Topic The New Power Word: Solace
Edited by Twistedmind on 3/31/13 9:16 PM (PDT)
The following is my analysis of the new Power Word: Solace talent on the patch 5.2 ptr. If you don't wish to know how I came up with my reasoning and simply wish to know what the conclusions are, they are:
The following is taken from data I collected today in the ptr:
Edited by Twistedmind on 1/13/13 4:52 PM (PST)
What this new thing is:
The new Power Word: Solace replaces Holy Fire (like Mindbender replaces Shadowfiend). It stacks Evangelism and if specced Discipline will proc atonement. It both removes the mana cost of a holy fire (5400 to 3780 mana depending on your current Evangelism stacks) and also gives back 1% of our total mana (this is affected by the 2% meta gem). This also is the replacement to the old holy fire glyph such that it makes the spell an instant cast. If you are not talented into this new Power Word: Solace, your Holy Fire spell will not be an instant cast and will revert to its original 2 second (before haste) cast time. The new solace uses the old solace animation.
While at first glance this might look like a great spell (or the return of the not-so-great old pally judgement), this new talent is actually problematic for both disc and holy. I have a lot of mathematical explanations and conclusions, each with a link to the appropriate section:
Disc Mana Comparisons: http://us.battle.net/wow/en/forum/topic/7592579788?page=1#3
Holy Mana Comparisons: http://us.battle.net/wow/en/forum/topic/7592579788?page=1#4
Holy Healing Per Second Comparisons: http://us.battle.net/wow/en/forum/topic/7592579788?page=1#5
Disc Healing Per Second Comparisons: http://us.battle.net/wow/en/forum/topic/7592579788?page=1#6
Potential Damage Comparisons: http://us.battle.net/wow/en/forum/topic/7592579788?page=4#64
Conclusions Drawn From Potential Damage: http://us.battle.net/wow/en/forum/topic/7592579788?page=4#68
What's All This Mean?: http://us.battle.net/wow/en/forum/topic/7592579788?page=1#8
What Needs to Happen: http://us.battle.net/wow/en/forum/topic/7592579788?page=1#9
Edited by Twistedmind on 1/6/13 8:35 PM (PST)
Disc Mana Comparisons:
The mana returns of 1% per cast, on a perfect 10 second cast in your rotation will net 6% mana per minute. Shadowfiend once every 3 minutes will net 8% mana per minute. For comparison, Mindbender every minute is 14.61% mana per. Every 10% of player haste, mindbender will gain another 1.461% mana swing per minute. So on paper, this looks balanced because both are 14% mana per minute, right? But it's important to note that the mana saved from not spending mana on a holy fire is not included in these percents. Also, for scaling, holy fire will always be fixed at 10 seconds, but mindbender will get an additional swing in at the 10% breakpoints (shadowfiend gets an additional swing in every 12.5%). Also, some encounters will get more mana from shadowfiend because of the length and the number of possible shadowfiend casts vs. the number of mindbender casts (fights that are just under 1 minute, 4 minutes, 7 minutes, ect.).
So what does this mean from a mana perspective? It means that the new Power Word: Solace is superior regen under perfect play. That much should be a given just by eyeballing the above percents. But casting solace later than the 10 second cooldown will adjust the mana gains. I wanted to know how late could I wait to cast a holy fire to get the matched mana gain + mana saved (with shadowfiend included) to stand against mindbender. Below is a chart of how often you need to cast Solace to match Mindbender's mana regen, depending on how many stacks of Evangelism you have (this includes Shadowfiend):
As you can see, even with terrible timing on pressing holy fire every 20 seconds (over 20 seconds per holy fire will also drop all of your evangelism stacks if you aren't casting smites or penance), you will be netting more mana from the mana returned+mana saved than mindbender. But say, what does this chart look like when mindbender gets an additional melee swing? Or when shadowfiend gets an additional melee swing?
Even in between the 10% and 12.5% plateaus where mindbender should be at its maximum potential dominance over the shadowfiend+solace combo, you can play with a poor ~18 second per Holy Fire and net the same mana as using Mindbender. But what about those special fights that favor shadowfiend over mindbender because of the fight length?
Poorer play is excusable under those conditions as well. It's really important to note that all of these calculations were made under the assumption that you are normally casting Holy Fire. If this is something that is left out of your spell choices, take these figures with a grain of salt.
Edited by Twistedmind on 1/6/13 8:43 PM (PST)
Holy Mana Comparisons
But what does this talent do for holy? Holy would normally not be casting holy fire because the spec does not have the added benifit of Archangel or Atonement. We must instead look at the mana savings from the perspective of the passive mana gained from taking time out of our rotation to cast these free spells. What I mean is if we're following the rule of always be casting, then casting Solace will save us certain amounts of mana per minute from our mana gained from spirit depending on how many globals we used per minute casting them. This number is also affected negatively by haste because it will speed up the globals, meaning reducing the time spent per minute casting mana-free spells. For the following I've chosen to model this at 8000, 10000, 12000, and 14000 spirit as well as choosing to model the 0%, 10%, 12.5%, and 16.6% haste breakpoints. The number of casts is also a factor, so I'm showing the behavior of every 20, 15, 12, and 10 seconds for casting Solace.
As you can see, the passive mana gains go up and up. Don't look at these numbers and say, "omg! 20k extra mana per minute! That's OP!!!11!" These numbers are a calculation of passive mana gained while casting solace. If you weren't casting solace, you'd be casting other spells that would be taking away from this passive mana gained. Spells such as...Prayer of Healing? Or simply your average mana spent per second over the encounter. For reference (and in case you want to visit the above table and figure this out) prayer of healing eats up 8100 mana per 1.5 seconds before haste. You would have to add the mana you would normally spend from the above values to find that the amount of mana you save from casting PW:Solace in their place. It should be clear that this is extremely powerful. This is assuming that you're casting moderately expensive spells (read, not heal) and that you're following the rule of always be casting, which does not include cancel-casting to either minimize overhealing or maximize healing per mana. So really, this talent is best for holy from a pure mana standpoint. Without the incentive for atonement/archangel, this is purely a hit this button and get mana mechanic. You should adapt to using this by either filling up the time you spend cancel-casting spells with pressing this button, or if you just wait for damage to happen and cast (having some dead time in your rotation but being ready to react quicker), cast this instead.
Also, if the fight has downtime where healing isn't as needed, these Solace casts can be cast with no assumed penalties (I'm not including preemptive rolling echoes of light or renews on people who might take damage). However, because Holy Fire is on a 10 second fixed cooldown, this new Solace is actually much more inflexible compared to the old version. So ultimately, matching solace spam to the down periods of the encounter was actually better with the old incarnation of this talent. But this new tier might have some boss where the down period for healing is every ~10-20 seconds for a brief instant (doubtful).
Edited by Twistedmind on 2/23/13 7:52 AM (PST)
Holy Healing Per Second Comparisons
*The following data is invalid now that solace provides healing when pressed. The amount of lost/gained throughput is now entirely depending on your spell selection choice. See the next post for a more accurate measurement of how this is affecting holy.
So the mana gains from this new Power Word: Solace is superior to Mindbender (accounting for shadowfiend usage) for both specs in all scenarios (accounting for the shadowfiend/mindbender haste breakpoints and the number of casts per minute). What about the healing per second gains/losses from using this spell? I'm going to look at holy first because every Solace cast is time away from healing. Here's a chart of 3-6 casts per minute at the previously listed haste values:
As you can see, this is quite a substantial loss in any theoretical healing rotation you could be doing assuming that you will always be casting and there will always be something to heal. Something to note about Mindbender and Shadowfiend in the above tables, I subtracted them from the values listed above because I have the weapon that makes them off the global cooldown. Should you want to include them in the above calculations, subtract the 1 gcd (modified by haste) for mindbender and add the 1/3 gcd for shadowfiend (in addition to the solaces). If the intention behind the legendary this expansion is to keep building on the weapons we have, I'm going to assume that our pets will be off the global cooldown until next xpac. Pvp would still be casting pets that are affected by this gcd, but pvp'ers really shouldn't be looking at power word: solace because the damage portion (and therefore atonement) is not affected by pvp power (subject to opinion, of course).
Edited by Twistedmind on 1/6/13 8:56 PM (PST)
Disc Healing Per Second Comparisons
Looking at hps is a little trickier for discipline because the spec has Evangelism and Atonement. Each cast of Solace is doing 3 things:
1.) It's doing a heal (that changes in value depending on your current Evangelism count) that shifts the ratio of the theoretical max throughput a little lower because holy fire isn't maximum throughput. In other words, if you're casting a Solace, that could have been 1 gcd spent on a bigger spell.
2.) It's increasing your hps by giving you a stack of Evangelism, that will later pay off as a 5% increase to your hps later during archangel.
3.) Talenting into the talent will give a slight increase to the hps of Holy Fire because of the cast time reduction from going from 2 seconds to instant (1.5 seconds) as well as freeing up a little more available time per minute for more hps.
The chart above for holy does a good job of showing the lost gcd's per minute, but we can't treat these as "lost" globals because the spell is still doing some healing through atonement. How much or how little depends on which spells you would have cast in place of the Solace. Drafting up a chart to illustrate this must account for personal spell choice, which varies too much from person to person. So instead, I'll say that the new Solace will do 44,565 to 53,479 healing per second (depending on Evangelism) with my gear, compared to Prayer of Healing which does 128,173 hps, Greater Heal which does 56,437 hps (did not account for train of thought/inner focus), and Penance which does 61,403 to 73,683 (accounting for Evangelism) hps. I'm not showing grace or archangel here because Atonement heals can also benefit from grace and archangel. Obviously getting Atonements to land on graced targets is difficult, so for eyeballing the previous spells you could multiply 1.3 on the single target spells or 1.06 (add 0.06 per additional graced target) for prayer of healing. This would give more weight to them and show a greater discrepancy in hps between them and pw:solace. Without Grace, and if we average the numbers from Evangelism, Solace is doing 32% of the healing compared to Prayer of Healing, 87% of the healing compared to Greater Heal, and 73% of the healing compared to Penance.
The amount of healing per second increases from archangel that casting each of these solaces is also going to change depending on how often you use archangel. If we remove the Penances and Smites from your rotation, we can draft up a chart that shows how much of an overall gain each Solace is giving depending on both how many Solaces per minute you're casting and how often Archangel is being cast. All of the following assume you didn't derp and drop off your Evangelism stacks:
So if you don't let your Evangelism stacks expire and you make use of archangel, you are increasing your theoretical hps by 1.5%-3% every time you cast Solace.
Edited by Twistedmind on 4/1/13 11:02 PM (PDT)
Disc Healing Per Second Comparisons Continued
*Edit: The following math was done when Holy Fire was a 2 second baseline. Now that it is 1.5 seconds, the following is invalid. However, I'm leaving it up for the sake of knowledge.
The reduction in cast time from going Holy Fire to Solace to see the hps gains is actually trickier than simply subtracting the 2 seconds from the 1.5 seconds modified by haste. This is because you can actually shift your healing per second numbers around with the number of smites cast during the Holy Fire/Solace dot if you are using the Glyph of Smite. What many don't know is that making Holy Fire an instant cast means that the dot loses 1 gcd from the possible 7 seconds of casting smites because of the instant cast nature of the spell. The dot is applied at the beginning of the spell cast (at the start of the 1.5 seconds global). If you don't talent Power Word: Solace, your Holy Fire dot will start ticking at the end of the cast, leaving you with the full 7 seconds to cast smites during it. However, you must chain cast those smites, which often isn't as advisable because of all the other "better" short term cd spells that you could be casting as a Disc priest.
The argument could be made that you're chain casting smites after a holy fire to build stacks for archangel as quickly as possible, which does see some situational use. But mostly, chain casting smites after holy fire is done if the priest is maximizing their damage output, the bonus healing is simply a bonus (and you should still break the chain casting of smites to cast a Penance as the priority order of damage per cast is Penance>Holy Fire>Smite). And one more variable to account for: there is a soft cap for fitting in more smites per Holy Fire dot. This number is 7.1429% haste (not accounting for latency), then 28.5714% (which is the 7.1429% under power infusion). If you are talented for Solace, your haste breakpoints are the exact same as the previous numbers (7.1429%), you will just have 1 less smite regardless because you're losing exactly 1 gcd from the Solace cast.
So first for comparing these, I have a chart showing the time saved per minute at our favorite haste levels and the number of Solace casts per minute:
So the amount of time Solace is saving by being an instant cast, assuming you'd be casting a holy fire anyway, is a moderate increase that gets better if you cast holy fire/solace closer to cooldown.
For hps with the healing gained from maximizing smites during the dot portion of holy fire/solace, it depends on the individual's gear and haste scaling. I have a few models for maximizing dps, so stripped out all the variables and modifiers, and looked at the rotation at 30k spellpower for both 0% haste and 7.1429% haste (I removed Evangelism and Critical Strike Rating as they would be equal theoretical percentages):
So the difference in healing done from going from reaching the haste cap for getting the extra smite during the holy fire/solace dot (7.14% haste), the gains are 8.74% increase hps. These rotations are the balls to the wall, max dps rotation.
Edited by Twistedmind on 2/23/13 8:11 AM (PST)
What's All This Mean?
Overall, you should be thinking that holy fire is on a 10 second cooldown, yet you only really need to cast it every ~20 seconds to match the mana savings of mindbender. This is overly good for disc because with both Archangel and Atonement, these are all positive gains for your healing per second, mana gained, and healing per mana. Even in the most strenuous encounters, the mana savings from casting holy fire and the gains from archangel are enough to justify this talent over mindbender or from darkness comes light. This is not only taking into account the mana that would otherwise be spent on holy fire, but also taking the data we found when looking at this from a holy perspective of more passive mana gains. If that wasn't enough to convince you, the healing per second gains from taking this talent are tremendous. If we assume a 12% time spent per minute casting this spell, the conflicting spells I listed in point 1 average to 64% of the possible healing that could be cast at the time of casting each solace. Now this isn't a concrete number as I didn't list a full spells breakdown, and the scaling I'm examining is heavy mastery, but bear with me, I'm just trying to illustrate the potential drawback in hps first. Those factors considered, we're suffering a 4% loss in hps. Now, add back in points 2 and 3. Point 2 illustrated that each solace cast will yield a linear 1.5%-3% increase in overall healing per minute per cast (archangel usage from 30 sec to 60 sec). This means a wild 4.5%-18% variance, but that's just from archangel gains. We need to account for the fact that we'd probably be casting holy fire anyway (and you should with a possible 18% gain from it), so this variance isn't simply a 4.5-18% increase in your hps for taking Solace. It's a 4.5-18% increase, minus the archangel percent increase. This isn't a simple number to find because we're starting to juggle so many variables. I tried to account for some of these by illustrating the gains made from the potential smite gains at the soft haste cap. For me, and this is really estimated, entirely subject to napkin math and my summed breakdown of my hps during garalon, this talent will pull a healthy ~6% theoretical increase to your hps at the expense of a more rigid spell selection. And the question is, will this be more of an increase in hps compared to mindbender or from darkness comes light? For this example, the answer is yes. But I'm still looking to use mindbender on fights that demand more aoe spell selections and from darkness comes light on heavy tank damage fights.
For holy, you really wouldn't be casting holy fire anyways in a pve setting, but solace is the talent that must be taken for maximizing mana, so long as you can spare the globals casting it. If you aren't spamming aoe's (non-aura fights), this is probably an increase to your hps as well, but it really depends on the fights and your ability to keep 5 stacks rolling. For fights where you can't just stop the intense aoe healing required, stick with mindbender and take from darkness comes light for juggling the procs and heavy serendipity play.
For both specs pvp, the mana savings are there, and the spell is an instant, but it still isn't benefitting from pvp power on the ptr. And arguably, mindbender is better burst and from darkness comes light is a better instant for healing on the move (remember my post from way back saying you had to be casting ~40% of your casts as flash, greater, binding, smite, or heal to justify that talent). But your pet can be cc'd, and the healing from solace isn't affected by resilience, so most of the ptr arena testers are actually preferring solace. Ultimately, I think it depends on your playstyle.
Edited by Twistedmind on 2/23/13 8:19 AM (PST)
What needs to happen
Discipline is getting too much value from atonement. Not just with this talent, but in general. The opportunity cost to build stacks of evangelism falls by the wayside because it was largely hindered (once upon a time) by the time it took to cast (and time spent not doing bigger heals) but also because of its high mana cost. The simple fact of the matter is that the mana cost is negligible with entry level raid gear spirit levels, and the possible mana returns for disc keep going up with every point after the rapture softcap. Also, it isn't as punishing to build evangelism now because the cast time is much faster than before. When disc has so many fixed cd's, faster cast times are going to allow for more wiggle room to maximize the number of possible spells cast. This new revision to the talent will assist with getting further away from these penalties absurdly because its further removing any mana penalties. Another looming problem with how efficient these spells are compared to direct heals is that they're making archangel/atonement play mandatory. My definition of mandatory is "must be played this way in every situation". What would really make this a tough decision (when weighing against fdcl and mb) is eliminating the reduced mana cost from evangelism building spells. This is my opinion, but I still believe that's the cleanest way to both stop the goofy playstyle of dropping all spirit in favor of crit haste and spamming atonements at 5 stacks evangelism for entire fights, and also the cleanest way of putting more weight on casting something besides smite for every fight (maybe even cast heal, gheal, or fheal!).
For holy, it's a neat flavor. But at the end of the day, disc is still getting slightly more value from it because disc can consume evangelism stacks for archangel. If holy could do something with evangelism stacks, it would be a really compelling choice to take this talent.
They could always make it work better for holy by having it proc a regen buff in addition to the 1% mana back. Like an extra .5% mana back over 5 secs. Something small to offset the need to not heal while using it.
Your analysis is solid, but you're missing one important issue: Twist of Faith.
In most of the more difficult fights, you've got secondary targets. Secondary targets on which your Holy Fire creates an 17 sec window of Test of Faith (10 sec duration, procs from initial cast and ticks 7 sec out). On primary targets, Test of Faith also matches the tendency of fights to ramp up damage towards the end.
Now, Discipline loses a fair amount from Test of Faith (it does gain some back through Atonement double-dipping) but Holy gets the fully benefit. Casting a Holy Fire (less than 1.5 sec) to gain 17 sec worth of +15% casting is a pure hps win.
However, there is the issue of mana - it would almost certainly be a net loss in mana regeneration (over Mindbender) to use this approach. How much you value that loss against the gain hps is somewhat subjective.
Edited by Twistedmind on 1/8/13 12:13 PM (PST)
That's true, I did not account for the potential for twist of fate, which arguably would be something you wouldn't use as holy (offensively) unless you had to cast a solace for the mana.
The uptime he's^ getting at is that the buff refreshes itself with every dot tick, so a single holy fire could provide 17 seconds of twist of fate uptime if the target being hit is in execute range but won't die during the 7 seconds of the holy fire dot.
Also, arguably, the 'problem' of not keeping uptime with twist of fate doesn't exist for disc because the spec regularly would cast offensive spells, either with or without the new power word: solace.
Its a fair point to bring up, but there are still many conditions that must align to use twist of fate that would work against using the new solace for this purpose. You still must use the spell as close to cooldown, but if you have to delay the cooldown because something is almost in execute range you're bleeding out mana efficiency. Not to mention the increased awareness required to properly execute and actually make use of this compared to the ease of power infusion/divine insight.
I'm a fan of Twist of Fate as holy, it would just be better if we had a more flexible cast window with Solace. This just doesn't exist with the fixed 10 second cd. So my vote is more support towards reverting Solace to its previous version (the telluric currents version) for holy spec.
That's true, I did not account for the potential for twist of fate, which arguably would be something you wouldn't use as holy (offensively) unless you had to cast a solace for the mana.
Personally, I think reworking the talent to: "Solace: Whenever you deal Holy damage, you regain 0.2% base mana; whenever you deal Shadow damage, you have a 25% chance to refresh the duration of another periodic Shadow damage-over-time effects (the effect with the longest remaining time is chosen)." would be a nice fix.
For Discipline, this would tone done the free-Holy-Fire-wth-mana-return you discussed above. Depending on how much you used Atonement (Smite and Penance also would receive mild mana returns, but nothing like the 8 * 0.2% benefits for Holy Fire), it could beat Mindbender or not.
For Holy, it would make Holy Fire massively mana positive in Chastise Chakra. So Chastise could be legitimately used as a mana recovery method (albeit a bit clunky).
For Shadow, it would replace the rarely-ever-used Insanity with an effect I have no desire to fully model for balance. :) But it's an oft-requested ability (essentially your Vampiric Touch and Shadow Word: Pain would constantly refresh each other - you'd have virtually no chance of refreshing Devouring Plague due to the short duration).
As long as we're doing wishful thinking, I'd like to see the "costs no mana" part of this spell removed and for discipline atonement heals to only do 75% of the damage done in a smart heal (instead of the current 100%). I'd like to see the old Solace return for holy only and for it to do 1% mana back, and for it to completely replace smite and "deal the same damage and interact with other spells and abilities in the same manner" (read, stack evangelism). It would be cool if chakra was a 15 second cooldown. It would be nice if the red chakra only affected holy damage spells. If spirit of redemption returned you to 10% of your hp at the end of its duration, I'd finally have a reason to consider holy for more than just high hps fights with bad prayer of healing range issues. Ideally, Rapture should return a flat 6.1% of our mana; no more absurd scaling, and extra icing for returning the multiple rapture pop bug. It would be nice if holy nova was a baseline, discipline only spell, that did something different from what it currently does (perhaps lifebloom-ish but with an aoe? like, after 4-5 seconds it explodes healing everyone within 10 yards of the target. novas on multiple targets would subtract the amount of bloom healing). It would be really nice if holy word: serenity just gave a flat 25% increased heal buff on the target, no more crit nonsense. It would be nice if guardian spirit and void shift actually worked reliably. It would be cool if lightwell finally was able to be /use macro'd like warlock healthstones. Wishful thinking ;)