I just forgot that Prayer of Healing is something Discipline priests cast too. You are correct that Prayer will beat Flash Heal by a wide margin here.

I think the core of this dispute is whether or not we as priests are running out of mana in Ulduar. I contend that mana hasn't been an issue for me on any fight so far, but you've had problems with many fights. How many healers do you bring to most fights? We generally bring 6 healers, or 7 if the fight is more about control than DPS. I suspect that we are bringing too many though, which would explain the lack of mana issues. If you are bringing only 5 healers to an average fight, then I can see mana being a concern. If you have 6 or 7 healers, then someone in your raid is doing something wrong if healers are out of mana.

On the subject of hard modes, most of them seem to be of the form "stack extra DPS and cut healers", so that's a situation where you might need more mana regen. However, they also have the form "this deals more damage", so you in theory need more throughput. Mana efficiency and throughput are always at odds, and I believe hard modes need to be analyzed on a case-by-case basis. We can't just say that "all hard modes require a greater focus on gear aspect X".

I'd like to take a step back here. Spell power is just a means to an end, just like spirit, crit, haste, and m/5 are. There is always a tension between how much throughput you have and how much longevity (regen) you have. Shorter fights tend to favor throughput while longer fights favor longevity. From that framework, I suggest Spell Power is the most efficient stat because it increases both Throughput and Longevity. If you could spend 2 itemization points on any combination of spell power, int (mostly all regen), and haste (all throughput), 2 points in spell power (roughly 2.1 spell power) will beat 1 haste and 1 int for both throughput and longevity. That's not to say there aren't fights where 2 int will be better (vezax) or 2 haste will be better (razorscale), but 1 haste + 1 int is never the right choice because 2 spell power beats it on both throughput and efficiency.

So when I hear you arguing for both haste because it increases Throughput and Int/Spirit because they increase Longevity, I'm dubious. You can't argue both sides. Either Throughput matters more, meaning you focus on spell power / crit / haste, or Longevity matters more, meaning you focus on spell power / crit / int+spirit. I wonder if the reason you are running low on mana is that you haven't focused on spell power. It's easy to fall into a trap where you get a lot of haste, which makes you feel low on mana, and then you try to compensate by stacking mana regen. Unfortunately this produces a situation where the heals are anemic, and you need to work a lot harder to keep everyone alive.

Related to this, it's worth noting that we as priest might need to start collecting two healing sets, one focused on mana efficiency and another on throughput. I think part of this dispute comes from applying a "one size fits all" gearing methodology to the variety of boss fights in the game. Neither a "hit these arbitrary thresholds" nor a "follow these arbitrary weightings" heuristic can do justice.


But we don't make decisions by itemization points, because we know that things with similar item levels can have vastly different utility. Even if you wanted to argue from itemization points, 65 int < 38 m/5, so if we followed that logic, we would falsely conclude Distilled Wisdom is worse than Pure Mojo.

At any rate, this is another aspect of throughput versus longevity. But this one clearly favors Frost Wyrm, because taking Distilled wisdom makes you lose 2 spell power for every 1 intellect you gain. If you want more longevity, it's better to make that sacrifice in gemming, where you only lose 1.15 spell power for every 1 intellect you gain.

But why is gemming the perfect situation to balance stats? Every other class in the game uses gemming as the opportunity to stack whichever stats give them the best returns. I know that there is a theoretical mathematical optimum where all stats are roughly equal in value. And yes, clearly the more spell power you have, the better returns you get from crit and haste. But I haven't seen any math proving that the current gear level just happens to be in this optimum. Rather, the math I've seen says quite the opposite-- some stats really do provide better returns than others, and they should be stacked.

Similarly, I confess I'm uneasy about putting the "21% haste soft cap" in the root post as a "really important guideline" without any math to back it up. I know you are sticking with it, but I prefer my choices to have a mathematical backing. If there's math to back it up, great-- we should post that. If not, then it seems like a harmful guideline that just encourages inefficient gearing (from a mana standpoint).

It depends on more than just your role...

Item 1 - Depends on the discipline of your raid healers to not heal just to heal. On Hodir I have been getting ~40k in ticks from PoH glyph while I am tank healing.

Item 2 - While I agree that you shouldn't be distracted or on gcd so much that your primary target could die. That doesn't mean you can't PoH since you can PW:S the tank then get a ~1.9 second PoH. Then you can PoH the group the tank is in that is 10 targets and you didn't really leave the tank that long.

Item 3 - Are you sure that it is really playing that big of a role? I have been watching for it and haven't seen it multi-proc that often but then I haven't been trying to exploit it either.

Well, unless you're gemming for stat-balancing with Haste and Crit then the only reason not to use Spell Power gems is if you have Mana problems. If all your sockets are gemmed for Regen and you still have Mana problems, thus requiring Distilled Wisdom, then you probably have bigger problems than the difference between the two flasks. It would appear that your raid takes too much damage, or that your healing isn't efficient enough, or perhaps that you have far too few healers, or maybe your gear is below par.

On the whole Flask discussion as a whole
If your raid is less disciplined, sniping HoTs, not avoiding damage quickly, forgetting Health-stones, ignoring Lightwell, using Immunities solely for threat issues and so on and so forth, then you will need more mana. Some of these are especially true during new encounters and will not be an issue in a few weeks. This whole discussion seems to boil down to the quality of the raid. If people make few mistakes and the dps is high, keeping fights short, then you don't need as much Mana. There's not an aweful lot the healers can do here. Their job is to keep people alive. If people make mistakes their job becomes harder and requires more mana. Particularly for new encounters having spare mana to cover for people's mistakes is a big asset to a raid and I would only reduce my available mana when I'm confident that I won't need the excess.