Or MoE could just be removed, as fire doesn't need nearly as many talents crammed into its tree as it has now. Arcane Concentration can be the mana regen for fire and arcane, and fire's ability costs can be dropped further if they want us to be able to actually DPS until evocate is available (or, crazily, make evoc's cooldown far more reasonable for poor fire.)

I would be very interested to see the sim numbers for a 7 minute patchwerk fight doing molten/scorch versus mage armor (potentially also scorch) for DPS.


(edit) also I said this over on the BB, but I'm not sure what mages are members, and I'm likely to get called a retarded schoolchild again for bringing this up out here as well but:

Casters already have resource management, it's called time. That time that you spend casting is what physical classes spend waiting on resource regen. Balancing instants with moving, knowing when to cancel a cast and move or when to keep standing there, that's resource management as a caster. Maybe it's not significantly engaging enough, in which case you add in more procs/chaos/spells, or you decide that part of the casters short-term toolbox is to regen mana.

Some also have blue-bar management like warlocks, where at least originally (I hear from a guildmate that at 85 they barely need to lifetap ever) they lifetapped once every 30-40 seconds, which meant it was something you planned for like a short cooldown and worked around. You might delay a tap (back when spellpower didn't increase it) due to trinkets up and double-tap later, or know there's movement coming up shortly so delay it.

Even arcane's mana planning is on a two-minute time horizon which personally I don't find engaging but I understand others do. But fire's cycle is a full four minutes and just doesn't offer interesting decisions or gameplay.

Aspect of the Viper was mana management: just because it was "mana management" didn't make it by definition a good design, and those who disliked it weren't just mindless monkeys who wanted to spam a single button and do top DPS, they were good players who had legitimate reasons to object to the design.

I'm not sure tying MoE to crit is still a viable design direction, given the brutal ratings hit at level 85. With low crit rates, it probably wouldn't be sufficient even to give a completely free fireball after a crit, at least not until perhaps higher tiers of gear where crit rates rise somewhat. Furthermore, with crits being rare, RNG plays an even larger factor in our dps, as a dry spell not only hurts dps (and mps from free hot streak pyros), but if we are made to rely significantly on a crit -> mana talent, also could seriously hamper passive regen forcing a mage into scorching until a mana CD is up.

From a flavor standpoint, giving free casts after a crit even if balanced and effective seems dangerously close to being a clone of clearcasting. It might be a better idea to tie a new regen ability/talent into the other focus of the fire tree, dots. Perhaps giving mana back when X dots are on a single target, or when combustion is up, or a fraction of the mana back each time any dot ticks/crits (where the relatively high number of ticks/second as compared to fireball casts/second would reduce the chance of dry spells).

All this is however straying dangerously close to wish-listing or trying to design the game for Blizzard, and ultimately the simplest and likeliest solution here (although the most boring) is to just adjust the spell costs.

I strongly dislike the idea of using mage armor as a fire mage, as it just seems really not fun to have a highly appealing molten armor which ties into talents and stats integral to a fire mage, and then have it simply be unavailable due to spell cost and regen balancing. An "always use mage armor" model seems to me to be a fallacy, as int/mana pool/replenishment scaling would eventually lead to optimal play where you swap armor mid fight and use molten armor an increasing fraction of the fights as gear improves. This seems unappealing both due to scaling issues, and also from a fun standpoint (even if we're balanced around it, it really isn't fun to feel like you're doing subpar damage for a significant fraction of the fight).

Assumption set 1 - Fire Mages are intended to be able to cast a normal priority system without relying on Scorch for mana adjustment purposes under normal conditions throughout an entire fight. Abnormal conditions (half your DPS dies) can extend the fight to such a duration where you will still have to rely on scorch. Scorch becomes a safety net moreso than a frequently used tool.

Result of that - Mana tools or costs need to be adjusted, as that is definitely not currently the case.


Assumption set 2 - Fire Mages are intended to, past a certain scaling point in time depending on available mana and regen, run out of mana if they do not use Scorches to extend the point further. This point is not necessarily within the duration of all or any encounter, but also is not necessarily outside of the duration of all or any encounters.

Result of that - Scorch needs to be used to allow the mage to continue casting. This can take three general forms -

Form 1: Cast normal priority list, fall through to scorch when out of mana and continue on blindly from there (Worst option from a DPS perspective)

Form 2: Cast a priority list that includes interspersed scorches at such a frequency that the overall priority list will allow you to be at or around zero mana at the close of the fight duration. (Better option)

Form 3: Cast a priority list that includes interspersed scorches at such a frequency that at 35% target health the mage can swap to a scorch-free priority system and maintain that system until target death which should coincide with the mage being at or around zero mana. (Best option)


In all likelihood, and to use Lhiveras analogy from the beta thread, I think assumption 2 is the most likely. We wouldn't have been handed a bolt and a wrench if we weren't meant to turn it. Regardless though, if we have discussions with these two basic separations of thought in mind both can be productive as both are potentially valid. Raid balance testing should show what the answer is.

I absolutely agree — I've also been trying to think of ways that MoE could be changed, but it seems that stepping on the toes of Clearcasting is nigh unavoidable. If it were to regen based on our DoT's (Ignites, LB, whatever), it can't be where it is in the tree as both of those talents are in deeper tiers.

Perhaps it can give an increased regen effect after a crit for 10-15s, or when > 2 targets are taking Fire damage — and just limit it to once per minute.

Boy, people who dislike the idea of using Scorch to manage mana are not going to like the results of the tests I just ran. The full explanation of my test can be found on the official forums, but I'll summarize here.

I lowered my ratings to what I deemed would be Cataclysm-appropriate levels: 21% crit before Molten Armor, 9.82% haste, and a mana pool two-thirds the size of my usual. If anyone wants to compare the relative mana pool of a character on the beta with what I was running, Fireball cost 2.4% of my total mana and Living Bomb cost 3.3% of my total mana.

After playing around for a while, I found that--after an initial Fireball-spam burn phase--alternating between mana-free casts and mana-costing casts allowed me to drift slowly down my mana pool. Mana Gem usage allowed me to bump myself back up periodically, but I still generally had somewhere between 5% and 20% of my mana left when Evocation came off of cooldown. Here's the kicker: I was able to cast continuously for sixteen minutes and only went out of mana twice--once after eight minutes and once at the end of the sixteen-minute period. I never spammed Scorch--other than than at the eight-minute OOM point, when I cast Scorch and maintained Living Bomb for twenty seconds until my Mana Gem came off cooldown and gave me enough mana to sustain my normal rotation until Evocation came off cooldown. In fact, I never even cast Scorch twice in a row unless it was immediately preceded or followed by two mana-cost spells in a row. Hot Streak breaks the monotony by doing extra damage on your "free" cast or letting you cast two Fireballs in a row.

At the end of the fight, Living Bomb and Fireball each accounted for roughly a quarter of my damage done. Scorch was at 15% of damage done, with Pyroblast! at around 10%. In terms of mana gain, Master of Elements accounted for slightly more than my mana gem.

Overall, the feel of playing like this was rather nice. It has a good rhythm, moments of disruption due to Living Bomb's timer and Hot Streak procs, and the capacity for burning with Fireball spam. It might be worth looking into--especially since, due to its interaction with our mana cooldowns, you wouldn't be able to simply cast X Scorches at the beginning and then switch to Fireball spam.

I am in a similar boat.

I am roughly 2% haste away from getting the 7th LB tick. However, the ONLY way I can achieve this is by a) retalent, dropping 2 points in piercing ice, and picking up 2 in nether pres. b) regear, which will involve dropping crit. c) reforge, which will also involve dropping crit.

With crit rates dwindling so low I cannot justify losing 2% more, just to pick up 2% more haste and a 7th lb tick.


I am not sure if the dps increase would be even worth it, but I have definitely been giving thought to trying it.

You ran this with a level 80 in WotLK, correct? I'm not sure of the spell cost differences or coefficients but since it's not a Cataclysm 85 mage with the current beta build, it's not exactly a fair test. If you can get access to the beta with a premade 85 and rerun a similar test, that'd be much more helpful.

I'm not sure I follow the point of your tests Bashram. We already knew that we could cast essentially indefinitely by weaving scorches into the rotation. As Zaldinar pointed out, it's only a matter of time before someone throws up a simulation that tells you the optimal number of fireballs between scorches for a given fight length. It was never a question about finding a viable way to do damage for the entire fight, but rather is it competitve.

The problems are that it's not clear as to whether it's intended to function in that way, and that it appears we are underpowered compared to other classes when we are forced to use scorches due to mana concerns. Fire has been designed for 3 expansions with scorch as a DPS loss to fireball, and the only reason scorch was ever cast in any expansion was for debuff applications. They finally gave us an alternative (HS) to applying said debuff only to force us to use scorch for another reason? That's like a slap in the face.

Think about the actual cast time you spent on Scorch (seems like a significant amount of time if it accounted for 15% of the damage, knowing that it's significantly less DPS than fireball). How much DPS was left off the table because we couldn't spend that time casting fireballs?

I tested 7LB ticks with 1/3 NP and 3/3 in piercing ice. With the 12% haste buff from T10 HS Proc i succesfully got 7LB Ticks. I even got 8 Ticks, when Blackmagic and my Engineer Enchants were up, it was about 1600 haste++ for 8 Ticks. And 1200Haste for 7 ticks with 1/3 in NP.

The major Problem i got with all this, that my crit chance got so low, that HS procced a lot less, so no 12% haste buff from 2T10 and then no 7 LB ticks.

Yes and no. I raided Naxx and early Ulduar with a hunter and the basic thought process is very similar to that of Aspect of the Viper. Mechanically it is different, but as far as the logic that drives it it's essentially the same. You want to use this ability early in the fight so you can burn the end of the fight down hard and quick.

I think that the form 3 should be revised, upon further thought and reflection, to be that you burn as hard as you can until Evoc is on cooldown, and then when your mana reaches the point where you will have to float at to be able to burn through the molten fury range you prioritize Scorch over Fireball to float around that number. There is no real magic ratio of scorches to fireballs, as the random returns from MoE and random lack of mana use from Hot Streaks will throw those off fairly significantly from fight to fight, but you as a person can easily say "I have 40% mana right now and thats my burn goal, scorch time...Oh, 45%, a few fireballs... back to 40%, scorch again".

You can accomplish the same basic goal by using scorches in your rotation earlier in a fixed ratio to fireballs used, but it will be significantly harder to do well. My sim is showing that doing a 1 scorch per X seconds type logic is sketchy at best and produces a lot of awkward variance at not a whole lot of benefit other than pushing that OOM point back a little bit. I'm working on teaching it to do the balancing act at a certain percentage of mana to see if that works out better numerically.

In the end all this may be not worth even thinking about though until higher levels of raiding, scorch and fireballs relative DPS when hot streak frequencies are taken into account are not overly disparate, so it may not be something we need to be concerned about for another nine months or so.

Don't bother with the hot streak. The Hot Streak Pyroblast! doesn't spread with impact, nor does it get included in combustion.

Rotation I've used, on the whelps:

Living bomb -> (non-hot-streak) Pyroblast -> Blast Wave (will proc impact) -> Impact Fireblast -> Reapply LB to main target ('cause Impact will leave you with only 2) -> Flamestrike spam

If you're really good at timing, and have good lag, you can get LB to explode from all targets with impact... which nets absolutely comical damage on those whelps. Unfortunately, they all decided to eat me very shortly afterward.