I don't get the regen nerfs...

Out of casting regen can be a significant source of mana, but this is complicated by the fact that it can also be really random due to clearcast procs and the nature of the fight itself. Something like Loatheb can be easier than intended because the fight asks you to pause a lot.

Now I remember the long ago healing strat guides and how clever they can make it sound to try and delay a heal to earn more mana. It sounds cool on paper. But the reality was things like stopcasting macros because you couldn't afford to wait a couple of seconds to see if the heal was needed.

We actually want for healing to be less spammy and more about decisions and coordination. But if we did that sans other changes, everyone would be out of fsr even more and the content would be too easy. Make sense? Making regen more consistent is also the key to making healing more tactical and less monotonous and frenetic.

Spirit is a great regen stat depending on your class. Many players never get out of the fsr until they overgear the content at which point mana really isn't what is slowing you down. I predict many players won't notice a difference especially on shorter fights like Arenas or trash.

I chose "monotonous" and "frenetic" specifically because they both imply that you are spending every GCD on a heal (and probably a big one because why not - you can both afford it and have the need to make it count).

What we would like is for the chance to pause, perhaps to consider the fight, coordinate or just prevent overhealing, to exist. Without these changes, those pauses will just fill your bar back up.

We think we know exactly how long you can non-stop heal without going oom. If you discount any procs or cooldowns that let you get a little OOFSR mana, it isn't as long as some people imply (assuming you don't overgear the content or have Divine Plea).

Couple of points...

1) I was trying to argue (perhaps too politely) that the notion that healing was fun when smart people could take healing breaks to allow for out of five-second-rule regen was a little bit of rosy nostalgia. The way expert raiders handled that was by healer rotations or stopcasting macros, and I don't seriously think a lot of people long for the return of those. (No doubt some do.) Sunwell and similar content were tuned to where you couldn't realistically stop casting for regen anyway. If you are stopping now, it's because Naxxramas is easy. On challenging content, all 4 classes already spammed heals pretty consistently. That meant the out of FSR breaks came mostly from clearcasting procs or encounter mechanics that forced you to pause. When that happened, mana felt limitless. As I've said, we would rather shift back a little (A LITTLE) from GCD-constrained healing to mana-constrained healing.

2) It sounds like some of you are saying that even on challenging content without replenishment and without leaving the FSR your mana is infinite. I am a slightly skeptical of such claims, but we'll see what happens in Ulduar. We think you will run out of mana sometimes, particular on very hard fights or if you aren't playing at your best, even with replenishment. Similarly, we feel like the paladins are noticing the Divine Plea change in PvE. If you aren't, then perhaps you are an amazing healer, or perhaps Naxx isn't challenging you much. Now, if it turns out regen is still just too good, then replenishment would definitely be our next target. But we don't think we're far off. This isn't a "to the ground" scenario.

3) This change won't change the relative values of Int and Spirit. Int was already a good regen stat (in part because it provides other bonuses). However, you are rarely ever choosing between a +100 Int chest and a +100 Spirit chest (though you might be with gems). More to the point though, the regen formula is SQRT of Int * Spirit * a constant. We are just lowering the constant. If Spirit was good for you before, it will still be good for you. If the only reason it was good for you before was because you took a lot of out of FSR breaks, well 1) again I am skeptical, and 2) yes, under that specific situation you will notice a nerf to your regen.

Overall, our goal is not to make healing more difficult across the board (though we do want mana to matter more). Our goal is to make healing more interesting.

Q u o t e:
It matters a fair amount for Druids because of how much the Spirit formula worked for Innervate.

I don't think it will suddenly make druids regem from Spirit to Int. We'll see.

If you mean these changes will nerf Innervate, yes, it will. On live Innervate can be 120% of your bar, say 25,000 mana. Historically it was more like 75% of your bar, and it should be closer to that with these changes.

Q u o t e:
And nobody's stood around taking healing rotation breaks since early MC, and no one here is arguing for a return to that.

I agree with that first part, Odes, but I think some people are arguing that the rotation breaks are what OOFSR healing was all about. We aren’t really pushing for a return to that.

Q u o t e:
Holy Priests are being forced into the kind of mindless boring spam that's long characterized Paladin healing. A play style they've been complaining about for years. Why would you want to encourage that? It's not fun, and to force it on a class/specc that was once the complete opposite of that is incredibly disappointing.

We don’t want that either. Most of my posts have been trying to say “We don’t want the healing game to be a spammy one.” (Again, some players are arguing they do want that, which is their prerogative). However, we are also backing off of the thing that prevents the spams to be the FSR game. I think a better solution is interesting spells. Swiftmend is a really interesting spell. So are Lifebloom, Penance and Beacon of Light. Prayer of Mending is a very effective spell, but I don’t think it’s a particularly fun one to cast. I do think Holy priests could use some work in the fun department, which has been a pretty consistent thread for the last several weeks on these forums, and something we are working on.

I also think some paladins do like the way paladin healing plays out. I’m not saying we want to adopt that model in every case, but we shouldn’t dismiss it either. (Similarly, I am not dismissing players who thought the FSR rule was fun, and in fact we aren’t removing it, just reducing its impact).

Q u o t e:
The reason these changes have to happen are quite obvious. If they're going to move to less-spammy fights, druids and priests with spirit will dominate paladins and shaman without. As soon as you move to a situation where you actually CAN pause for more than a couple of seconds priest and druids regen spikes, where paladins and shamans stay static.

This doesn't necessarily make druids and priests overpowered, but it makes it unbalanced. If they balance it around paladin/shaman regen, either druids/priests have to keep casting constantly, or else they get to make use of the 5sr. How frequently they get to make use of the 5sr regen is dependant on:

How frequently the target avoids hits/resists damage.
How frequently they get clearcasting procs.
Who else is healing, and whether they land a heal.
Whether they have cooldowns up like inner focus.

When you factor those unpredictable effects in, you can't make a strong model of priest/druid regen over the course of a fight, so you can't design a fight that will make a class go OOM if they play poorly, or if they undergear the content or whatever. You can only make a fight that challenges healers to not screw up.

Paladins and Shaman have pretty much static mana regen. They have some cooldowns to get mana back, but nothing as variable as the 5 second rule.

If you base your fights around paladins and shamans going OOM, you need to base your fight around priests and druids going OOM at about the same rate. And then when you take into account a streak of tank avoidance, letting you cancel heals, clearcasting and inner fire letting the priest hold off on spending any mana for 15 seconds, that lucky streak translates into a boatload of mana regen that the paladins and shamans have no way of lucking into. In this respect, it makes fights on average too easy for those classes with higher in combat mana regen.

If you balance it around the average mana consumption, then when you have a bad streak where you can't cancel more than every other cast, and you don't get clearcasting at opportune times, you're going to run short on mana early.

The problem is simply, the 5sr adds too much variance into your regen that non-spirit classes don't deal with. Blizzard wants to make fights that challenge your mana pool, but they have to be able to model what your mana pool should look like over the course of the encounter. When luck based factors can greatly affect that, there's no way to model it.

Thanks, Thorene. Well said.

Q u o t e:
I guess at the end of the day, I'm not necessarily concerned about the o5sr nerf - a lot of druids think we shouldn't be doing the o5sr dance as it is, and obviously it's been stated now that the devs would like there to be less focus on this method of healing/mana regen as well. I guess my concerns lie in the current mana regen model in general, and the fact that it benefits two healing classes while penalizing the other two.

The goal is to have mana last about the same for all healers. We don't think many players would be that interested in a style where you heal crazy good for a short period and then run OOM. How classes manage their mana varies a lot, and we are making tweaks to it for 3.1 to try and keep them in line. For example, the shadow fiend needs to be more reliable.

I am still not following the logic that Int now trumps Spirit by even more. It may have been a better stat already, but these changes shouldn't affect it that much unless you A) skip Meditation and Intensity, or B) relied a whole lot more on OOFSR regen than the average raid healer of reasonable skill.

I will break from our normal practice and go ahead and provide the numbers, just to make sure nobody is guessing about the details when doing their estimates:

-- The amount of base mana regen granted has been reduced 40%. We called this "Spirit" in the notes, since most players associate OOFSR regen with Spirit, but in reality Int factors into the equation as well and we only lowered the constant, not the relative contributions of Int or Spirit. In retrospect, this probably caused more confusion than it alleviated, but mana regeneration is a fairly technical concept.

-- The effects of talents that provide mana regen while casting have been increased by 67%. This includes: Arcane Meditation, Improved Spirit Tap, Intensity, Mage Armor, Meditation, Pyromaniac, and Spirit Tap. For example, Intensity and Meditation are now 17/33/50% mana regen while casting (up from 10/20/30%). For most dps classes who never got much mana from OOFSR in the first place, the results should not be noticeable. Boomkin may be a possible exception because of Innervate, and we'll take a look at that.

-- This should leave mana regeneration while casting (even the contribution of Spirit) relatively unchanged, but reduce mana regeneration while not casting by 40%. (If you don't understand our logic, I suggest you re-read the excellent post I quoted above.)

-- Since paladins don't rely on any of those abilities for mana regeneration, we lowered the healing penalty of Divine Plea to -50%. We are also likely to make Spiritual Attunement provide less mana for non-tanking paladins. We are not touching Illumination for the moment. Nor are we lowering the effects of Replenishment (though as I have suggested, it would be our likely next target if we aren't happy with the results of these changes).

Q u o t e:
Yes GC, you seem to be completely ignoring the PVP implications of this. You're leaving the mana regen of paladins and shamans the same (the two strongest pvp healers) and nerfing the regen of priests and druids (the two weakest). The implications of this are huge.

I said the goal was to have regen about the same. In our tests all four healers run out of mana at about the same time if they aren't able to get out of the five-second rule, and if the paladins aren't gaining mana from being healed and using Divine Plea at every cooldown.

We think Holy paladins are dominating healing in Arenas too much, but some of these changes should help that. (Plus, paladin defenses just tend to help them survive when burst is high. This is a trend that happened even in previous seasons.) We don't think the success of Resto shamans in Arena is necessarily a mana sustainability issue, but has a lot more to do with what other classes can do to priests and druids. (Which is also not an ideal situation.)

Q u o t e:
I have a question. You talk about priests, druids and paladins quite a bit. Unless I missed something, you have yet to talk about Shaman regen specifically.

I thought I mentioned this at one point, but we were pretty happy with Resto shaman mana regen. They were our target and we wanted to bring priest and druid regen (by making non-casting regen lower) and paladin regen (through DP and SA) closer to that point.

Q u o t e:
Can you clarify what you mean by "these changes"? I assume you're not saying that nerfing spirit-based OFSR regen (or SA) will hurt Paladins in PvP more than priests/druids. Are you talking about unannounced changes or something I have missed?

Sorry, that was unclear. I meant some of the changes to lower burst damage so that defensive classes aren’t so uber in PvP.

To those players who feel like they already have trouble managing mana now, it's hard for me to evaluate each case to know what is going on. My only suggestion is you need to treat mana regeneration as a real mechanic and not just go for throughput in all of your gear and talent choices. We are going to get all of this on the PTR so that we and the community can run additional tests to make sure that all four healing classes feel valuable and effective. Ulduar is definitely going to shine a harsher light on PvE balance in all forms and we're just going to have to deal with problems (real or imagined) as they come up.