What I wouldn’t do for a blue just straight up telling us “In heroic blues, you should have X stamina & Y prim stat and weapon should add Z AP/SP. You will have ~2000 secondary stats total, with each ranging from 300 to 900, depending on your gear choices. In the next Tier, this should be …”

            Also, your stat shifts are an interesting conceptual idea, but in the end, it’s an illusion of choice. You will still value an item in its total based on your DPS increase (or HPS or TMI). Your 40/60 Haste/Mastery item might as well be a 70/30 Haste/Crit item, if it gives players the same metrcs in the end. Both changing the stat distribution and adding Versatility create extra choices for gear creation, but with just stat distribution you will still want to go for that maximum stat balance. That might be 70/25/5 Mastery/Haste/Crit or 30/55/25, but in the end, you have enough gear slots already to balance out stats as you like. Atleast Versatility would add a extra dimension to your stats and make it harder to find the point closest to that “free choice” optimal point or, through it’s extra benefits, make you find a balance between survival and output.

            But Versatility itself is such a boring stat, outside of the recovery ability boost for tanks (which is virtually the same as just +AP), it changes nothing in how you look at the other stats or use abilities.
            Atleast Mastery for specs such as Frost DK or Enh Shaman change in value purely on a fights mechanics. “Hey, Haste sims a bit better for single target, but Mastery works better on adds or if I need to move away, so maybe it’s actually good for bosses 2,3, 6,…” is an infinitely more interesting mechanic than just “Hey, my numbers are bigger”.

            PS: I also think that the buff distribution for Versatility is completely fooked. The only ones bringing it are Plate DPS and Druids, which means that in WoD a 10man guild that doesn’t run plate DPS (like we currently do), will flatout do 3% less DPS and require 4.5% more healing output when the Druid is taking a night off. I would expect classes with more Versatile spec choices such as Monks or Shamans to also bring it, if only so you it’s likely your healer team will provide the buff that probably matters most of all to them.

            Damn, that turned to be a long “short reply”

          • Theck says:

            Quoting you:
            “Also, your stat shifts are an interesting conceptual idea, but in the end, it’s an illusion of choice. You will still value an item in its total based on your DPS increase (or HPS or TMI). Your 40/60 Haste/Mastery item might as well be a 70/30 Haste/Crit item, if it gives players the same metrcs in the end. Both changing the stat distribution and adding Versatility create extra choices for gear creation, but with just stat distribution you will still want to go for that maximum stat balance. ”

            I don’t agree.

            Your rationalization for why different stat shift combos only gives an “illusion of choice” applies just as well to adding another stat. It doesn’t matter whether your choice is between a 70/30 haste/crit item and a 60/40 haste crit item, or between haste/crit and haste/versatility items. In both cases, the player will want the item that gives the largest power increase.

            What matters at the end of the day is, e.g., “how many distinct plate chest pieces can we create for players to choose from?” It matters very little what method you use to provide those distinct options, just that the options exist and that players are still willing to use the “non-ideal” choice (again, because it’s enough of a power increase to warrant doing so).

            Adding a stat that ends up being a universal “dump stat” is worse, IMO, than just providing different stat/split combinations of existing stats they do want.

  3. Talarian says:

    Hmm, I’m concerned about the sawtooth graph. For a boss that’s swinging only once every 2 seconds, that’s nearly a 35% – 50% swing in Resolve in that 2 second period. I get that timing your abilities is an important aspect of high-level play, but that seems a bit problematic if your heals hit for like 20% less if you miss that tiny window. Of course, if they tune Resolve to be more dependent on health than incoming damage, then it becomes moot, but I don’t see that happening. Vengeance’ smoothing made it a lot less timing-dependent in terms of ability usage.

    • Theck says:

      It’s not anywhere near that size. Resolve decays over 10 seconds, so in 2 seconds it drops to 80% of the value it started with. If you started with 100% Resolve (from damage only) immediately after the hit, it would drop to 80%. So it’s only a 20% fluctuation by definition, and less than that if you add the stamina-based contribution.

      I do agree that it encourages a little bit of very tight timing micromanagement, but ultimately I think the effect of that will be pretty small. And if it really only updates once every second (it’s possible that I’m mistaken about the 0.5 second update interval, or that 1.0 is the new value in a new build), then it’s really about a 1-second window you need to hit, which isn’t too bad.

      • Zambar says:

        Is that really going to be a problem?

        If we talk about keeping an hot up 100 %, with snapshoting gone you dont really care about that.

        If we talk about an emergency heal because you took a big hit and are at 20% health, well then you usually click your heal before even 0.1 seconds have passed :D

        • Theck says:

          Yeah, it only matters for the initial WoG cast; the lack of snapshotting makes it irrelevant for EF. I think most players have a reaction time longer than 0.1 seconds, but probably still less than 0.5. So I don’t think it’s a big deal.

          • Gupsy says:

            Double latency + lag could be a problem in some places around, out of US at least :D
            Not a big one in this case. But – as I’m usually playing around 300ms latency in raid environment (it means sometimes >800 peaks) – I believe it’s something deserve some thought. Anything amplifying latency difference issues is a minus, as this is a worldwide game.

          • Airowird says:

            While you could somehow make Resolve a flat number for 2s before falling down, it would not help for the Resolve gained from previous hits, which will still be a significant portion of the decrease. I know a slow connection can be an issue, but with the updates already being every 1s + damage, I simply can’t see a fix for latency that would not negatively influence the game play in it’s totality.

            And thinking of when exactly you use your abilities, absorbs will be an interesting dilemma when it comes to timing, as they are most powerful right after big bursts, but you’ll want to use them right before …. Hmmmm, I wonder…

      • Hamsda says:

        I think that Celestalon also wrote that Resolve will be recalculated on every damage taken event, which might explain the more frequent updates you have seen, though I don’t know how you tested it.
        On the other hand I might have just imagined that part :X

        • Theck says:

          Yeah, it does, which is what makes it tricky to test. I was letting a mob hit me and then stunning it with Fist of Justice and trying to observe the tooltip update rate. But I was just eyeballing it at first, and then the alpha servers went down so I couldn’t continue testing with a stopwatch.

  4. Theck says:

    Update: Just did some quick testing on alpha regarding the stamina contribution. It’s definitely using the full 7.7%, so the tooltip is just calculating using floor().

  5. Airowird says:

    Theck, while you compare Resolve ramp-up to Vengeance, you should also show the power-down effect!
    As you said, Resolve builds AND decays faster, so while right now you could tank swap out and keep healing through your X stacks of dangerous DoT, this may not be true with Resolve, thus only shifting the healing required from one tank to the other on tank swaps.

    • Theck says:

      Very good point, though note that DoT ticks still give Resolve, so it won’t decay away to nothing. In theory, a really hard-hitting DoT will still be fueling your Resolve, though probably not enough to single-handedly heal through it (like was often the case with lingering Vengeance). I think that’s sort of the intent though – less tank self-sufficiency.

      • Fetzie says:

        Don’t forget that abilities that ignore armor (like more DoT effects) give 2.5 times as much resolve, so a heavy-ticking DoT could well contribute more Resolve than a boss’ melee swing.

        • Theck says:

          I hadn’t forgotten. :) My point was that with Vengeance, a hard-hitting DoT could give you enough Vengeance to out-heal it thanks to the silly scaling of Vengeance & EF. That is almost certainly not going to be the case with Resolve, because the magnitude of the healing bonus is smaller.

  6. Homsel says:

    What’s your opinion on the new role of mastery as a dps boost for all tanks with these new mechanics? Do you think, as a result of quite delicate balancing issues, progress guilds (especially for the first tiers) will tend towards more mastery oriented tank classes? (Or maybe the other secondary stats will make up for this, most notably haste for us paladins.)

    • Theck says:

      I think it’s necessary to prevent the reverse from happening. If mastery didn’t provide DPS, then tanks that prefer crit/haste would be doing more DPS than tanks that prefer mastery. The only solution to that is to make sure mastery gives a DPS boost that’s roughly equivalent to an equal amount of haste or crit rating.

    • Çapncrunch says:

      Yeah, the whole point was to make it so that all secondary stats provide both a survival and dps benefit for tanks. Crit, Haste and Multi-strike all inherently increase dps, so they had to give mastery a dps benefit as well.

  7. woodknight says:

    Hi Theck, I am very curious about the title, what does ‘Velvet Resolver’ mean?

  8. Çapncrunch says:

    On the subject of threat stability between tanks, as you said this ironically becomes a problem due to the removal of the offensive component of vengeance. Not surprisingly this reminds me of some of the reasons for why vengeance was even put in the game in the first place: originally the intent was to allow tank dps/tps to scale over the course of the expansion in order to allow them to keep up with the dps/tps scaling of the dps specs (also worth noting that during Cata the scaling of vengeance was based on our stamina rather than on boss damage, mainly because the vengeance cap was basically a given on any fight, so it was a gear-based dps/tps increase rather than the laundry list of issues that vengeance has had during MoP, in a lot of ways Cata Vengeance was just so much better than MoP Vengeance).

    Anyways, my point is that in terms of stabilizing threat Vengeance was a good idea, once you established aggro it was relatively easy to maintain it. And I think this idea is also something they could continue to use in the future. What if Resolve continued to increase our threat, without affecting our damage done? It could act as a threat multiplier to help ensure that accidental loss of threat isn’t a constant concern. Then it’s just a matter of making sure that the ramp-up and decay of resolve is fast enough to stabilize threat within the duration of a taunt.

    Or another option (same effect but different means) would be to make each tank’s passive threat bonus also be modified by damage taken. I would say to just merge them directly into Resolve, but I’d think that our threat bonus would need a larger baseline value and smaller scaling component compared to Resolve’s smaller baseline amount and relatively larger damage-based component. Also it could have a different (faster) ramp-up/decay rate than Resolve in order to make threat more responsive during taunt-swaps. Like maybe Righteous Fury (defensive stance, blood pres, etc) only increase our threat generation by 300% baseline, but when taking damage from a boss it can scale up to 500%.

    tl;dr, they could continue to allow our damage intake to influence our threat generation without creating the dps issues that vengeance had.

    • Theck says:

      Yeah, there are a number of ways to accomplish that. My suggestion was that taunts apply a mutually exclusive 5-minute “Extra Mad” debuff that increases threat generation against that target by 10%-30%. That way whoever taunted automatically generates a little more threat to overcome tank DPS discrepancy caused by class or gear.

      That said, a mechanism does exist to overcome this: chain taunting will continue to apply the 3-second Fixate buff that increases threat generation by a factor of 3, so you can keep taunting every 8 seconds to stay ahead on threat. I think that’s sort of a lame solution, but it does work (ignoring taunt DR, I guess).

      Once we can test things in beta I think we’ll see exactly how big a problem this is, but I wouldn’t be surprised if something gets implemented to help fix it.

      • Çapncrunch says:

        That “Extra Mad” debuff ain’t a bad idea. They could probably actually replace the 3 second threat buff after a taunt with that. The main reason for the 3s buff was because of vengeance’s effect on damage output, but with tank dps stabilized we don’t need a temporary boost anymore, all we need is a bit of a boost to our sustained threat.

        I really don’t like the idea of having to taunt on cd to keep from being pulled off of, has too much of a possibility of leading to accidentally causing the boss to go taunt-immune. Taunts by design are only intended for pulling aggro off of someone else, not to maintain aggro (I know some use things like taunt+bubble to maintain aggro while bubbling a debuff, but that’s a niche case and a whole ‘nother can of worms regarding bubble as a tank). Plus, like I said with our dps stabilized the 3s buff doesn’t really serve a purpose anymore, so trying to use it to solve a different problem seems kind of lazy.

        And yeah, thinking on it, we don’t really need threat generation to scale with damage intake, far more complicated than it needs to be to solve the issue at hand. So yeah my new suggestion is to scrap the 3s threat buff after taunting and replace it with your “Extra Mad” debuff.

      • Airowird says:

        “Extra Mad” – You just insulted his mother! Oh my …

  9. Gupsy says:

    I can’t understand what is the “use” of resolve. It doesn’t address any problem, like vengeance did (it compensate the threat problem of dps scaling faster then tps, the tank too viable to pvp etc.). The important part of vengeance for the game mechanics is the damage increase, the “super + to auto heal” is just an error they did in MoP, it address nothing.
    Now we will have back all the problems vengeance was introduced to solve plus the problem it introduced with MoP vengeance, plus the new ones this “the most stamina you have the worst it works” (the risk is a healer will have much more problem healing a Mythic dressed tank than a lfr tank; equipped tank will probably die too fast, but would take less damage).
    Again, what is this “resolve” for? Why not to go back to the old times (possibly given tank some more base dps :p), removing this new mechanics that improve nothing?

    What am I missing?

    • Zambar says:

      Well the dmg part(and ofc tps) is nothing that needs to be solved by resolve since in wod all defensive stats give all tanks offensive value aswell. Because of this tanks tps should not lag behind as much as in earlier expansion.

      Regarding the need for tanks selfhealing to scale with incomming dmg, I think it is purly down to the fact that if they tune our selfhealing to fit a weak hitting boss then our selfhealing will be more or less worthless on a hard hitting one and ofc if they tune it to hardhitting they the weaker one will just be lol…

      I have to say it is interesting that resolve got one part that goes up the more health(stamina) a tank have and one part that goes down the more health a tank have. Is the first part of resolve formula going to be enough to not encourage tanks to have as little stamina they can to get the most out of the second part of the formula?

      • Airowird says:

        I’ve done some napkin-ish math (syou can try out yourself if you like, spreadsheet link is up higher) and here is a rough idea as an answer:

        For Stamina to be an increase in Resolve, the boss damage needs to be so low you can pretty much absorb/heal EVERYTHING with just the base ability.

        In other words, yes, having better gear will reduce the survival you have, which makes sense, because the word ‘Resolve’ indicates you’re making an extra effort against more challenging opponents.

        PS: This also counters the reasoning as to why Resolve should not be pre-passives/armor, as the major argument against it was the “feel weaker on gear upgrades” that defended the current Vengeance model.

        TL;DR: You will have a reason to use DPS over Stamina trinkets on farm content.

        • Gupsy says:

          I don’t think you got my point. I know I’m so bad at expressing in English that quite for sure is my fault.

          My post was all about design choice behind it. It means nothing it is coherent with the meanings of a word, or that other mechanics are addressing problems the removing of vengeance introduced.

          A partial answer is where Zambar says: “…if they tune our selfhealing to fit a weak hitting boss then our selfhealing will be more or less worthless on a hard hitting…”
          But it is partial, as there are so much simpler things you can do to address it not involving such a complicate and troubling solution (ie simply working on % of hit instead of raw values, or a low healing for lol content to be soloed plus a % absorb on another ability for hard content, there are plenty of nice solution that address this and more without drawback).

          I think I found an answer (not the answer ofc!), and posted it in http://www.tankspot.com/showthread.php?82134-WoD-Versatility-Resolve-a-Damage-Scaling-Concern&p=572308#post572308 as all this is not strictly related to theorycrafting, and I fear is a bit off topic here, where Theck is working so hard on numbers :p

    • Theck says:

      @Gupsy: Vengeance in Cataclysm was designed to address a tank threat scaling issue. What you’re missing is that it hasn’t really served that purpose since the redesign in the beginning of MoP. Boosting the threat modifier to 500% (and later 700% in MoP, and now 900% in Warlords) pretty much made tank threat vs DPS threat irrelevant.

      The primary purpose of Vengeance in MoP was to balance percent-based active mitigation (i.e. SotR) with point-based active mitigation (i.e. WoG). SotR naturally scales up with boss damage – it always blocks e.g. 40%, no matter how hard the boss hits. If WoG just healed for a fixed amount, it would almost always be superior to SotR for weak-hitting bosses and almost always vastly inferior on hard-hitting bosses. MoP Vengeance served as the scaling mechanism to make both worth using. It also performed the secondary purpose of bringing tank damage up in raids.

      In WoD, Resolve takes care of the Active Mitigation scaling. They’re bringing baseline tank damage up significantly across the board to cover the second part of what Vengeance did, which will be nice because we’ll be able to solo more effectively.

      @Zambar:
      The stamina-based part of Resolve is by far the weaker part, and I think that Airowird is correct that increasing stamina will generally reduce your Resolve in a boss encounter situation.

      However, keep in mind that your maximum or average Resolve value isn’t a good estimator of your survivability. Stamina traditionally gives a large survivability contribution, both in terms of EH and spike survival. If our self-healing capabilities are being severely curbed in Warlords, having more health may well be a bigger survivability increase in both areas despite slightly reducing our Resolve.

      @Airowird:
      They’ve said that they don’t want Resolve to feel weaker as you get gear upgrades, and while Stamina does in fact do just that, it’s probably a minority player here compared to armor & avoidance. In other words, even though getting more stamina will make your self-healing through Resolve weaker, it probably isn’t a very large effect. Post-mitigation damage, while making a lot of sense, might be a much larger effect depending on how it’s calculated (almost certainly if you’re ignoring avoided attacks, less certainly if we’re just talking about armor mitigation).

      And stamina is generally a pretty good survival stat anyhow, so a slight negative scaling via Resolve might not matter much in the grand scheme of things. Once we have more information about L100 gear and boss damage, we’ll be able to sim these sorts of things out and look for potential problems.

  10. Çapncrunch says:

    On the subject of Resolve’s negative scaling with Stam.

    I think this is actually a necessary thing. I know, I know, I must be crazy, right? But hear me out.

    I was considering why the normalization was in there in the first place. Now naturally they need to scale the raw damage down to some meaningful number to use as a multiplier. Also it needs to scale in a way so that lower-level content doesn’t give meaningful multipliers to higher level players (ie Garrosh should give meaningful amounts of resolve to level 90 tanks, but not to level 100 tanks). Now, they could use a level-based constant like they do with other things like stat-ratings, which would accomplish that goal without causing resolve to scale negatively with gear within the same level.

    However, as I was thinking about this it occurred to me that this might not actually work, and that resolve might actually NEED to scale negatively (at least a bit) with gear. See all of the comments regarding the negative scaling of resolve all leave out an important factor: resolve isn’t the only thing that determines how strong our self-healing is. Our self-healing also scales with our gear (and since our dps is supposed to scale much more closely to that of dps specs, then our self-healing should scale in the same way).

    So if Resolve wasn’t normalized to a level-based constant, then at the start of the expansion we’d have less ap, less haste, less mastery, less multistrike, less everything, which means our self-heals would be relatively weak. Bosses would also hit kinda weak so resolve would be low. So we have weak self-heals with a low multiplier. By the end of the expansion we have more ap, more haste, more mastery, etc which means our self-heals would be stronger. Bosses will also be hitting much harder so resolve would be high. So we’d end the expansion with strong heals with a large multiplier. It’d basically be an exponential growth. So the health-normalization basically acts to keep self-heals from scaling out of control. Sure that new upgrade has more stam on it which means resolve will be lower, but it also has more strength and more haste/mastery/multistrike/etc which will increase healing to compensate (and that’s before even considering the other survival benefits of the stam and other stats beyond self-healing).

    TL;DR: I think the whole point is to help keep self-healing’s relative power mostly stable throughout the expansion. Instead of scaling to the point of self-sufficiency like it did in MoP.

    Now this does bring up an interesting question that Gupsy also brought up: Do we actually *need* resolve? Self-healing and absorbs can potentially be balanced to scale with gear enough to keep up with %-based mitigation over the course of the expansion. However, this would mean that when gearing up within a single tier/difficulty (ie boss damage is static, say if we’re only looking at the final boss) self-healing would continue to improve as gear improved, but %-based abilities would stay basically the same. Then when moving to the next tier (so boss damage goes up, but at first gear is the same) %-based abilities would get “better” while self-healing would remain the same, then again as gearing up withing that new tier %-based abilities would again stay the same while self-healing improved.

    But with Resolve, and with it being normalized against health they could potentially keep self-healing more stable within a single tier/difficulty (so the increased healing from stats is canceled out by the reduced resolve due to health), just as %-based abilities are. Then when moving to the next tier/difficulty the increase in boss damage will cause self-healing to jump up in a similar way that %-based abilities do, and then again both will remain mostly stable again.

    (I know this part I sort of simplified things down as if each tier/difficulty was only a single boss, but hopefully you get my point regarding the impact of increasing gear against the same boss as well as the effects of moving from one boss to another with the same gear, and how health/gear normalization helps cause both self-healing and %-based abilities to behave the same as eachother in both situations).

    • Airowird says:

      So, we totally forgot about something …. Attack Power

      Generally, without a lot of sockets available, Stamina mainly comes from gear, based on item level.
      The same gear also brings, based on the same item level, an amount of Strength with it. So how does gear progress work on Resolve?

      Let’s simplify the mechanics to this:
      Resolve = Stamina / Q + R / Stamina (R = Resolve from boss, based on boss raw DPS)
      Attack Power = Strength
      Stamina = G * Strength (trinkets excluded, G can vary slightly depending on class’s +X% Stamina passives)
      15 ilvl increase = 15% more Strength

      All abilities scale almost purely (base of 1 dmg/heal, woop!) with Attack Power. Because of this, the Mastery AP increase can be ignored for the Resolve discussion, as it is just multiplicative. I will also ignore scalars for abilities, because that is just tuning (again, the point here is to see how Resolve & gear stats work together)

      So to poor it all into 1 formula:
      Heal/Absorb = Attack Power * (1 + Resolve)
      = Strength * (1 + Stamina / Q + R / Stamina)
      = Strength + Strength * Stamina / Q + R * Strength / Stamina
      = Strength + Strength * Stamina / Q + R / G
      = Strength + Strength ^ 2 * G / Q + R / G
      = Strength ^ 2 * G / Q + Strength + R / G

      ->> So looking at this, we have 3 parts, let’s start at the back:

      R / G
      This means that the Damage part of Resolve is only affected by a fixed (per class) scalar. In other words, Recovery abilities (should) scale linearly with incoming damage. (more at the end)

      Strength
      Obviously, without Resolve, your abilities should still scale, so this is working as intented.

      Strength ^ 2 * G / Q
      This part is relatively small, because Q = 65250. Bringing this one together with the one above, you’ld see:
      Strength * (1 + Strength * G / Q). For that last part to be even a 10% influence, you need about 4500 Strength. Something that is unlikely to happen even if we went up to Tier 20 this expansion.

      So while we will see Resolve go down slightly as you gear, your actual abilities will still be doing the same amount of damage recovery. The exception here is Stamina trinkets (and gems), but in my opinion this is a trade-off for having the extra health as buffer for your healers.

      I also want to point out that for most tanks, not only does Mastery increase this power through AP, usually Crit, Haste and Multistrike do as well either through procs or generally allowing for more resource generation. This means that your power increases just as much as a healer’s will. e.g. if healers are doing 50k HPS and you’re doing 10k without Resolve, we can expect that by the time they hit 60k HPS, you will do 12k. On top of that 10/12k, you will recover an extra amount of damage, but this is linearly related to boss DPS. In this way, Resolve accomplishes its goal perfectly. Active mitigation both through damage mitigation/reduction abilities and heal/absorb recovery abilities scales (almost) perfectly with DTPS.

      There is only a single differentiator left: Armor
      Tank Armor is a passive, guaranteed, scaling reduction on boss damage. It is a major part of your gear improvements, and in its current state, changes the balance between R and actual DTPS too much to allow for a stable environment. Right now, R ~ DTPS / (1 – armor_reduction). That means, at 50% Armor, R is linear with 100/50 = 2 * DTPS, while at 55% this is 2.222 * DTPS, a 11% increase in Resolve ~ DTPS ratio. That means an upgrade might make you take 100 DPS less on the same boss, but make you require 130 HPS less. This, I simply cannot watch get implemented.
      R ~ DTPS is already a fixed ratio for magic damage, it should be the same for physical, or you’re limiting boss design, pure and simple.

      Can this be compensated? Sure, bosses can hit slightly harder (and then a wee bit more, because it gave you extra Resolve), but as Armor goes up, so does the percentage of DTPS you can survive on your own and this is exactly one of the issues with Vengeance right now. Maybe we won’t be hitting the level of self-sustaining tanks we do now, but if you heal enough yourself so that a Mythic group can use 4 healers instead of 5, that’s a lot of DPS gained.
      And yes, Bonus Armor, which you’ll find as much as your healers find Spirit will be slightly reducing Resolve. But being the only guaranteed damage reduction from a stat, that seems a good balance, especially considering Blizzard has already indicated that bonus armor will be worth taking over other stats (and they tend to not look at stuff like TMI, but just at overall reduction)
      The only major issue I see here is Guardian Druid Mastery … but as a single spec’s Mastery, that (to me) seems a far smaller problem than the scaling of a core mechanic that impacts group composition, and thus boss design as a whole. Especially when changing the Mastery to grant X% Armor and 2*X% extra self-healing is nearly-perfect compensation and relatively simple to implement.

      • Gupsy says:

        @Theck, yup you are right, but I’m not missing it, as I said imho it was a clear design error, I’m not interested in it. They used an already in-game mechanics, designed to address a precise issue, to address something completely different. Probably designer had been a bit too conservative. I think MoP was by far the worst expansion WoW ever had. Just hoping they are not going to repeat the same errors in WoD, as i really love this game.

        @Çapncrunch: I agree, I think it is the same idea I wrote in my post at Tankspot. It can be a bit counter-intuitive, but with fine balance it can be great to help with the hard step we have at the beginning of each new tier.

        @Airowird: too much numbers for me :D I’m looking for the “why” it is there, not on the “how it works”. Theck did it for us already. As this is a game, not reality, the why is not confined to a theological discussion. I think understanding the row scope of a low they introduce in a game is an important things, as it can help us focus on the important part of it, and not just flame on something we didn’t really understand. Anyway for sure to understand it we need to analyze the mechanics in the global contest, and this is your objective in this post if I understood it right. Summarizing it we can say: resolve is a mechanic that smooth a bit the iLevel slope from one tier to the next. And this is its uniqueness. At the same time they count on it to cover the issue removing MoP vengeance bring back: balance the percent-based active mitigation with the “fixed one”.

        After this I can say my bigger concern about it changed. It is not about the retro-scaling of resolve, it is all in the weak mechanics of this balancing. It was not working in the whole MoP expansion, almost a “lol” at the end (worst then the “too much avoidance stacking” at the end of TBC, and it was really a huge and lollable mistake). Balancing it is all related to balancing AP scaling, so it is tightly coupled with a general balance issue. My fear is it introduces more constrains on this, making it harder to achieve (specially class wide), while in the meantime it amplify any given difference at any point of progression. Always consider that as players aren’t on the same progression point a nice balancing implies you have to balance a curve not a point!

        • Airowird says:

          If we want to talk about “why”:
          Resolve provides extra healing based on damage taken without making the heal feel useless when not taking damage. Number-wise, the math was just to show that if you keep Strength and Stamina on gear at a same rate (something that has been true for all blue/epic gear so far), gaining gear upgrades will not lower the number you see popping up. So at no point will the end result feel weaker when you gain more gear.
          I also think that Resolve should be an extra indication as to how challenging a boss is to someone IN THEIR CURRENT GEAR. Similar to how Berserk timers give your DPS an indication as to how their total output is changing during raid progression.

          On balancing:
          AP scaling is not a major issue. AP will only ever come from your gear’s base Strength/Agility, similar to how your healers get SP from Intellect. Because nearly all your Stamina will come from the same gear, we can consider those together as a single package. Without the base Stamina Resolve, all tank recovery abilities can essentially split into 2 independant parts: One that scales only with gear and one that scales only with incoming damage, never both. The part that scales with gear goes up at the same rate your healers’ heals go up, so this is pretty much “As long as you are playing right, you gain 10-20% more healing”. The second part works exactly as your Shield/Avoidance mechanics do, reduces X% of the damage you take. (Assuming they change the armor thing)

          The thing is that right now, boss DPS needs to scale upwards based on the following:
          * Healer overall HPS = Healer gear
          * Tank’s Active Mitigation uptime = Tank gear (e.g. more HoPo already is more reduction!)
          * Tank Armor = Tank gear (more Armor = more Resolve from the same DTPS)

          This is all multiplied with eachother, so that boss DPS curve you need to balance against is a 3rd power of how fast player gear increases. e.g. 50% more stats = 237.5% extra boss DPS. I doubt this can be balanced against the assumption of 5 healers in a mythic raid group. Without the Armor+Resolve part, this would only be 125% boss DPS increase, about half of what it would be otherwise!

          On the “how”
          What Theck didn’t touch is that the big issue with Vengeance/Resolve & gear upgrades is Armor (and to a lesser extent, the damage reduction from Versatility). More Armor = larger % of damage taken you can heal. This is true for Vengeance now, which has lead to the lol tanks we see in SoO, and while self-healing will be less powerful in WoD, the Resolve mechanics will still make it scale up faster than boss damage will increase.
          This actually adds contraints, because Resolve coming from taking magic damage does not scale with Armor, so every single boss will need to have its melee damage adjusted upwards so the sum of magic dmg Resolve and melee Resolve is enough to counter X boss DPS. It will grow towards tanks taking larger portions of their HP each boss they advance, just like non-DR avoidance caused in tBC, mana regen in Cataclsym and Vengeance does in MoP. They had to change how the first two work, it’s time to learn from the mistake they made with Vengeance now.

          I’ll say it again:
          If an item upgrade reduces the damage you take by X%, the external healing you need should not be reduced by more than X%, or you will grow towards immortal/lol tanks as we have now in SoO. This problem is caused by how Armor (doesn’t) affect Vengeance and it will be the same for Resolve in its current state.

          PS: Forget my statements about Guardian Druids, I forgot they changed their Mastery, so it already works post-armor and removes yet another argument as to why Resolve should be calculated on pre-armor numbers. And yes, I’m on a crusade against this pre-armor mechanic, because I’ve predicted it to go wrong 3 years ago with Vengeance, and I’ve been proven right more and more in the last 8 months, which is aggrevating!!

          • Gupsy says:

            OK, I see your point on AP=HP for scaling. Probably you are right that making abilities depending on AP or HP is really similar, now.

            What I was saying is just to use the same mechanics for both, capping and scaling on the same variables, so you are sure they will remains always balanced (or unbalanced) together.

            Now the prob is those two abilities aren’t at all on the same curve. As healers HPS scale up similar to DPS, boss damage will need to follow that curve, and so will our hps. I’m just changing your point of view from “tank” mechanics, to others. So DTPS is not scaling up because of armor, but because of HPS scaling, and Armour is the things that keeps the tanks in par with this scaling.
            My fear is we will have many micro adjustment just to keep things going (the resolve mechanics is there mostly for this). Just to keep tank dps aligned with 75% of pure dps they will have to change much during progression. And you are balancing against a point in the curve while only a minority of your player base is there. As all those “abilities” are connected each other – and usually each class will have his own connection under the hood – tweaking one will move other unwanted things around.

            On this regard Armor is not a big concern, as it’s scaling is quite linear, the “bonus” limited to tank only slot. For sure it is one of the factor that manifestly bring some concern about balancing to the table, but it is not at all the only one, nor the worst.

            Decoupling abilities as much as possible, separation of concerns (defense % of HP, dps % of AP/SP, HPS % of SP) and capping based on linear scaling variable (like HP, SP and AP i.e), possibly using asymptotic curves can make easier keep all this balanced.

            All joined with a lower curve of item progression. Ie they can fix it around a maximum of 20% intra and between tiers. Doing so at the end of the expansion (start+4 tier) we would have a x2 from start. Balancing a curve with such a range is not easy, but doing it with a x10 is nearly impossible. And I’m not hyperbolic, the actual rate for MoP is more then 10 times! The OP of EF and Vengeance lollable is not at all a uniqueness. We had the lock soloing of Immerseus, the 2.5M average dps of Locks on protectors, the solo healing of 25H… Out of single outperformers the x10 things is an average things. The following numbers seems unreal to you?
            In MoP from starting equip to end:
            DPS: start 500k
            HPS: start around 20k, end >200k

            With such a scaling I think it is quite impossible to balance more then a single point, not to talk about PvP.

          • Theck says:

            @Airowird: While boss damage needs to scale up with all of those factors, you’re presenting a false dichotomy between that and healer output. Boss DPS scaling and Healer scaling will scale roughly the same by necessity.

            Or, put another way, a healer’s HPS will increase somehow with gear – the exact details aren’t that important, because with the new design paradigm they’ve talked about, that’s the independent variable this time around. Boss damage will be increased such that the tank’s damage intake *after* all of those effects (Armor, Versatility, Avoidance, Active Mitigation) is an appropriate challenge for healers.

            That shouldn’t be all that hard to balance at all, and really is a more logical method than trying to balance players around bosses (which is how MoP and prior expansions have been done) given that bosses have far more free parameters to work with that don’t affect PvP.

            @Gupsy: Resolve *is* using the same mechanics for both. The amount mitigated by SotR increases linearly with boss DPS (raw or post-armor-mitigation; it’s irrelevant since they’re both multiplicative). Resolve adds a multiplicative factor based on boss DPS to self-healing – in other words, it ALSO linearly increases WoG with boss DPS.

            In other words, it’s the simplest and most direct way to keep the two scaling at exactly the same rate across the entire expansion. In fact, Vengeance was quite successful at that in MoP.

            Where Vengeance faltered is the sheer self-sustainability it gave tanks. Because Vengeance gave AP directly, in later tiers it was possible to literally have 10x-15x as much attack power as you did without Vengeance, which meant your self-healing got 10x-15x better. That was the root of the problem you’re describing.

            However, there’s no reason to assume Resolve will cause the same problem. In fact, there are a few reasons that indicate it won’t. First, it’s calculated entirely differently, so it’s not possible to game it and build up a huge reserve. It adapts much more quickly to your current damage intake level. The elimination of DoT/HoT snapshotting also means you can’t exploit the mechanic by casting a large Eternal Flame when you’re at peak Resolve.

            Perhaps more importantly, it scales significantly less quickly with boss DPS than Vengeance did. While I normalized the plots for clarity, the effect of Resolve is about a factor of 10 smaller than Vengeance. In other words, a situation where Resolve gives you an extra 50% healing is a situation where Vengeance would give you an extra 500% healing. Another way to put it is that in my sim, a boss that gives you 500k Vengeance AP gives you about 50% Resolve.

            And note that, just like Vengeance, this can be tuned down with a single parameter. They could decrease the coefficient of 0.25 in a later patch to apply ad-hoc diminishing returns to compensate for armor inflation in exactly the same way they decreased the Vengeance coefficient (from 0.02 to 0.018 and eventually to 0.015).

            In fact, one of the reasons that coefficient wasn’t slashed more significantly in MoP was because Vengeance tied into tank DPS, which they didn’t want to slash. But again, Resolve fixes that by decoupling the two concerns.

            Furthermore, keep in mind that tank self-healing is being slashed across the board. Our baseline self-healing will be considerably less noticeable than in MoP, which is then being multiplied by a much smaller value (Resolve) than in MoP (Vengeance).

            In other words, I think the devs are keenly aware that our self-healing went from 10% to >100% in MoP, and have taken a number of steps to curb that with Resolve. I’m not sure what leads you to believe that this new system isn’t balance-able, because it seems far more flexible and accommodating to the designers than any previous system.

            I’m not sure what your concern is about item level scaling; in pretty much every expansion we’ve doubled or tripled our DPS from the beginning to the end. That’s just how WoW item scaling works. MoP was on the high side thanks to ilvl inflation to be sure, but not unreasonably so. Most of the feats you cite are extreme outliers that have much more to do with encounter/class mechanics than they do with raw power scaling.

          • Airowird says:

            Ok, I agree, bosses need to be designed around players (and somewhat to the same extent, raid groups, as preferred/expected composition directly affects individual players). And yes, bosses will need to grow exponentially, because its raw output has to be = incoming HPS / (1 – Tank reduction).

            The issue that currently exists is that incoming HPS is not a steady balance between tank HPS and healer HPS. Incoming heals are divided into 3 categories:
            a) X*(b) Healing, X based on % of boss damage (through Vengeance/Resolve),
            b) Base tank self-healing, based on gear
            c) Base healer healing, based on gear.

            Note that if we consider (b) and (c) linear to gear ‘power’, (a) scales from the base amount, the amount we can use our ability through resource generation (also based on gear) and an additional time through gear, because that coefficient X scales with 1/(1 – armor reduction). If we were to express our gear ‘power’ from earler as a factor P, total healing received by a tank is:
            HPS = P^2 * A + P * B + P * C
            HPS = P^3 * X * B + P * B + P * C
            HPS = (X*P^3 + P) * B + P * C
            It should be clear that any noticable power increase will shift the HPS more towards tank sustainability (B) and away from actual Healer requirements (C).
            As you have proven in your Vengeance with a Vengeance blog, there is an (1-A)*(1-M) magnifier for the self-sufficiency abilities. Atleast this (1-A) factor needs to go to. (1-M) is debatable, because not all tanks mitigate equally, and some stats (e.g. Riposte) boost avoidance while others (Haste for most) boost resource generation and thus recovery abilities without touching mitigation.

            If we are to compare ShotR and WoG, I believe the design intent is to have ShotR as an efficient, reliable(*), preventive, measure, while abilities such as WoG are created to recover from sudden RNG spikes or burst damage and generally less effective in terms of damage removed per resource. This means that there already is a sort of DR on Active Mitigation (B, sorta) as an attempt to partially balance out resource generation

            (*)Monks and Druids are prone to RNG on their ‘buff’ ability, thus will always require some use of recovery tools, but both Paladins and DKs can not keep up their ability for extended durations of time. This is why (I think) Shield Block has a cooldown.

            To point to your own example: Blizzard has needed to reduce the Vengeance coefficient by 25% (0.02 to 0.015) simply because we were getting too far out and we still have issues with it, because comparing armor between a blue-geared tank and one in full SoO heroic gear, the latter one takes half the damage purely on armor alone. Outside initial progress I can’t recall any boss (except Thok) that we 3healed in 10man Heroic, even though raids in Tier 14 were pretty much required to have you bring 50% more healers during a time our tanks healed for 33% more than they do now in the same gear!

            Will Resolve have the same issue as Vengeance, no, because, as you said, it no longer is required to balance tank DPS and it’s only effect is tank self-healing. But if scaling is far enough out of balance that they need to reduce the scaling factor each patch, tanks will see their heals drop regardless of what Blizzard does, either through patches (GOT NERFED QQ), or because they equipped some bonus armor (game mechanics says you don’t need it anymore/offset for it being an otherwise superior stat)

          • Theck says:

            First, let’s make a clear distinction here between passive self-healing and active mitigation. WoG is Active Mitigation – it competes with SotR for resources, and thus needs to scale appropriately. That’s what Resolve accomplishes as a baseline. The self-healing of WoG shouldn’t really “count” as tank self-healing in a sense, because it always comes with a significant opportunity cost (less SotRs). Ignoring quirky set bonuses and such for the time being.

            Where Vengeance caused problems (and Resolve has the potential to cause problems) is with the passive self-healing we get through Seal of Insight, Eternal Flame or Sacred Shield, and possibly Empowered Judgments. Because those have no opportunity cost, and don’t represent the tank choosing to forego mitigation in favor of healing.

            There’s no doubt that Resolve will cause those passive sources to heal for more as the expansion progresses. And that will naturally mean that the tank will require less external healing, at least as a percentage of intake. On that, at least, we agree.

            But you seem to be assuming that for things to be properly balanced, the exact percentage of intake should be fixed across the entire expansion. And I don’t necessarily agree with that.

            For example, if a tank is passively self-healing for 10% of his DTPS at the beginning of the expansion and *doubles* that to 20% by the end of the expansion, why is that necessarily a problem? The vast majority of his healing intake still comes from healers, and he’s nowhere near the point of self-sustainability.

            The healers still aren’t going to slack off 20% of the time. It’s not like they are being depriving of gameplay. We’re just creating a little more wiggle room for error, which is what gear upgrades do naturally anyway.

            The point here is that we don’t really have enough information to say that Resolve is any way broken. Yes, it scales nonlinearly, but so do various other mechanisms in the game. As long as the end result is reasonably bounded, that’s still very well within the realm of “balance-able.”

            It’s still very early yet in the alpha phase, let alone beta. If tank self-healing looks strong enough that they anticipate Resolve creating self-sustainability problems in T19 gear, they have plenty of time to nerf it.

          • Airowird says:

            Ok, we’re apparently looking at self-healing stuff differently from eachother. Considering both WoG and Seal heals benefit from haste, crit, etc., I see all self-healing abilities, WoG and Seal alike, as consisting of two parts:

            On one hand, you have the base healing they do without taking damage, this is based purely on the gear you have, scaling roughly the same as healer gear (+ a wee bit stamina). Compare this to survival talents or glyphs, like Bloodthirst heal, Shaman Glyph of Flame Shock (30% of damage heals you) etc. This part is pretty much part of the “healer team total HPS” you’re contributing to, because it’s in effect even when you’re not tanking. It’ll be higher than the DPs healing, but then again, you’re not doing as much DPS as they are (according to design intent)

            On the other hand, you have the healing done through Resolve. I currently expect this to be at less DPS removed than the ‘main’ mitigation button such as ShotR, Shield Block etc. but overall, these ‘recovery’ AM abilities should scale linearly with damage taken, just like the prevention ones do.

            So to me, if you are healing (as example) 15% of DTPS through WoG, with 10% coming from the ‘base’ value and 5% coming from the Resolve part, I expect the following things to be true in a balanced environment:
            a) That the 10% will roughly remain 10%. (Scales roughly same rate as healer gear, Stamina Resolve balances out Spirit a bit, but gneerally this can be ignored as a major balance issue)
            b) That the 5% will go up roughly at the same rate as the amount of damage ShotR prevents. So if, through Mastery, you go from 15% to 30% reduction from ShotR buff, then WoG needs to go from the 10% + 5% to 10% + 10% = 20%, not 30%. Of course crit might benefit WoG more than ShotR and maybe Mastery will buff ShotR more, but overall, if you’ld be getting 100 rating of everything, it should be fairly equal in relative/% progress.
            Currently, this is not the case with the alpha mechanics of Resolve, mainly (imho) through the consequences of % Armor reduction hyperbolically (literally!) buffing Resolve.

            For clarity, this is what stuff means to me:
            Prevention = Shield Block, ShotR, … All the abilities that prevent X% physical damage taken within a certain time frame
            Recovery = Shield Barrier, WoG, Rune Tap, … The stuff you press when you’re low on HP and a big hit is coming in or simply because it works on magic damage.
            Passive Heals = Blood Craze, Seal, … basicly just stuff that happens without pressing buttons.

            If it’s still unclear as to what I am trying to express, send me a PM on Tankspot and we’ll set up a VoIP moment somewhere, because I’m not sure I can explain it any better in text as I have already.

          • Gupsy says:

            @Theck:I agree with you about Resolve mechanics in itself. As I told in the first post my concern about it are all gone. I like how it works, I like the little “regressive” part too, is a fine tuning plus. So resolve in itself is perfect.
            The prob is what a similar mechanics implies. Its presence means they are planning to maintains the same mechanics of scaling they have now for overall abilities: at the end of the various multiplicative effect we will have exponential scaling. This is not a “needs” of the game, is just how it worked since now.

            So, resolve is really fine if you adhere to the actual scheme, but the actual scheme as proven itself as practically unbalanceable in the long run. With MoP the last and worst example of it.

          • Theck says:

            Not sure why, but reply got un-nested. See bottom of comment page.

        • Theck says:

          Wait, what part do you consider a “clear design error?” The incredible scaling of Eternal Flame (and other such abilities) with Vengeance had more to do with those abilities than the Vengeance mechanic itself.

          Are you saying you don’t believe that there needs to be a mechanism to keep e.g. WoG competitive with SotR over several tiers of content? Or that you believe it can be achieved or done better in other ways, i.e. without a Vengeance/Resolve mechanic?

          (Note that I don’t really have a dog in that fight; I’ve argued both sides before).

          • Airowird says:

            I think he means that the error was to refurbish the Cata DPS Vengeance into a self-healing balancer, which (clearly) didn’t work out as well as a completely newly designed mechanic could have.

          • Theck says:

            I’m not sure I’d agree with that, given that it worked out pretty well for the primary function (keep WoG & SotR competitive). They’re keeping the successful part of it (Resolve) for WoD, and axing the parts that caused problems (the damage output part).

            I guess I could agree with the argument that a completely newly designed mechanic (aka Resolve) would have been better for MoP as well, but a lot of that is only “clear” in hindsight. Though it’s true that many of us pointed out the potential issues with the AP conversion well in advance (mostly in terms of gimmicky tank DPS stuff).

          • Gupsy says:

            @Theck
            Yes I means it, an it has to do with the abilities for sure, I agree on it too.

            They keep the abilities fixed in stone, trying to invent a complex mechanics to manage a wrong ability design. In design you must fix the goals not the implementation.

            I.E. if the intent was to have two abilities on par, competing on the same resource, but one for physical and preventive, the second one a generic “healing/absorb” for magic alike. You can make both capped on % of tank health. If you really want those strong just chose to cap those ability on the % of damage received too, something similar to the actual resolve, but that didn’t scale with content (in % vals), as I don’t think it is a design goal having tank that absorb/heal/reduce 10% at the beginning and 110% at the end of the expansion, I think is just a flaw of the design. And where are we going? Just on a curve with an asymptote, something like diminishing return was. I.e. f(HP,HIT) -> 60% of Total HP. At that time you can tweak the abilities, making them interesting and differentiate them for different tank, having something that works under the hood you can free the imagination.

          • Airowird says:

            Actually Gupsy, I looked at this exact issue, and if they make Resolve only scale with the damage you take after armor, that 10% healing coming from tanks won’t even hit 11% on a scaled up boss. Changing one number in the current formula seems easier than capping stuff based on several cases. Also makes it easier to understand, because it’ll be based on the actual damage you would take.
            If I got time this week, I’ll make some matlab graphs about it and post it on tankspot (or send Theck the info). Maybe it will be more clear then

          • Theck says:

            @Gupsy: I’m not really sure I understand your argument. They tweaked both the abilities and the Vengeance mechanic in the Cata -> MoP transition. And in any event, the choice they made was the most logical and *least* complex way to do it. If your desire is to balance an ability that scales with boss damage (SotR) with one that scales with points/AP (WoG), then the logical interface is some bonus that depends on boss damage.

            I don’t think that instituting a % max health cap on both abilities makes more sense. That would artificially create another imbalance based on cap itself.

            And in any event, I think you’re being a bit hyperbolic about the design goal varying from 10% at the beginning of the expansion to 110% at the end. The Vengeance/healing interaction certainly got out of hand, but I doubt that was by design. And the reasons for that have been identified and rectified in the Resolve design.

            In short, I pretty much flat-out disagree with your analysis on this topic. I think Resolve is a good mechanic, and arguably what Vengeance *should* have been replaced with going into MoP.

      • Çapncrunch says:

        @Airowird:
        I’m not sure I agree with some of your statements, such as Armor being some unique factor in the equation, or that Magical Damage does have a similar scaling issue as physical damage does.

        As far as Armor being unique, it’s honestly not. You’re basically saying that armor is different and a problem because it reduces the damage you take. But there are lots of stats that reduce the amount of damage we take. How is damage reduction from getting more block chance different than damage reduction from armor? Or avoidance? Or increased AM uptime? They all basically equate to X stat rating = Y% reduction in DTPS, exactly the same as it does with armor. The size of the multipliers are different, but it’s ultimately the exact the relationship.

        The actual problem that you are on about is that more gear reduces our damage taken and increases our healing done, and for some reason you seem to think that armor is the only cause of the first…which it certainly isn’t.

        “If an item upgrade reduces the damage you take by X%, the external healing you need should not be reduced by more than X%”

        The ONLY way to accomplish this is to make self-healing not scale with gear at all. To literally make it so that as your gear improves you take less damage and heal less damage.

        For example let’s say that with a given set of gear a boss does 1000 dps to you (not raw, I’m talking about actual DTPS), and your self-healing is say 200hps. So you need 800 external hps to stay alive, that’s 80% external heals.

        Now you get better gear and the boss now only does 900 dps to you, a 10% reduction in damage taken, so by your logic this means that our external healing needed CANNOT be less than 720 because if it’s any lower then our external healing went down by more than 10%. This means that our self-heals can’t be any higher than 180. Which is lower than they were before.

        That’s literally the only way to solve the “problem” you are talking about.

        • Gupsy says:

          @Theck: the problem origins from the abilities, not from the vengeance part. Resolve change the vengeance part, fixing it a bit indeed, but the problem is still there, ready to come up again. To make it simpler my statement is:

          a) do not mix different factors on different goals (AP is unnecessarily pushed in DTPS)
          b) use same dependencies and behavior for abilities you want to be micro balanced (ie SotR, WoG and variation)
          c) use DR like func, so you avoid the fix cap prob that brings reforge to the table, you push the players towards auto-balance attitude, compensate a bit with excessive iLev scaling (ie any func in the form of f(x) = (M-1)x/(x+(M-1)+k), you can easily tweak the slope with k and the limit with M)

        • Thels says:

          @Capncrunch: The problem is that Resolve is based on pre-mitigation damage, and we mitigate damage exponentially, rather than lineary.

          As we progress through the expansion:
          - Our armor goes up, reducing physical damage intake.
          - Our strength (or agility for druids/monks) goes up, increasing avoidance.
          - Our secondary stats go up, increasing avoidance, block chance and most notably active mitigation.

          So as time progresses, the pre-mitigation and post-mitigation damage starts to skewer more and more in an exponential way, and that’s not even including that our healers are also getting better at their job. By removing one or more of the above options, the values skewer less exponentially and more linear.

          Naturally, they don’t want to penalize us for keeping our Active Mitigation up, or preferring certain secondary stats over others (not to mention that the effects of secondary stats vary so much from class to class), that they can’t cut out the part affected by our secondary stats.

          Since both crit and primary stats affect avoidance (as well as druid AM, or did they change that?), working with avoidance might be a bit finicky. Also, while not really an issue for Paladins, effects like Shield Barrier and Blood Shield could be penalized for having a parry streak just before a big effect that you want to mitigate.

          Armor however is clean and clear. It scales naturally as you progress through the expansion with gear, works the same for all five classes, and affects every physical hit exactly the same, without lucky or unlucky streaks like avoidance. Therefor, it’s a lot cleaner to ignore Armor, than it is to ignore avoidance, block chance and active mitigation.

          While not perfect, letting the difference between Resolve and active mitigation be skewered by 2 different causes rather than 3 different causes reduces the problem. It doesn’t remove the problem completely, but it’s a step in the right direction.

          Personally, I’d rather see Shield Barrier, WoG (prot only), Blood Shield, etc receive a bonus based on a fixed percentage of your own health.

  11. Yûko says:

    you sure that resolve doesn’t change our thread level? As i will quit tanking (after 5-6 years) it’s nothing that bothers me much but i have seen stuff which wasn’t … lets say intended :) …. seal of light -> saphiron -> just see how other people give you healthread for free …

    With resolve the dmg output won’t be buffed at all but when i see how a vengeance loaded eternal flame could get you loads of adds @imerseus or can even whipe you if you forget to turn off rightenous fury at garosh hc p2-p3 addspawn how will ef thread will be with resolve in there and all tanks will be on par dmg wise, time of autoattack and hoping of not overaggroing?

    • Gupsy says:

      Just speculation, but as they are giving us SP = AP, generally we will have more people below 100% for longer time, mana is still of no concern for us, we will have height crit due to crit=parry … Will smart paladin return on real versatility? Is it possible we will start healing on tank swap?

    • Thels says:

      Resolve is probably still affecting healing threat, but this is usually insignificant to DPS threat, except indeed for picking up fresh spawning adds.

  12. Theck says:

    I’m not sure this example makes any sense:

    “That the 5% will go up roughly at the same rate as the amount of damage ShotR prevents. So if, through Mastery, you go from 15% to 30% reduction from ShotR buff, then WoG needs to go from the 10% + 5% to 10% + 10% = 20%, not 30%. ”

    To be abundantly clear, let’s consider two scenarios. First, let’s take a boss that hits you for 100 damage after armor. Let’s also say through gear upgrades your SotR goes from mitigating 15% of a boss’s hit (15 damage) to 30% of a boss’s hit (30 damage).

    Your WoG before those gear upgrades might’ve healed for 20 damage, let’s say 15 from the “base” value and 5 from Resolve effects. For the sake of argument, let’s say that those gear upgrades doubled your health and AP since they (more than) doubled your mastery. So now your WoG heals for 30+5=35 damage, not quite double what it did before (thanks to Resolve) and a smaller percentage of your health than it did before (because you doubled your health).

    Now let the boss hit twice as hard. Hits go to 200 damage after armor, SotR mitigates 60 damage. If we double the Resolve portion from 5 to 10, WoG is only healing for 30+10=40 damage, not fast enough to keep up with SotR. Since we doubled damage after armor, it should be even higher before armor – let’s say even 4x (unrealistically high). That still only puts us at 30+5*4=50 damage. It’s clear that the Resolve portion of WoG needs to increase faster than you describe to remain competitive.

    Note that shifting more of the base value into Resolve (ex, going to 10+10 -> 20+10 -> 20 + 20 -> 20 + 40 in these scenarios). Only barely allows it to keep up in the extreme case. And that’s ignoring the fact that SotR has a base value as well.

    In any event, I think it gets the point across well enough for SotR vs. WoG. What matters is that WoG heals for enough to make it worth considering, which generally means “heals for more than SotR mitigates from one attack, but probably less than SotR mitigates from two attacks.” That’s a really broad target to hit, and that’s even a soft upper limit (because a heal can be rendered useless by external healing, where mitigation is generally not). It really doesn’t matter if WoG gains on SotR over the course of the expansion thanks to Resolve inflation as long as its kept within a reasonable range, which should be completely possible to do through coefficient tuning.

    It’s a tighter constraint when comparing two mitigation abilities, like Shield Block and Shield Barrier, because the upper limit is more strict. Again, I’d say the target to hit is that Shield Barrier needs to mitigate more than two blocks worth of mitigation (so about 60%+ of an attack) to be really competitive, and possibly more. But if it mitigates more than four blocks worth, then it’s quite possibly too strong (ignoring smoothness concerns, which may not be a good omission at all). But again, I think this target is obtainable by tuning the Resolve coefficient properly.

    Don’t forget that “exponential scaling” isn’t necessarily a bugbear. As long as you keep $x$ confined to the appropriate region, $e^x \approx 1+x$.

    • Gupsy says:

      Lol, good point :D

      Ok, I adhere to your optimistic view, even if I’m a bit sheptic about it… The main reason is I can’t do anything to change it :p

      After all they managed it before Cata (but for the avoidance in late TBC) and went completely out of any reasonable limit with MoP.

      Hoping they will revisit iLev scaling this time (=keep x in the reasonable range).

      • Theck says:

        It’s good to be skeptical. They can and do make mistakes. But at this point I think it’s more important to use that skepticism to identify places we should watch rather than proclaiming that anything is irrevocably broken. We’re early enough in alpha that there’s no guarantee any of the numbers we’re seeing have been tested beyond a simple guesstimate.

        The inevitable self-healing creep that the math predicts is something we should watch during alpha/beta to make sure it doesn’t get out of hand. If it looks like it will be a problem once numbers have settled down more, that’s the time to break out the pitchforks and torches.

        Just keep in mind that they re-balance things every tier nowadays, so they have ample opportunity to correct things mid-stream if necessary. In the case of Vengeance, I think they couldn’t correct things *enough* simply due to how interconnected all of the components were (specifically tank DPS and self-healing gains). In an ideal world, the separation of those components should mean they have more leeway to tune Resolve to keep it from going overboard.

    • Airowird says:

      Basicly, gear upgrades can be put into 3 categories:
      a) Primary stats (Stamina + Strength/Agility/Intellect)
      b) Secondary stats (Mastery & co)
      c) Armor

      We have been referring to (a) + (b) as ‘power’, namely the throughput increase healers and DPS see.
      Because for healers effective boss DPS is all that matters, let us assume that eDPTS scales with power as well.
      I am also taking the liberty of assuming that all raid members get geared equally.

      Now let’s look at how damage compared to our max HP. Because Stamina is fairly balanced compared to other Primary stats, we can remove all primary stats from scaling

      Healer HPS scales with power, so in %HP they only scale up with secondary stats.
      Similarly, boss DPS as % tank HP only scales with secondary stats.
      The preventive AM buttons (ShotR etc.) scale with secondary stats twice; once from boss DPS, once from being able to use them more (or them being more efficient)
      The base part of our heal/recovery AM button scales with secondary stats.
      The Resolve part scales with secondary stats (because it is % of base), secondary stats again (%boss DPS/HP) and Armor (takes boss raw damage rather than effective).

      So our base heals scale at the same rate as healer HPS. This is basically the raid-total HPS, as you don’t need to be tanking to heal yourself for the base amount. Preventive AM abilities scale quadratic, but are limited to physical damage and spread out over time (as partial reduction or RNG buffs). Recovery AM scales quadratic with secondaries AND scales even more through Armor.

      With B & C being the stats described at the start and P & R being the ability scalars, you can compare preventive vs recovery AM abilities as followed:
      P * B^2 =?= R * [B + B^2 * C] (divide by B)
      P * B =?= R + R * B * C (and again if you want to write it in function of R)
      P =?= R * ( 1 / B + C )
      P / R =?= ( 1 / B + C )

      So this means that to be truly equal in power, 1/B + C needs to be somewhat constant
      Because we get a certain amount of crit & Mastery for free (S) and the amount of extra secondary stats from gear is roughly linear to the amount of armor on gear (G), this means that
      1 / (S + G*C) + C = P/R = constant for values of C ranging a relatively large range.

      Obviously, at C = 0, the constant is 1/S
      So at 50 armor reduction, C = 100% -> 1/S = 1/(S+G) + 1
      >> multiply with (S+G)
      (S+G)/S = (S+G)/(S+G) + (S+G)
      1 + G/S = 1 + S + G
      G/S = S + G

      So, I admit, it IS possible to balance Resolve to a certain point, and there is even room to play, although I’m skeptical about how well that goes, as going further away from the balance skews the linearity.

  13. woodknight says:

    Hi Theck, this might be an outdated question. If you have answered it in one of your previous posts, a link to it will satisfy me.

    So the in-game description of vengeance is ‘Each time you take damage, you gain 1.5% of the unmitigated damage taken as attack power for 20 sec’. It feels to me that vengeance should grow or drop linearly rather than exponentially. Could you enlighten me how to draw the difference equation and get its exponential solution from that one sentence description of vengeance.

    • Theck says:

      I’m not sure if I described it in a previous post, but the old Vengeance calculation worked like this:

      NewVengeance = OldVengeance * (Duration Remaining) / 20 + VengeanceFromNewAttack

      Which is pretty clearly a difference equation for your Vengeance amount V:

      $V_{i+1} = V_i * K + C$

      So for example, if you had 100 Vengeance at time t=0, and 2 seconds later a boss hit you with something that gave 15 Vengeance, your new value would be:

      $V_{i+1} = 100 * (20-2) / 20 + 15 = 100 * 18 / 20 + 15 = 90 + 15 = 105.$

      To see how to get an exponential from that, consider that $V_{i+1}=V_i+\Delta V$, and re-arrange the equation:

      $\Delta V = V_i (K-1)+C,$

      which very clearly has a general solution of the form $V = \alpha e^{\beta t} + \gamma$.

      • woodknight says:

        I see, thank you for the patience. So actually the in-game description of vengeance is not accurate, should be ‘Each time you take damage, you gain 1.5% of the unmitigated damage taken as attack power for 20 seconds scaled by duration remaining/20′ …

        • Theck says:

          Tooltip descriptions aren’t always chosen for accuracy. In the case of Vengeance, most players don’t actually care how it works under the hood. So they opted for a very general description that gives the relevant numbers without really describing the calculation. Resolve’s tooltip is similarly vague.

          Note also that in both cases, the developers described exactly how the calculations were performed in forum posts, so that we’d know how to properly model it.

Leave a Reply