Mage Armor isn't terribly expensive. In my gear it's about 5% of my mana, but losing 5% of your mana at the beginning of a fight is a lot less troublesome than losing 5% of your mana when you're already low and your Shaman just hit Bloodlust. Through my experience the change is never justified since, as mentioned, it just makes you run out of mana more slowly. If you still manage to run out you're pooched - Mage Armor or not.

Mage Armor simply isn't a dependable reactive ability which is why it's unfair to compare it to Lifetap and AotV.
The closest we have is Evocation and, as mentioned about a million times, we can't control it without being penalized.

A Warlock can choose to Lifetap until a certain percentage of his mana is restored and continue DPSing. If he didn't Lifetap enough he'll just tap more. A Hunter can do the same with AotV. Furthermore, none of these classes are really penalized for regenerating mana at the "wrong" time - with the exception of a dumb Warlock who is comfortable with Lifetapping to 10% health in an AoE-heavy encounter.

If Blizzard's idea of "mana management" is Mages timing their Evocation between Razuvious' shouts then I don't understand why Warlocks and Hunters have much more dependable and forgiving mana-regen mechanics.

The difference here is that mage armor is not an active mana regen source. It is also hardly something you can use reactively. In other words, comparing aspect of the viper with mage armor is like comparing oranges to apples.

It's active in exactly the same manner as Aspect of the Viper, which has been dragged in to this. Now, I'll agree that you can't use it reactively, too. But it's aspect of the viper. It's just a terrible unbalanced and horribly planned AotV. But both are things we switch to over better DPS things to recover mana. It just happens that AotV works and Mage Armour is horrible. This doesn't mean that it shouldn't be compared, it just means that the comparison can't be "this provides the same benefits as AotV," it's "this is the same type of spell as AotV."

If it were beefed to 2% of the mage's mana pool per second (4x as strong as replenishment) would it still be useless? I doubt it. It would be much more reactive, in the same sense as AotV. It's just not giving adequate returns.

Aspect of the Viper really shouldn't be compared to Mage Armor. Aspect of the Viper is dependent on two variables. One is your damage output; the other is your maximum mana. Mage Armor, on the other hand, isn't dependent on your damage or your total mana. It's dependent on spirit (and to a lesser degree, intellect, which indirectly, in this case, is related to your total mana).

Because, currently, a Frostfire mage has no in-combat regen, and spirit is so godawful (26 spirit was giving me 5 in-combat Mp5 at 30% regen, for instance), there is no reason to have spirit on any of your gear if at all possible. But because this one fight runs longer than normal, and you may need to switch to mage armor towards the end, you're left with an abysmal in-combat regen. The problem is that you aren't regenning mana faster than you can expend it. In fact, you aren't doing that even if you were using Mage Armor from the start. That would trivialize mana regen.

The purpose of Mage Armor, of course, is to delay the time it takes for you to run out of mana. Using it when you are out of mana, or almost out of mana, is the worst possible time to use it. It doesn't give you an instant boost in mana like a mana gem or mana potion does (or even evocation). Instead, it costs more than another frostfire bolt would.

But Aspect of the Viper is entirely reactive. It requires no part in selection of gear to work because it is a percentage of your maximum mana (along with your damage) that is returning mana. It's not reliant on a stat you don't have in the first place. It also costs no mana on its own. Yes, you are doing 50% of normal damage, but think of it this way: at least you are able to still do damage. With Mage Armor, you are regenning at a constant rate, but still trying to cast at the same rate. You may as well just start wanding to get out of the 5-second rule, and keep molten armor on for the 5% crit (which has a detrimental affect MoE on top of all things).

The short version of the difference is that when a hunter needs to switch to Aspect of the Viper, they do so until they've recouped an appropriate amount of mana to continue the fight without needing to switch. And they can switch back at any time. When a mage needs to switch to Mage Armor, it's a calculation of how much longer they will be able to cast before they completely run out of mana. AotV regens mana so you can continue fighting. Mage armor delays the time it takes for you to run out of mana. That is why they should not be compared.

Ghostcrawler made several posts on the Priest forums, yet are also relevant to the topic at hand here for mages: World of Warcraft - English (NA) Forums -> ETA on Priest talent trees?

It's a continuation of the trend: post about mana to healers, never post in the context of DPS.

I would sleep better at night if I knew that there was an understanding up there that:

(1) dps mana and healer mana serve two different functions, and
(2) dps aren't structured to be able to use mana the healer way (we don't have "that one big nuke" we can save up for)

As it is, the track record of beta means those posts make me nervous. Every time they posted about that, they nerfed mana regen because it was overpowered from a healing perspective, but there were few (and much-delayed) changes made to DPS.

JoW was the sole "dps-only" mana regen source, which was what could set our regen apart from healers, and it just got gutted for mages, to the tune of -130mp5.

And the consolation prize we got was that a stat we avoid (because its value is often zero, and barely useful on some fights) will now give us minor regen through talents. So I can gear up to 450 spirit just to offset the JoW loss (not counting the -3% spell costs that used to be on the talent) if I want to be mana-neutral, or I can just take the loss -- but either way it makes me wonder how more mana-hungry builds are supposed to compete.

You're seeing the forest for the trees. There's no reason that mage armor couldn't be redone like aspect of the viper. Aspect of the Viper doesn't really depend on damage output. It regens a set amount upon successful attack, irrespective of the damage done or type of attack. In effect, the regen is approximately constant for when you are attacking, and around half as much when you are not attacking, changing a little bit with haste. So, basically, the core difference is that AotV scales with maximum mana instead of spirit. Is that really an insurmountable difference? I think not. Of course, this doesn't address that Blizzard wants mages to get some use out of spirit. But, there's no reason that your passive regen has to be the place for it (cf. warlocks, spirit affects only our active regen).

In my opinion you're being abit pessimistic. I instantly saw this as a good and welcome change. The pyromaniac change + the JoW nerf, allmost cancel eachother out, even with base spirit, meaning we're less dependant on other classes to get the same regen. I'm guessing hardly anyone had any mana problems in 25 mans with replenishment etc. But in 10 mans where you couldnt depend as much on other classes, we could have a much harder time. (10 man KT for example).

About the new viable full fire build we'll see how mana hungry it is, but I'm amazed they've actually given us 3 very good and viable raid specs. It's the first time since WoW released!

Realistically, I can't see Mage Armour being changed to be like AotV. Mage Armour is currently (and, in my opinion, intentionally) a long-term choice. You choose Mage Armour if you think you're going to OOM on a current fight. As has been stated, the problem is is that there are no fights long enough to warrant it, and even if there were, Mage Armour is the not powerful enough to warrant the DPS decrease. It's less MP5 on almost all fights than a Mana Potion. It's way, way less MP5 than Mana Gems.

For Mage Armour to be a good offensive (DPS-increasing) choice, it needs to mean that you get to cast long enough to offset the 5% crit damage bonus from Molten Armour. Doing some really rough napkin-math, assuming solely FFB spam, you would need to cast a full 6% more Frostfire Bolts to reach the same damage, at a cost of more than 5% dps. With MOE, Hotstreak, and a full-up FFB rotation thrown in, this changes even more.

Spirit is not useful. For it to be so, either it needs to do something more than passively regenerate mana or the whole mana/regen/replenishment balance needs to be drastically changed to something where if you don't use spirit, you will go OOM on a 2 or 3 minute fight. Say, perhaps, increase all spell costs by 200% and increase spirit regen by 500%...but this would also require rebalancing most classes damage and mana costs, so I don't think it's at all likely.

Think Spirit is practically useless, what about Arcane Intellect/Arcane Brilliance? All it gives is 900 mana at highest rank.
It just seems too weak being only worth 900 mana when your mana pool could be like 15K and higher.

I think something is off when a tank's health can be 30K and mana is around 20K.

Yes, arcane intellect is weak. Is there a point to that statement?

Exactly. Unfortunately, at this time, Mana Regen is not a solid, tangible use for spirit. I like your idea of a spirit-based, high regeneration but with cool-down mage armour, but I don't think it's something that blizzard is considering. Armours, for both mage and warlock, are long-duration passive buffs. For spirit to be useful with armour, it needs to affect all of the armours in a meaningful way for boss fights. It could also be an interesting way to balance specs.

Warlocks get a meaningful use out of spirit with Fel Armour. What if Mage Armour had a passive haste buff tied to spirit somehow? What if Molten armour had it's crit chance and reactive damage tied to spirit? What if Ice Armour had it's armour amount and oh, say, chill time affected by sprit?

Making it so that spirit is a desirable stat solely in terms of mana regen means changing the whole mana usage/regen/replenishment/five second rule so much that I can't see it being plausible for Blizzard to do. Especially because they will also have to re-balance all of the mana-using classes around it.

It would have to consume no gcd too.

Those are very important points. Combined, they mean that the entire "5-second rule" mechanic is useless for mages.

A healer can plausibly cast a heal, wait a while, and cast another big heal. The demands on a healer are are governed by the damage that the raid takes, and in a mythical world of "infinite haste, 0 cast time, 0 gcd," a healer would still cast, then wait, then cast again. If the individual heals were large enough compared to damage taken, then an infinite-haste healer could spend a fair fraction of time out of the 5 second rule.

In contrast, there's absolutely no benefit for dps to pause like that. An infinitely-hasted dps (mage) would cast until out of mana, gem, evocate, potion, repeat, wand, repeat. The 5-second-rule mechanic is useless, because there is no sane rotation that gives a mage time "not casting," except when the fight mechanics force breaks. Even without haste, the only way a mage could spend time outside of the 5sr is if she were to rotate channeled and un-channeled spells. (Normal Arcane Missiles, rotated with non-Hot-Streak Pyroblast would do it, for example. It would also be nuts.)

Blizzard-chaning also spends some time "not casting," but there the mana drain is so large that any sensible regen won't even prolong the damage-time.

So the problem with mage mana regeneration is that there isn't any. Mana gems, potions, and Evocation are more like "spare pools" of mana, rather than ways of regaining it when necessary. Once those are all tapped mages simply have nothing to turn to. The problem is only hidden by the gigantic mana pools available to current mages.

Finally, a quick idea. Blizzard seems fond of haste, so what about a "Aspect of the Viper" for Mages (replacing current Evocation) that slows cast time by 100%, rather than reduces damage done? The PvP utility is negligible if it also increases cooldowns on instant-casts by the same fraction.

I like the idea of reactive switching armors to adapt during a fight. Heck I´d settle for just making both armors into something of a choise boss by boss. Biggest problem to this design as I see it is currently the glyphs. Glyphs aren´t designed to be switched battle by battle so giving mages two major armor glyphs directly contradicts using mage and molten armor anything like AotV. Neither fire or frostfire has the glyph slots to spare and arcane only has them to spare for lack of options in the glyph department. Offcourse if the armorglyphs had only a very small impact on mage and molten armor instead of their current 66% boost you could design these spells more like AotV but major glyphs only having minor impact on their spell doesnt work as a concept either.

Long story short I´d really enjoy an AotV like mana management system for mages, but current glyph design directly contradicts anything like this.

Actually the problem with applying the concept ot AotV to mage armours is that the armours regenerate too little mana and give you too little dps in comparison. I mean AotV reduces damage done by 50%, molten armour is how much dps?

They'd have to make mage dps revolve around having molten armour and give mage armour a heft dps penalty which isn't clever, mage armour is widely used in PvP.

It just won't happen, this is why I think an evocation revamp would seem more natural. Designing it more like life tap then as in lose time for mana, maybe with a short term cooldown to add some more judgement to it.

There is a difference between Arcane using the 30% mana regen and Fire with the 30% mana regen. In the Fire tree, there is Master of Elements which is a mana cost reduction of 30% for spells that crit, 25% if you have Burnout for fire spells.

So if your crit rate is 30% and you get 30% mana back, your mana pool is effectively 9% larger. Based on my lvl 79 mage, without mana regen abilities, I would go OOM in 19 to 20 casts of fireball. With 18 Arc/52 Fire spec, I would have 27 fireballs cast before using Evocation or a mana gem.

And then I base the numbers if Pyromaniac was 3% mana cost reduction, and I would have had 26.6 fireball casts. Gee, that change did a lot of good. Maybe if I had better gear...

And my point with Arcane Intellect and Arcane Brilliance being so weak, Blizzard could do something with those to match what priests get out of PW:Fort and the talented Spirit buff. Maybe tag on to Magic Attunement a boost to AI and Arcane Brilliance.

I think Evo is the perfect candidate to fix some of our problems. And without much of an overhaul at all. Reduce the amount of mana it regens and remove the cooldown completely. Say, 30% over 5 seconds, no cooldown. Still channeled, still interrupted by dmg. You are still doing 0 dps while regenning mana, but you can use it whenever you need, 2 ticks here, 3 ticks there. Makes it far more interactive and far less frustrating.

ETA: I wouldn't tie it to spirit at all either. Just a flat percentage based increase.

So the answer to mage regeneration is... to make mana pools even larger, so regeneration is less important?

Really, the question we should be answering is this:
"I'm glad that mages have X, because even if my mana pool were half the size it is now, I'd still be a valuable contribution to a fight."

Right now, there's no good answer. This is bad. It makes mages valuable only while they have mana, and it also means that Blizzard is restricted in encounter design. To keep mages functional, they have to handle mana drains in encounters with kid gloves. It isn't quite so bad with healers because they can crank down damage done if they lower healer mana, but mages have to stay order-of-magnitude-comparable with other dps classes, come hell or mana drain.

Armageddon Raid Scenario: All mana, enery, rage, and runic pools are drained by 50% of damage done (for mana, scaling appropriately for energy/rage).

epoh: Any mana regenration model that doesn't include an increased use for spirit only solves half the mage's problems, namely the part where we have a useless stat on our gear. Also, your evocation has indirect effects on mana drains, mana usage, and non-evocation forms of mana regeneration in pvp. The goal is to have us in a place where managing mana is important, going oom is not "*sigh* wand time" (unless you tie wanding into the regen model), and spirit has a use.

Your statements are inaccurate. Firstly, MoE gives mana back based on the spells base cost, not the cost you pay for which maybe modified by current Pyromaniac and current Elemental Precision. In essence, you're paying for (as low as) 94% of the cost and refunding 30% (or 25%) of the cost. Not refunding 30/25% of the 94%.

Secondly, your crit rate would not be 30%. Assume a moderate crit rate of about 520, giving 27.5% base Fire crit with AI. Add 5% for Molten Armor, 10% for i-Scorch, 3% for ToW (or equivalent). Then add 2% for FFB specs glyph or 8% for Fireball glyph + Focus Magic.

By my account that's 47.5% FFb crit, refunding 26.6% of the spell's cost (12.7% economy) or 53.5% FB crit, refunding 25.77% of the spell's cost (13.8% economy). Significantly better than your 0.6-per-20 Fireball. Closer to 2.8-per-20.

Don't forget spell haste and mana regen rate. Spell Haste reduces the amount of mana you regenerate before you use up the amount of mana that matches your mana pool. And with mana regen, any spirit on a mage is because the gear had it.

My lvl 79, pre-raid, mage has 181mp5 OOC mana regen and 12K mana pool with 9.38% haste and 18.86 fire crit chance without Molten Armor (+3% crit chance). The numbers will be lower.

Also, I do not assume for perfect raid conditions because not every mage wants to take the time and gold to respec and reglyph as well as that boomkin and elemental shaman might not be there for a week.