Blizzard could have added some intrigue to talent selection if they offered a viable place for those other 2 points, but based on my presumed raid spec, if I don't take Nether Vortex, my choices are Imp. Counterspell, Imp. Blink, and Prismatic Cloak, none of which are really exciting. Perhaps Cloak offers a more tangible, if almost negligible, benefit, by reducing damage taken, but 4% damage reduction is not game-changing for PvE.

From the few level 85 tests I have ran (using the level 80 dummies to simulate being hit-capped), You can sustain around 6.5-7.0K with pre-heroic gear, with various rotations; at one point, I was alternating ABx2 and ABx3 cycles during the Evo cooldown, extending AB on Clearcasts, following a cooldown/proc burn to 40%. My gut tells me that at higher gear levels, ABx3 and Abx4 will be viable, and eventually we'll get back to ABx4, like we do on live, as our mana pools start to push 80-90K, and higher.

Personally, I think the idea of having a talent to modify mirror image would be cool - making it so mirror images use arcane blast instead of frostbolt, or something where using mirror image is more appealing to an arcane mage.

There is a minor glyph that does exactly this. Not that 25% damage to your Mirror Image nukes will make much difference, but it's better than nothing.

Does the glyph actually make your Mirror Images actually benefit from your school specialization? Last I heard, it was just a cosmetic change, but then again Fire Orb recently started benefiting from specializations, so I suppose MI could have changed as well.


As for Glyph of Conjuring, it's awesome that it makes it affordable to reconjure gems during combat, and is without a doubt a necessary glyph during any fight where you'd need more than three charges.

It's nice that a 2 stack/clear looks like it will be mana-positive, though a lot of our efficiency still relies on clearcasting and missiles procs. Though by now that could probably be fixed with tuning.

I have been trying to validate this statement, but have been unable to.

Query: Can AB2ABr become a mana positive rotation? If so, at what mana levels?
Setup: Use Mage Armor (unglyphed), Replenishment, Blessing of wisdom, base spirit regen, 100,000 mana pool, assume 0 haste (best case scenario)

Analysis:
MP5 sources:
Base spirit regen: 870
BoW: 326
Replenishment: 500 (1000 over MP10)
Mage Armor: 3000

Total MP5 gain: 4696
Total MPS gain: 939.2

MP5 drains:
AB0: 1393
AB1: 3831
ABr: 2090

Total mana drain: 7314 drained over 6.5 seconds
Total MPS drain: 1125.231 mana

Δ Mana = Mana Gain - Mana Drain = 939.2 - 1125.231 = -186.031
Δ Mana(with MA glyph) = 1059.2 - 1125.231 = -66.031

Conclusion: At 100k mana, AB2ABr is still not a mana positive rotation.



Further analysis:
Query: At what mana pool size does AB2ABr become a mana neutral rotation (assuming 0 haste)?

870 + 326 + (0.005*X) + (0.03*X) = 1125.231 * 5

0.035*X = 4430.115
X = 126,575.857 mana

with MA glyph:
X = 4430.115/0.041
X = 108,051.58 mana

So at ~126k mana (~108k with MA glyph), AB2ABr will become mana neutral. I don't know if we will reach that mana level at entry level raid gear, or even tier 1. Furthermore, as tier gear increases, so will haste, which will obviously negatively impact this analysis.


In short, I do not think it is feasible to assume that we will have access to a reliable mana positive rotation, even at higher gear levels.

I think the original idea still stands. The "mana positiveness" of a particular rotation is governed solely by RNG. If clearcasting procs and/or if AM procs where you can use it at an AB1 stack, are pretty much the only ways we can achieve mana positive rotations.

Arcane mages cannot rely on having an on demand mana positive rotation.

edit: Added Mage Armor glyph results.
N.B. I am still trying to ascertain the viability of using the MA glyph over ABr glyph, so if it turns out that MA glyph is optimal in most situations, things may change, but that still does not account for haste.

I wasn't attempting to argue there was no RNG involved. We can't just ignore RNG when discussing mana efficiency. Free Missile Barrage procs, Clearcasts and 30% mana return from Master of Elements crits play a big role. It's not ideal that we have minimal control over this but we can't just rule them out the calculations.

In reality in a raid environment (and barring a string of bad luck, which yes we're poorly quipped to deal with), if an Arcane Mage uses Missile Barrage as soon as it procs and Arcane Barrage after every one-stack Blast if it is off cooldown, the Arcane Mage in question is going to hover damned close to 100% mana until the next Evocation / Mana Gem.

We have a 40% chance to proc Missile Barrage, a 10% chance to proc Clearcasting and in a raid probably around a 30% chance to proc Master Master of Elements. Yes they're RNG but over the course of an average fight you'd have to be exceptionally unlucky to not receive some benefit. We can't just count them out the sims.

----------------------------

I'm not arguing that Arcane doesn't need some additional mana tool that can actually be controlled but I don't think we should be assessing mana efficiency without including the mana efficiency talents and abilities.

As for a counter to bad RNG, perhaps directly tie a mana efficiency talent to bad luck, ie if your Missile Barrage talent does not proc within 3 casts you gain X buff, which returns mana equal to X% of damage done by you for X seconds.

I know. I wasn't directly arguing against a particular point you made. I was more just using your post as a launchpad to explore and get some numbers on what kind of mana pool sizes will be needed to make our lowest reliable rotation mana neutral.

As far as reality goes, yes, with fair luck, we will be able to hover at a particular mana % point for a certain amount of time.

As a side point, I'm not certain if MoE will end up being as useful as we want it to be, at least not in all scenarios. It only returns mana for an ABr crits (~627 mana), for AB it returns mana based off of the mana cost of an unstacked AB cast (~418 mana) irrespective of which AB stack you were at when you crit, and doesn't return mana for AM at all (as expected).

If I were to hazard a guess, it may end up emphasizing crit for conservative rotations (where we may be using ABr more), but this would be like walking a tight rope, since as gear scales, so will our ability to cast higher stacks of AB during "conserve phase" which naturally means we will be using AM more often (since it will be available more often). Meaning that MoE's actual usefulness (in terms of influencing DPMs to a significant degree) may only be in that very short period of time where people are at or just below entry level gear, where they heavily rely on AB2s to conserve.

Though I guess since a little mana is better then no mana, coupled with the fact that there really is no other viable alternative talent to spend points into, Arcane mages will just end up picking MoE anyway.

As far as counting these procs out in our sims is concerned. I am certainly not doing so. In fact, the model I am using for them in my sims is as close to the real thing as you can get (i.e. I am not just mathing out their benefit). I haven't done MoE yet though, since again, I am not sure how useful it will end up being in the long run or whether it will end up having any meaningful impact over the course of a fight.

edit:
And then there is also Arcane Potency. In some ways, I see the synergy between Arcane Potency and MoE. My hope is that MoE is useful enough for us to try to use it for conserve phases, so as to allow us to syncronize its use with ArcPot. This may lead to sub strategies evolving where we actually use POM in conserve rotations to force ArcPot, hence making use of MoE to regain more mana. Though all of this is contingent on whether MoE is all that useful in the first place. Though even if it is, all its giving us is a chance for a chance at regaining mana (chance for ArcPot to proc normally then chance to crit, combined, to give us some mana back), which is yet more RNG when dealing with mana gains.

Either way, I guess we have some work to do in figuring out how useful MoE actually is.

While I won't argue that hard math may make MoE a lackluster choice due to the low amount of base mana our 1st AB costs, and the theoretical lack of quantity of Abarr's used, the process of elimination makes it almost mandatory.

Assuming you took only DPS talents, and put 1 point in Imp. Blink (which may have a DPS gain, depending on the particular fight), then Burning Soul & Piercing Ice, you'd have 4 points to place amongst Imp. CS, Imp. Blink, Prismatic Cloak, Nether Vortex, MoE, Imp. Fireblast, Early Frost, and Shatter. Of these, few have potential DPS gain:

Imp. CS: none
2/2 Imp blink: marginal, and highly situational
Prismatic Cloak: in theory, the removal of the fade aspect might allow faster DPS.
Nether Vortex: no increase except when attacking non-tanked targets, which in raid environments should not be happening. Further muddling that scenario would be the case where you switch targets before the first is dead, in which case the auto-slow does not apply, since it is still on the previous target.
MoE: Mana regen through crits
Imp. Fireblast: limited DPS gain; requires the situational need to cast an instant spell when Abarr is on CD
Early Frost: none, as you should never be casting Frostbolt
Shatter: assuming you'd rather not just start with AE, one might start an AoE phase with Frost Nova, than the first wave of AE should be almost a guaranteed crit, not sure if the trade off of the low damage on Nova makes this valid or not. I'd guess not, especially once our gear starts pushing us up into the 40% and higher crit range.

Logix, would you like to start a new Arcane thread? Something that can be fleshed into what the Fire/Frost ones are morphing into: more dedicated spec, talent discussion, optimal ways to play, cycles etc.

This thread has a lot of great discussion, with some about the aforementioned. Much of it has got lost in other discussion, and the OP is now outdated and dwelling on desgn. It's time to accept Arcane in Cataclysm, for what it's worth, open up consolidated discussion on that - and move forward with that in mind.

The thread can stay on with a role as a place to discuss problems and pros/cons for Arcane from design and thematic standpoints. Things that really would just clutter up the new thread. Furthermore, I myself am dissappointed with Arcane's state (disinterested as a result) and don't really have the heart to update this thread too closely moving forward to Cataclysm.

This thread will be renamed, and Logix's new incoming thread will take over the name "Arcane in Cataclysm" - and become the place to discuss the current state of Arcane from a ingame, optimal playing perspective. It won't dwell so much on the thematic problems and issues raised here. This thread can remain on as a place to discuss things like that, however.

Yea, I have started putting together the information for a thread like the one you describe. It is incoming shortly.

Since Arcane is now very mana dependent I assume we shouldn't just blindly stack sp+haste anymore but stick to a certain amount of haste which will make you able to conserve a certain amount of mana after a burn phase. I haven't gotten around to test Arcane PvE yet but I was wondering how we should be dealing with Blood Lust / Heroism now, as it seems a rather 'dangerous' buff.

Perhaps we won't even be able to maintain mana on a 2AB AM/Abarr rotation during BL, should we be resorting to casting as many AM's in between ABs/ABarrs while BL is up? How would this affect our DPS and could we even get into situation where we'd be off best canceling BL (seems like the craziest idea ever)?

Just some questions that popped up :O!

Reforging Haste into Mastery comes to mind, eventho now at 80 I hear we don't really have mana issues at all.


Edit: I just regemmed to reach the hit cap, replaced a few int/haste ones into pure int - then reforged all my gear with crit/spirit into mastery and then did the same for haste on the other few gear pieces without crit. Leaving me with 915 haste >.< Sound about right? My DPS on a target dummy is 9.5k~ after 6mins of casting.

Sounds about right?