A Eulogy for Silver

Silvering Vet Dungeons made me a better player.  It made my group of players, three of whom I didn’t know from Adam going into our first run, a cohesive team.  Each dungeon in turn was a challenge that steadily bent, with practice and sustained effort, from seemingly neigh-impossible to almost easy.

Was it disheartening to wipe at 1% a few times on Forgemaster?  Yes.  Was I often worried that someone on the team would give into frustration and want to call it an early night?  Yes.  Was it scary when I noticed my hands shaking with excitement/nerves as we were forced to sit through Mordechai’s (seemingly endless) roleplay while the timer rapidly ticked down?  Yes.  But damn, it was fun.  Challenging, exhilarating, rewarding fun.

—————

Disclaimer: This is my personal (visceral) reaction to the news that the Vet Dungeon attunement requirement will be reduced from Silver to Bronze.  I will likely post a follow up with more general, dispassionate discussion of the change and its impact on the game. But for now I’m just posting how I feel.

The Times They Are A-Changin’

Okay, the latest (Sabotage!) drop changes might not exactly rival the 1960’s in terms of their broader cultural significance, but they are likely to change up the playstyle of Healing Medics across Nexus, so from the narrow weltanschauung of this blog, they warrant a Dylan reference.

Mending Probes nerf

As was expected the part of our healing kit that was deemed “overtuned” was Mending Probes major tier upgrades.  The T4 detonate got nerfed from 50% of the HoT potential to 35%, the number of T8 splash heals was lowered from 3 to 2 per probe, and the size of the these splashes was reduced even further from 100% of the base detonate to 75%.

The table below compares Mending Probes T8 + Flash T4 combo, pre and post patch, to some other abilities.

Ability 20-man AvgHeal/ Player (k) 5-man AvgHeal/ Player (k) “Cooldown” Range (m) Cast time (s)
OLD MP T8 + Flash T4 5.2 17.4 10.2 6 1.0
NEW MP T8 + Flash T4 3.3 9.6 10.2 6 1.0
Crisis Wave T8 1.4 5.8 2.5 15 1.25
Crisis Wave T8 Burst 4.3 17.3 5.0 15 2.5
Reverie T8 (> 35%) 3.4 6.9 6.0 25 0
Reverie T8 (< 35%) 4.5 9.0 6.0 25 0

In 5-mans the AvgHeal/Player drops from 17.4k to 9.6k, a substantial (45%) reduction.  This puts the combo only slightly above Reverie T8 (on targets < 35% heath) despite the longer cooldown and significantly lower range.  Things look worse for 20-mans where the AvgHeal/Player is now only 3.3k, even below Reverie T8 on players above 35% health, and again on an ability where players need to be tightly stacked rather than spread out.

Some players, including me, suggested altering the detonation mechanic so that a player could be hit by at most one splash heal, which would reduce the healing in 5-mans but not 20-mans.  Note this would reduce the 5-man AvgHeal/Player to 10.5k, 10% higher than the new value.  So I would classify the change, on 5-mans, as “major” but not completely game-breaking.  However, in a 20-man environment, MP T8 + Flash T4 now looks pretty uncompelling, which is disappointing since Medics don’t exactly have a lot of skills in our kit that are useful for healing large groups of players (our only other ability that hits more than 5 targets is Rejuvenator).

You will note I’ve added Crisis Wave (CW) T8 Burst to the table: this assumes you cast Crisis Wave three times in a row, taking advantage of the major tier upgrades that make the third cast instant (and off GCD) and Actuator and Focus free.  The burst healing per target put out by CW T8 x3 is almost the same as MP T8 + Flash T4 pre-Sabotage, and has a less restrictive positioning requirement.  Of course, it comes at a cost of four Actuators, a 2.5sec cast time (as opposed to 1sec), and 40% more Focus (86 vs 61, unAMP’ed).  So with these changes should Mending Probes T8 be replaced by Crisis Wave T8 on our LAS?  Possibly; it’s definitely something I will be experimenting with in the coming days.  Although Crisis Wave does appear to do the job for burst (with some drawbacks), I will certainly miss the Focus-efficient healing of the Mending Proves HoT, and the fact that you could place the probes on someone and then run out of range to do something else and the HoT would still be ticking on them 12sec later.  Of course, with Crisis Wave now on your bar, if you are at full Actuators and the tank is close to full shields, you can afford to spend some of those Actuators on a Crisis Wave to heal those DPS that normally would have been taken care of via the probe HoT.

Reboot Changes

Reboot will now trigger when the target is below 30% shields, but will hit for only 135% of Support Power, down from 225%, a 40% reduction.  Using typical stat values1 (and taking into account that Reboot can crit), the average Rebooted (base) Shield Surge will go from healing for 8.9k to healing for 6.8k, “only” a 23% nerf (since the Shield Surge healing itself is unchanged).

How much does this change matter?  One way to evaluate it is to consider how often you will be casting a Rebooted Shield Surge if your strategy is to only cast Shield Surge when Reboot will proc, and assuming you are healing a target taking steady damage.  The math says in any given time period you will end up casting Shield Surge 8.9k/6.8k – 1 = 30% more frequently2.  That’s a pretty significant increase in the number of times you are casting a Rebooted Shield Surge, and I think you will feel the difference in Focus burn rate.  That said, although the heal from Reboot was nerfed, the fact it triggers off of someone at 30% shields means it’s easier to maintain a higher average shield level, and hence a higher average Effective Health, although clearly it will now cost you more Focus.

More to come…

The Devs specifically mentioned they wanted to “update Triage to function as a powerful single-target emergency heal.”  Let’s look at what needs to happen for Triage to earn a spot on the bar.

For Triage to be viable as a single-target emergency heal it needs to be a targeted heal.  Period; full stop.  They tried something else, and it has issues.  They might be able to cook up some other way of Triage selecting a target, but let’s face it, no matter what the mechanic there will be corner-cases, so short of having Triage read our minds, the only way it is guaranteed to hit the target we want is for it to be targeted.  As Hick’s said, “it’s the only way to be sure.”  And sure is what we want in an emergency heal that will only hit one target.  Do I relish having to disengage mouselock (I currently use Steer), select a target, and reengage mouselock?  Nope, I’m definitely not looking forward to it (although having the mouselock toggle bound to the G-shift key on my Logitech G600 makes it pretty seamless).  But I will do it if the heal is good enough.  And so will you.

How good does the heal have to be?  Well, it has to hit for enough to make having it on our bar worthwhile compared to other abilities.  With typical stats, Emission T8, Flash T4, and Shield Surge all hit for around 3.5-4k, but hit 5+ targets, so I think Triage has to be well above these numbers, even at base.  To get some more points of reference, Esper’s Phantasmal Armor hits for 9k at base (but with a 45sec cooldown), and Esper’s Mending Banner and Slinger’s Astral Infusion each hit for around 7k.  So it seems like something in this neighborhood, say around 8k, would be compelling3. That’s around double our multi-target alternatives and gives us an incentive to cast Triage over them when we want to concentrate the healing on one target.  Triage currently splits its healing 50/50 between health and shields, and I like that aspect (why heal health or shields when you can heal both!) and see no reason to change it (especially in light of the Reboot change).

What about the cooldown?  45sec seems about the longest the Devs go (e.g., Barrier, Phantasmal Armor).  This seems a bit on the long side for our only single-target heal, with something in the neighborhood of 20-30sec feeling more reasonable.  So let’s say 25sec as a ballpark number.  If one were then to (ab)use it by casting it on cooldown to supplement a normal single-target tank healing rotation of Emission T8 x2 + Shield Surge T0, it would only increase the single-target sustained HPS by around 7%, so it won’t give us much better single-target sustained healing, it will just allow us to ramp it up a bit when needed.  Since this added throughput shouldn’t come for nothing, the Focus cost should be high enough that adding Triage on cooldown increase the overall Focus burn rate, which sets a lower bound of around 16.  A Focus cost of 25-30 would make using Triage on cooldown (rather than saving it for when it’s needed) somewhat Focus punitive, without making the Healing Per Focus so low that you favor casting other abilities purely for efficiency’s sake.

These suggestions are just one possible set of numbers that I think would make Triage interesting.  You could certainly lower the cooldown and increase the Focus cost to allow a Medic to significantly ramp up their single-target healing for a short period of time, while severely draining their Focus, something akin to our ability to increase group healing when needed by spamming our spenders (Crisis Wave and Shield Surge, with Emission in between of course) at the expense of a good chunk of Focus.

As for major Tier upgrades, I like the idea of it being a source of Actuators, since you will still want them in a single-target healing role because of the need to cast Shield Surge.  The current Tier 8 upgrade of gaining additional healing on low-health targets is fine, especially appropriate for an emergency heal, although it won’t really change up our playstyle.  If T8 did something like add a short-lived buff that increases incoming healing, it would require the Medic and his teammates to react to this to gain the full benefit of the ability.  Another random thought is the base cooldown could be longer but one of the major Tier upgrades could give Triage two charges (since Medic healers don’t have any multi-charge abilities).

TL;DR

  • The burst healing from Mending Probes T8 + Flash T4 has been nerfed by 45% in 5-man stacks, slightly less-so in 20-mans. Although the nerf looks mostly warranted in 5-mans, it really harms Medics’ already limited ability to heal large (10+) groups. Crisis Wave T8 might replace Mending Probes T8 on many an LAS.
  • A Rebooted Shield Surge now hits for around 23% less. If you’ve gotten used to casting Shield Surge whenever Reboot will proc, you will be casting it 30% more often and will have to be mindful of the increased Focus burn rate.
  • If Triage is to be a viable option for an emergency single-target heal, it has to be targeted, hit for something in the neighborhood of 8k at base and either have a moderate cooldown (25sec) and Focus cost (25-30) or a lower cooldown and a higher Focus cost.

—————————-

1 2,000 Support Power, 20% Critical Hit, 175% Critical Severity.

2 If the incoming DPS to shields is D, you will go from casting once every 8.9k/D to once every 6.8k/D, which means in a given time interval T your number of casts goes from T/(8.9k/D) to T/(6.8k/D) which is an increase of [T/(6.8k/D)]/[T/(8.9k/D)] – 1 = 8.9k/6.8k – 1 = 30%.

3 This would be achieved by Triage healing for 72.51 per level plus 166.5% of Support Power (total, with a 50/50 split of this healing between health and shield).

Critical Reboot Required!

You may have read, even right here on this very blog, that AMP-based heals do not crit.  But for every rule, there is an exception, and for this rule the exception is Reboot.  As first brought to my attention on the official forums by Reglitch, Reboot, since it supplements an actual ability rather than being a separate heal in its own right (such as Hypercharge), adds to the healing of Shield Surge before the crit roll and therefore its healing can be amplified by the crit.  I have done extensive in-game testing to confirm this with the help of my ever-eager research assistant (aka, my pocket-stalker-tank wife).

The upshot is that, depending on your Critical Hit and Critical Severity of course, Reboot is around 15% better than I thought it was (and I thought it was pretty good!).  I have updated numerous posts, as well as the Stat Weight tool to account for this.  Thanks again to Reglitch for challenging the conventional wisdom.

As an added benefit, the testing showed that the natural RNG variation on an ability (Shield Surge at least) seems to be capped at +/-10%.  That is, if your Shield Surge tooltip says it will heal for 4,000, then without critting it will heal for between 3,600 (90%) and 4,400 (110%) with what appears to be a somewhat uniform distribution (more testing would be needed to have sufficient statistics to confirm the shape of the distribution).  I am guessing this +/- 10% range is universal (although it certainly could vary…), and I don’t think I’ve seen it mentioned before, so I thought I would put it out there.

Probing our OP-ness

The Devs recently lit up the official forums by announcing changes in the works for each class.  For Medics, the changes to DPS seem especially (although not universally) welcome, but there were also a few changes on the healing side.  Reboot is going to proc on targets below 30% shield, rather than zero shield, which essentially eliminates the often frustrating interaction between Reboot and various classes’ small shield heals (in addition to Stalker’s Frenzy T4 bug, which has been especially frustrating for yours truly).  30% seems a bit of an overkill, and I’m guessing it might be reduced before it hits live.  Dual Shock will be castable while casting, which is a nice buff to an ability I admit doesn’t see a lot of action in my Action Set.  It looks like they are reducing the duration of Interrupt Armor (IA) across the board, which will hit the IA buff to Protection Probes from the Solid State AMP and Rejuvenator T4.

The forum post also included the somewhat cryptic comment: “Specific portions of the Medic’s AoE healing kit are currently over-tuned and will be adjusted.”  The most likely target for this incoming nerf seems to be the Mending Probes (MP) T8 + Flash burst-healing combo.  Is this nerf warranted?  How should it be implemented?  Let’s take a look at some numbas.

Consider a standard-issue Medic: level 50, with 2,000 Support Power, 20% Crit, 175% Crit Severity, with Mending Probes T8, Flash T4, and the all three Cooldown reduction AMPs.  In order to have some context, let’s compare this to a similarly-stat’d Esper with Reverie T8 and Soothe T4.  For the Medic’s burst, we assume they put out Mending Probes and then immediately detonate with Flash.  They can do this once every 10.2sec, which is the cooldown of Probes with the assumed AMPs (we assume Flash is off cooldown because of the Flash T4 upgrade).  For the Esper’s burst, we assume Reverie with 5 Psi Points, which they can generate in 5sec with 5 casts of charge-2 Soothe T4.  Taking into account the 1sec GCD for Reverie, the Esper can perform this feat of golden-lantern goodness every 6sec.  We are only going to compare the healing on the burst, not the healing done in between bursts.  The table below compares the average heal per player1, assuming no overhealing, for two cases: a 5-man group and a 20-man group.  I’ve included two sets of numbers for Reverie depending on whether the targets are above or below 35% health (which determines whether the T4 bonus healing applies).

Ability 20-man AvgHeal/ Player (k) 5-man AvgHeal/ Player (k) “Cooldown”
MP T8 + Flash T4 5.2 17.4 10.2
Reverie T8 (> 35%) 3.4 6.9 6.0
Reverie T8 (< 35%) 4.5 9.0 6.0

Let’s first look at the 20-man case.  The Medic’s average healing per player (5.2k) is 52% higher than the Esper’s (3.4k), but only 17% higher on low-health targets, and the minimum time between bursts is 70% longer.  In addition, Reverie has a 25m range with a large circular telegraph, whereas MP T8 + Flash requires targets to be stacked up within 6m to get hit by the MP detonation splash.  The Medic burst will also have a more varied distribution of healing.  Reverie will hit 10 of the 20 targets for 6.9k (9.0k below 35%), leaving 10 of them without any healing.  MP T8 + Flash will hit 10 of the targets for 3.6k from Flash T4, detonating probes (hopefully) on 5 of those targets for an additional 3.5k, and then send out 15 splashes of 3.5k heals that might hit 15 targets, or potentially fewer (since a given target can be hit by multiple splashes) leading to more uneven healing (and likely some wasteful overhealing).  So, in summary, in a 20-man group Medic’s burst is somewhat more powerful, but on a longer “cooldown,” has a much more constraining positioning requirement, and a more random variation of healing distribution among the targets.  This doesn’t scream “overtuned” to me.

Now let’s consider the 5-man case.  The Medic’s average healing per player is now 153% higher than the Esper’s (94% on low-health targets), with the same difference in minimum time between bursts.  This is a significantly larger gap than in the 20-man case; a 17.4k per player burst heal is very (over?)-powerful.  The difference between the 5-man and 20-man cases is all due to the mechanics of Mending Probes T8.  Detonating the probes generates 20 3.5k heals, 5 from the Probes themselves and 5*3 = 15 from the splash healing.  In the 20-man case these 20 heals are distributed among 20 different targets, producing an average healing per player of just 3.5k.  In the 5-man case, you still get the 20 heals, but they are shared among just 5 targets, producing an average healing per player four times greater: 4*3.5k = 13.9k2.

If a nerf to MP T8 + Flash is coming, it seems like it would be appropriate to nerf the healing in 5-man groups without reducing the output in 20-man environments.  One way to do this would be to prevent a player from getting healed by more than one splash from the Mending Probe detonation.  For example, getting healed by one splash could apply a short-lived (a few sec?) debuff that prevents the player from getting healed by another splash heal.  If splash heals also “preferred” targets without the debuff, then when Mending Probes are detonated in a stack of 5 players, they would each get one splash heal, reducing the average heal per player from 17.4k to 10.5k, a 40% reduction.  This is still higher than Reverie, as I think it should be given the much stricter positioning requirements and longer cooldown, but it’s not 153% higher, only 52% higher, exactly the percentage difference as in 20-mans.  With this change, the average healing per player for a 20-man stack would be unchanged, since there are 20 available targets for the 15 splash heals so none would be wasted.  This change would have the added benefit of “smoothing out” the splash healing in 20-man stacks, since you wouldn’t have some targets getting hit by multiple splashes and some getting left out entirely.

Of course, Mending Probes T8 + Flash might not even be what the Devs were referring to when they talk about our overtuned AoE healing kit.  Perhaps they mean that Emission’s Support Power scaling with Tiers is so good, as are the Major Tier Upgrades, that Emission T8 is almost mandatory in a healing build.  Or perhaps it’s something else entirely, like Crisis Wave.  No, it’s probably not Crisis Wave.

TL;DR

- The coming Reboot change is very welcome, but I’m thinking the 30% will be reduced before it hits live.

- The average burst heal per player from Mending Probes T8 + Flash is much higher (3.3x) in a small group than in a large group because the number of splash heals received does not change with group size.

- If Mending Probes T8 + Flash is what the Devs have identified as our over-tuned AoE healing, they could reduce its (overpowered?) effectiveness in 5-mans without changing its (reasonable?) behavior in 20-mans by preventing players from getting hit with multiple splashes.

———————-

1 This is per player, not per target, so 20 in a 20-man and 5 in a 5-man.  Reverie hits 10 targets, MP + Flash hits potentially all 20, although I guess with bad RNG as few as 10 could get heals.  When comparing two abilities that heal a different number of targets, it seemed appropriate to normalize by the total group size rather than the number of targets.  Note this is equivalent to comparing the total healing done, but makes the numbers more intuitive because you can compare them to the average health pool.

2 The remaining difference comes from Flash, which hits 50% of the players in the 20-man case and 100% of the players in the 5-man case.

Weight, Weight…Do Tell Me

Once you’ve dinged 50 and moved from replacing your gear every few levels to gearing up for Elder Game content, you will probably start paying more attention to the tradeoffs between swapping gear piece X for piece Y, or because of creative Devs, your “Ethically Questionable Coat” for your “Chestwrap of Writhing Vines.”

Stat priorities for the healing Medic have always been pretty clear. Support Power is king because it translates directly into bigger heals, Insight is queen(?) since every point of Insight is 0.5 points of Support Power, and Moxie is next because it adds both Critical Hit Chance and Critical Severity. Brutality gives Critical Severity, but not Hit (it also increases Strikethough, but the value of that stat for a Medic is almost zero unless you are really worried about deflects reducing the effectiveness of things like the Scalpel! Forceps! AMP). Finesse (Deflect and Deflect Critical), Tech (Assault Power), and Grit (Base Health) all give things we shouldn’t care too much about. Sure having more health and deflect will increase your survivability, and you can’t heal if you’re dead (unless you are a Priest with Spirit of Redemption, but that’s a different game), but you are never(?) going to gear for survivability since by far the two most important levers you have to be more survivable are to avoid avoidable damage (player skill/knowledge of encounter), and to keep the heals coming to yourself (i.e., all those throughput stats). Focus Recovery Rate is an important stat, but folding that into a list with the others presents problems which we will discuss below.

Although stat priorities are clear, stat weights are less clear, and the subject of the current post. Yes Insight is better than Moxie, but when you start comparing two pieces of gear that have both, and asking questions like: “is it good to give up 20 Insight for 60 Moxie,” you need to know how much better. We need to transition from the ordinal to the cardinal (ask your math teacher).

Stat Weights 101

To compute stat weights we need to know how primary stats translate into secondary stat ratings (easy, Carbine tells us), how secondary stat ratings are converted to actual secondary stats (e.g., how does Critical Hit Rating affect Critical Hit Chance) and how secondary stats impact healing throughput (relatively straightforward, see this post). The rating to stat conversion needs to be deduced from in-game experimentation. As far as I can tell, the relationship is linear (i.e., no diminishing returns) for Critical Hit Rating and is equal to 0.0155%, e.g., it take 65 Critical Hit Rating to increase your Critical Hit Change by 1% (at level 50).

Critical Severity Rating (CSR) has diminishing returns, and the current best guess at the formula was worked out by Dalrea and Naokai:

Critical Severity = 221.4286% – 144.126%/(CSR/1000 + 2.01777)

The final step, translating secondary stat changes into healing throughput changes, has two complications. The first is that healing from AMPs cannot crit, although Reboot can, because it is its own special snowflake.  Therefore, the more your healing comes from things like Hypercharge and Emergency, the less valuable Critical Hit and Critical Severity are since they don’t impact these healing amounts at all. In the results below I will assume 5% of your healing is from AMP effects.

The second wrinkle is that the value of Support Power (and hence Insight) depends on your ability Tiers. At Base, a given percentage change in Support Power is worth the same percentage change in healing throughput across all abilities (as discussed in detail in this post), but this beautiful symmetry is broken when you start adding Tier points. For now we will ignore this effect, which is equivalent to assuming all abilities are untiered.

With the additional assumptions of a level 50 Medic with 2,000 Support Power, 20% Critical Hit Chance, 175% Critical Severity, and a Critical Severity Rating of 630 (which implies you have some AMP points in Critical Severity) stat weights, normalized to Insight, are as follows:

Stat Weight
Support Power 2.00
Insight 1.00
Moxie 0.51
Brutality 0.13
Crit Hit Rating 0.76
Crit Severity Rating 0.27

The first takeaway is Insight is worth about twice as much as Moxie. Note the stat weight for Moxie is just the average of the stat weights for Critical Hit and Critical Severity (obvious from the definition of Moxie), and Critical Hit is worth almost 3x Critical Severity. Brutality is almost worthless given it only gives Crit Severity (with a 0.5 multiplier) and a stat that doesn’t help us heal.

Stat Weight Calculator

Now that we have some idea what stat weights look like, we get to what you probably really want, which is a calculator for computing stat weights given the details of your character, LAS, and ability usage. The ability usage data is necessary to properly evaluate Support Power (and hence Insight), and I recommend using a healing meter (e.g., GalaxyMeter) to measure this. To get your Critical Severity Rating you can mouse over the Critical Severity Percentage in your Character Sheet and the rating will be at the bottom of the tooltip.

The sheet is here, and I think it should be pretty self explanatory. Inputs you should change are in green, data you shouldn’t change are in red, and everything else is formulas.  If the sheet is hard to read (diagonal grey bars almost everywhere) just uncheck View -> Protected ranges.

The sheet has a toggle so you can choose to include the value of approaching the next Milestone in the Primary Stat weights.  For example, if you are at 600 Insight, 90 more points of Insight will get you to the next Milestone and give you a bonus 20 Support Power.  Therefore, in this range, the value of each point of Insight is increased from 0.5 Support Power to 0.5 + 20/90 = 0.72 Support Power.  If you prefer to ignore Milestones except when the gear swap you are considering takes you through some, you can just toggle it off.

Focus Recovery

Focus Recovery is certainly a very useful stat, but it is difficult to compute a relative weight for it versus other stats since it increases longevity whereas the others increase throughput. The two are certainly related: you can generally increase your throughput by using more Focus-intensive abilities, but that obviously reduces your longevity and vice versa. Valuing this tradeoff, in general, is difficult. For a given encounter, if you have enough Focus to last the entire fight, even casting Focus-intensive abilities, but not enough throughput, then no amount of Focus Recovery is going to help you. In fact, you should probably replace some Focus Recovery on your gear (e.g., from Runes) with Insight. Alternatively, if you are able to keep up with damage on a fight but find yourself going OOF mid-way through, you clearly need to cast less Focus-intensive abilities, somehow reduce incoming damage (e.g., changing strats or just being less bad), or move some Insight into Focus Recovery Rate. So, in general, there is no “right answer” for the tradeoff between Focus Recovery Rate and other stats: if you find yourself often Focus constrained, get more of it, if you typically end fights swimming in Focus, get less of it.

The calculator linked above does allow you to get some idea of the relative importance of Focus Recovery Rate versus healing throughput stats, but requires you to define the tradeoff, in terms of what percentage healing you are willing to give up to get an extra +1.0 Focus Per Second.  How much does +1.0 Focus Per Second really buy you?  Here is a handy formula: if your total Focus pool is F and you are currently running OOF after around T seconds, an extra +1.0 Focus Per Second will increase your longevity by T/(F/T – 1).  For example, if you have F = 1,000 total Focus, and you currently run out of Focus after T = 5min = 300sec on a given encounter, an extra +1.0 Focus Per Second will allow you to last an extra 300/(1,000/300 – 1) = 129sec, so your longevity goes from 5min to a little over 7min, which is clearly a big change.

TL;DR

- Typical stat weights are given in the table above. Insight is our best stat (outside of Support Power) by a decent margin.

- More detailed values for your specific character/LAS/ability usage, can be obtained from this calculator.

To Tier 4, or Not To Tier 4 (Shield Surge)

Shield healing is one of the things that sets Medics apart from the other two healing classes (along with, of course, having by far the coolest weapon).  In this post I want to look at Shield Surge Tier 4, whose major upgrade increases shield mitigation by 25% for 8sec.  However, this will first force us to take a careful look at shield mitigation and the general tradeoffs between shield healing and health healing.  If you have a firm understanding of shield mitigation from my earlier posts, you can skim/skip the first section (although you will miss the cool graph).

Shield Mitigation

By default, shields take 50% of the incoming damage and pass the other 50% through to health.  The passive part of a Medic’s innate (Energize) gives a 15sec buff that increases shield mitigation by 25% whenever you land a shield heal.  So provided you are regularly casting Shield Surge, a Medic’s target will likely have a “base” shield mitigation of 75%.  The Buffer buff from Shield Surge T4 will take this to 100%.

There are still some comments on the official forums and elsewhere that talk about shield healing being only 50% (or, including Energize, 75%) “as effective” as health healing, implying that the shield mitigation number in some way translates directly into how “good” shield healing is relative to health healing.  This is simply not true, as I argued in my first post, shield healing is almost always just as good as health healing, even when shield mitigation is less than 100%.  By “as good as,” I am using a metric of how much damage the target can take before dying, which I will call “Effective Health” (EH for short).  Now, “almost always” is not always, and I will explain the exceptions to the general rule below.

At the risk of belaboring the point, here is a little thought experiment that drives home the central idea from my original post:

Suppose a player has 1,000 health and no shield.  You would say she has a health pool of 1,000.  Now imagine something (say an add-on) always colors 50% of the remaining health bar Green and 50% Blue, no matter what the total remaining health.  Initially she has 500 Green Health and 500 Blue Health.  What happens as she starts taking damage?  Well, if she takes 100 damage, she is down to 900 total health, 450 Green, 450 Blue.  If she takes another 200 damage, she is down to 700 total health, 350 Green, 350 Blue, etc.  The Blue Health pool is absorbing 50% of the damage and passing 50% of the damage through to the Green Health pool.  Note this is absolutely identical to a player in WildStar with 500 health, 500 shield, and 50% shield mitigation.  So saying shields are not a “real” health pool at 50% mitigation is equivalent to saying the “top half” of a player’s remaining health bar isn’t “real” health because it only ever absorbs half the incoming damage (passing the other half through to the “bottom half” of the remaining health bar).

So even at 50% mitigation, shields are equivalent to health?  Most of the time, yes, but not always.  They are equivalent in the above analogy because we never considered healing (or other means of hit points regenerating), so the remaining shield (i.e., Blue Health) was always exactly equal to the remaining health (i.e., Green Health).  The equivalence is broken because you die when your health is zero, no matter what your shield is.  This brings us back to the concept of Effective Health, which is defined as  “how much damage can the target take before dying.”  If the target has current health H, current shield S, and shield mitigation M, then Effective Health is:1

If the ratio of current health to current shield, H/S, is above this “breakpoint,” (1 − M)/M, then Effective Health is just the sum of current health and current shield, which makes intuitive sense.  In these cases, you won’t ever die with your shields up in the next hit (after the next hit, we would need to reevaluate if H/S was still above the breakpoint or not).  If the ratio of current health to current shield is below the breakpoint, Effective Health does not depend on the current shield level S.  This is because we are in the zone where the target’s health is low enough that she can be killed in the next hit while still having some shields left.  If you are in this second case, or more correctly, if you will be in this second case (H/S < (1 − M)/M) after the next Shield Surge you are thinking of casting, you might want to rethink it because that additional shield will not move your target any further away from death (i.e., it will not increase her Effective Health at all), whereas a health-heal certainly would.

Since we are approaching the 1,000-word mark, it’s time for a picture.  Let’s imagine a target whose max shield is 30% of her max health, which I believe is a pretty representative value.  The plot below shows the possible values of the target’s current Health percentage (x-axis) and current Shield percentage (y-axis).  At any point in time, the current state of the target corresponds to one point in this (unit) square.  Given the Effective Health discussion above, we can divide this square into regions where shield healing is useful for increasing Effective Health, and where it is not.  I will assume a mitigation value, M, of 75%, which represents the “typical” shield mitigation for a target who is being frequently (at least once every 15sec) healed by a Medic (without Shield Surge T4).  The green region is where shield healing is as effective as health healing in increasing Effective Health, and the red region is where it is not.

TT4ONTT4

Note the only assumptions I’ve made are the ratio of max shield to max health, and the shield mitigation percentage.  The triangular region where shield healing is ineffective is pretty small (for the geometrically inclined, it’s just 5% of the area of the square).  The implications of this for Shield Surge T4 will be discussed below.  One final caveat: we have ignored overhealing considerations in this discussion.  For example, if you are deep in the green region above, where shield healing is effective, it still might be better to cast a health heal over a shield heal if the target is at 95% shield and 50% health.  But hopefully that is obvious.

Shield Healing vs Health Healing

Before delving into Shield Surge T4, let’s discuss a couple of basic things about health healing and shield healing.  First, in general, shield healing is less Focus efficient than our “baseline” health healing abilities.  The table below compares Healing Per Focus (HPF) for our primary shield heal to some of our health heals, assuming a level 50 Medic with 2,000 Support Power, 20% Critical Hit, and 175% Critical Severity healing a single target.

Ability HPF
Shield Surge T4 92
Shield Surge T4 w/Reboot 160
Emission T8 347
Mending Probes T4 286

The healing abilities I’ve included for comparison are what I believe are typically used to do the bulk of baseline health healing on a target taking sustained but not spikey damage (e.g., the tank in your 5-man taking mostly white hits from a trash pack)2.  We see Shield Surge T4 is only about 30% as efficient as Emission T8 and Mending Probes T4 without Reboot, and still only around 50% as efficient with Reboot.

Of course, the compensation we get for this lack of Focus efficiency is shield healing allows us to effectively maintain our target’s Effective Health at a significantly higher level (around 30%) than we would be able to if we didn’t heal shields at all (in which case they would very quickly be gone, and on a tank, highly unlikely to ever naturally regenerate during combat).  Maintaining a higher Effective Health makes overall healing easier because it gives you more of a buffer to deal with any combination of: a) damage spikes on the target (via mechanics or player mistakes), and b) being unable to heal the target for a period of time, either because of range issues, cc issues, or you being away healing someone else (or prying the cat off your keyboard).

So to review: shield healing, even with Reboot procs, is less Focus efficient than our baseline health heals, but it helps maintain a higher Effective Health, which can be hugely beneficial.

Shield Surge T4

We are finally in a place to take a good look at Shield Surge T4.  I am going to focus attention on situations where the Medic is the only healer present (e.g., 5-mans), and hence the primary source of all healing.  Assuming Shield Surge is cast regularly, Shield Surge T4 increases shield mitigation from 75% to 100%.  This has three major effects:

1) It makes shield healing always useful in increasing Effective Health.  In the graph above, the red region disappears; at 100% shield mitigation casting Shield Surge will always increase Effective Health even if your target is at 1% health,  since it is now impossible for them to die with shields up.

2) It makes shields deplete faster by 25%, since they are taking more damage.  If the rate of incoming damage was such that the shield was broken in 4sec by incoming damage, it will now break in 3sec.  This means that to maintain shields, you will be casting Shield Surge more frequently and likely procing Reboot more frequently.

3) Until the shield is broken, there is no damage to health.  This means that Emission, which we need to cast at some point to build Actuators for Shield Surge is going to be 100% overhealing except for the damage that gets through in between the time shields are gone and we Reboot them.

The first effect is unambiguously a positive, but how big a benefit is it?  It will help when your target is at very low health, but hopefully you seldom find yourself in that spot.  Of course, even if it only helps once in a dungeon run, it may be the difference between a wipe and a kill (or a medal and no-medal), so just because it doesn’t make a difference often doesn’t mean it can’t make a big difference.

The net effect of the second point is a little subtle and depends on your Shield Surge strategy.  We will consider two cases: casting Shield Surge only when Reboot will proc, and casting Shield Surge whenever it would result in zero overhealing.  The second strategy is clearly more Focus intensive but maintains a higher Effective Health.  In the first case, 100% of your Shield Surge casts will proc Reboot (by definition), and therefore the only difference between Shield Surge T4 vs not T4 is casting more often and hence burning more Focus.  For the second case (Shield Surging as often as you can without overhealing), I’ve run some simulations and it again appears T4 results in a higher Focus burn rate (the cost of more frequent casting negates the “savings” from a higher percentage of Reboot procs).3 How much higher depends on how much effort you put into maintaining a high amount of shields, which will of course depend on whether you are trying to prepare for a damage spike (or just the unexpected).

The third effect is really the flip side of the second: in addition to spending more Focus on less-Focus-efficient shield healing abilities you have less opportunity to leverage the efficient health healing from your builder, which again, you have to cast at some point for Actuators, whether there is a health deficit or not.

Bottom-line, I think Shield Surge T4 hurts Focus efficiency in exchange for eliminating the “red zone” in the graph above, making shield healing always useful for increasing Effective Health, even on very low-health targets.  It also costs 5 (or 8, depending on how you count) Tier points, which are certainly a precious resource.  On balance, is it “worth it?”  I think that is open to debate, but I think it is less useful than some people seem to think.

TL;DR

- Shield healing is almost always just as effective as health healing in terms of increasing Effective Health, even without the Tier 4 major upgrade from Shield Surge.  The exception is when the target is below around 10% health, in which case there is a danger of them dying with shields up in the next hit.

- In general, shield healing is great because it allows Medics to increase the maximum Effective Health of their targets, but at the expense of being less Focus efficient than our baseline health-healing abilities

- If you spend Tier Points for Shield Surge T4, you are reducing your Focus efficiency in exchange for making shield healing effective on low-health (< 10%) targets.  Whether or not that is a good trade is debatable, but that is the trade.

—————–

1 The term “Effective Health” is often used in a different context, combining the size of a health pool and damage mitigation percentages (e.g., armor, resistances, etc.)  to allow one to compare the tradeoffs between increasing the overall number of hit points and damage mitigation stats.  In the context I am using it, I am ignoring damage mitigation since health and shields are both subject to the same damage mitigation stats (as far as I can tell).  So mitigation doesn’t impact the discussion above at all; if you like you can assume all the damage I am talking about is “post mitigation.”

2 Additional assumptions made: you are above 250 Focus, so the T4 major upgrade on Emission is not triggered, you are triggering the Emission T8 major upgrade 80% of the time, and Mending Probes are not detonated.

3 To get some idea why this is the case, we need to rock a little math.  Let S be the healing of an average Shield Surge, R be the amount of healing done by Reboot, xT3 and xT4 be the fraction of Shield Surge casts that proc Reboot assuming T3 or T4 Shield Surge, respectively, and cT3 and cT4 be the rate of Shield Surge casts per second assuming T3 or T4 Shield Surge, respectively.  In the T3 case, the shield HPS is cT3*(S + xT3*R), and analogously for T4.  Now, in steady state (and assuming no overhealing) the shield HPS has to be equal to the incoming shield damage rate, and given shield mitigation is 75% for T3 and 100% for T4 we must have cT3*(S + xT3*R) = (75%)*cT4*(S + xT4*R).  We are interested in cases where cT4 < cT3 (i.e., we are casting Shield Surge less often with T4), which implies (after some algebra): xT4 > (1/3)*(S/R) + (4/3)*xT3.  Using the known values for S and R (and taking into account the fact that S is a little different between T3 and T4), this becomes xT4 > 40% + (4/3)*xT3.  So for example, if you were getting 25% (= xT3) of your Shield Surges procing Reboot with T3 Shield Surge, you would have to get 40% + (4/3)*25% = 73% of them procing at T4 (assuming the same rate of incoming damage) to net be casting Shield Surge less often and hence burning less Focus per second.  This is a high hurdle to overcome without resorting to only casting Shield Surge when Reboot will proc (in which case xT4 = 100%), but that contradicts our assumption that you were casting non-Rebooted Shield Surges to maintain higher Effective Health.

10 Tips for Surviving Early Dungeons

It seems like everywhere you look these days: the official forums, reddit, or across the blogosphere (and don’t forget the twitterverse!),1 there are lots of discussions about WildStar’s first two dungeons: Stormtalon’s Lair (STL) and the Ruins of Kel Voreth (KV).  This post will provide some tips to help Medic healers survive their first taste of WildStar’s challenging 5man content (I haven’t tried Riot in the Void, but judging from Hycrest Insurrection I am taking the view that the first Adventures don’t count).  This is by no means a comprehensive guide to the mechanics of the bosses and trash mobs; there are plenty of solid ones out there already.

Succeeding in these instances does require some level of competency among all players present, but you don’t have to be uber-skilled or geared to finish them, you just have to be prepared to wipe some and learn from your mistakes.  For a data point, I went in the first time to both with around 600 Support Power and 2.4 Focus Recovery Rate at Level 20 (with my Stalker Tank wife at around 300 Support Power), and three PUGers, and felt I had sufficient throughput and Focus efficiency to deal with everything the dungeon threw at us.  If you have a Support weapon (i.e., paddles with Support Power > Assault Power), and so does the tank, things should be doable (in principle).

As previously discussed, a core Level 20 build would be:

Emission T3, Crisis Wave T3, Flash T2, Mending Probes T3

Leaving you three free LAS slot for some combination of: Urgency, Recharge, Paralytic Surge, Antidote, Barrier, Assault abilities, etc.

1. Flank your Tank

I find the best “default” position is to be to one side of the tank (facing them of course).  If you imagine a clock face with the boss/mob in the middle of the clock and the tank at 12 o’clock, you should either be at 10 o’clock or 2 o’clock.  This allows you to: heal the tank, stun/DPS the boss/mob, see what telegraphs are headed the tank’s way, and avoid having to move for the frequent frontal-cone shaped telegraphs aimed at the tank.  If the tank needs to dodge a simple telegraph she will likely either strafe toward you or away from you, rotating around the boss, and you should just follow her forward or backwards to maintain the same relative position.  Obviously you will often have to interrupt this dance to get out of a telegraph yourself or get in range to heal a DPS, but after doing either of those you should attempt to reset to this default position.  Position matters a lot in WildStar, and requires more of your mindshare than in MMOs with targeted abilities.

2. You Know It’s Urgent

Speaking of positioning, your short range is your Achilles’ heel, but that can be somewhat counteracted by attaching some wings to those feet every 20sec.  Urgency especially shines on those encounters whose mechanics generally cause groups to spread out: Grond (due to his spinning knockback, Thrash, and the need to kite adds into traps) and Forgemaster Trogun (the combination of players chasing down green orbs and fire everywhere tend to make this fight pretty chaotic from a group positioning perspective).

3. Don’t Emit Defeat

Emission, if spammed carelessly, will drain your Focus pretty rapidly.  With all due respect to Alec Baldwin, you should not Always Be Casting (ABC); if the tank is close to full health and doesn’t appear to be taking a lot of incoming damage, stop casting for a second or three and then shower her with your healing goodness.  Small pauses will, over time, lead to greater longevity.

4. Recharge early, Recharge often

Recharge is not your “oh sh*t I’m OOF, better call in the relief pitcher” button.  If you are using Recharge for the first time in a pull when you are out of Focus you are doing it wrong.  Focus has a 45sec cooldown (lower if you have the Tier 1 Cooldown AMPs) and should be used as early and as often as possible in a fight provided you are not so flush with Focus that some of the Focus gained will be wasted (at base it restores 60 Focus over 5sec, so if you are below around 925, Recharge away).  After this first Recharge you will likely use it on cooldown until the end of the fight unless you are distracted by intense healing (although remember in exchange for the 2sec cast time you get a 15% Support Power boost for 10sec to compensate).

5. Know how to Burst

When you need to really pump out the heals, I recommend: Crisis Wave x2, Energize, Flash, Crisis Wave x2.  This should bring health bars from a pretty low level back to almost full for all five targets (if they are in range of course).  Note Flash comes after Energize so it can also benefit from the 17% Support Power buff from our innate.  This assumes you are at full Actuators to start, which won’t always be the case, but if your need for burst healing is coming from a predictable source of damage, you can try and line this up once you know the encounter.

6. Bring some Deeps (rarely)

I recommend keeping Empowering Probes up (once level 21) to help with DPS, unless there is a DPS Medic in the group doing this for you.  Other than that, the only encounter I feel it’s good to have some Assault abilities in your LAS is Slavemaster Drokk in KV, to break your tether (this is assuming you aren’t high enough level to access Calm).  Otherwise, you have to wait for a kind DPS to come help you, and by that time someone may have eaten a Suppression Wave (or three) and be in desperate need of a heal.

7. Don’t Forget the Dispell

The only debuff that is just about essential to dispel is the group-wide DoT from the Voreth Darkwitch trash mob in KV.  Letting this tick too long will certainly wipe your group, so make sure Antidote is handy if any Darkwitches are patrolling nearby.  Note, seeing the debuff is pretty challenging with the default UI, since it only shows up on your character/target frame (along with the dozen other buffs/debuffs you are likely to have active) and not anywhere on the party frame.  Although Bijiplates (What?  You don’t have Bijiplates?  Go get it.) has a “cleanse alert” checkbox, I’ve never seen it alert me to a cleanse (anybody got this to work)?  I keep Grid up off to the side to monitor who has debuffs (it has other utility for classes with targeted heals but as a Medic I don’t know what those are).

8. Barrier to Entry

On Blade-Wind the Invoker, the first boss in STL, there is a Subdue (i.e., Disarm) at the phase transitions (first one occurs at around 85% health, second one after killing the last add) that can cause some problems if people are taking damage while you are trying to figure out whether that glowing mass of blue-white stuff you are sprinting toward is a pair of paddles or claws.  Remember Barrier gives Interrupt Armor to up to three targets, and Protection Probes with the Solid State AMP gives Interrupt Armor to five.  This will prevent the Subdue and completely remove one source of potential wipes.

9. Potent Potables

Hopefully your tank has some Serpentlily Instant Healing Medishots (which heal for 2.8k) buried somewhere in her bags (if not, you can instantly establish some goodwill by offering her some).  You can bring some Serpentlily Medisprays for yourself which are basically AoE healing potions that heal yourself and up to 5 targets within 10m for 2.2k.  Note the debuff from the Medispray will not prevent the tank (or anyone else you heal, except yourself) from using another Medishot.  The Instant Medishots are cheap and easy for an Apprentice Technologist to make, so they should be readily available on the AH.  The Medisprays are a little more costly, but not exorbitantly so, especially if they spare you the repair cost from a wipe.

10. With a Little Help from Your Friends

If you are struggling with raw healing throughput, which I think is most likely on either the add phase of Aethros in STL or KV’s Forgemaster if he eats a few too many green orbs with his Wheeties, it’s fine to ask if a DPS can throw out an offspec heal or two (e.g., a well-timed Phantasmal Armor can be a tank-saver).  For the Aethros adds, another way to reduce incoming damage is to chain stuns through the group, provided you do this so enough are off cooldown to interrupt Aethros himself after the gauntlet.  The whole point of the dungeon is to accomplish something as a team, right (even if it’s a team of mostly strangers)?

If you have other tips and tricks, I would love to hear them in the comments.

————————

1 There should be a name for the collection of discussion forums/news sites/blogs/twitter feeds, etc. that you monitor to keep up to date on topic X.  The “internet” seems a little too broad (and dull), we need something catchier.