I touched on this upthread; arcane is in a unique position where it has access to 3 potential rotations that could all provide equal DPS, and allow the mage to pick up some flavour talents in other trees (at the cost of puritan arcane talents). The tree is very close to achieving this right now, but it needs a little more tweaking to get it dead right. The rotations are, obviously ABar/AB, ABar/FB and ABar/FrB. All 3 are coming close, with arcane and FrB needing a little bit more to push them in line (this problem should be addressed in the respective trees - AB getting more support in arcane and FrB having a little more from early frost. Since frost DPs needs a small buff to be in line with fire this would seem an ideal solution. One idea would be to change Improved frostbolt to a 3 point talent at 0.2/point). The second problem in reaching this sweet spot for arcane is that a streamlined arcane tree still has too many 'essential single-target' DPS talents. If I cut it down to 53 points for improved scorch I'm losing 2-3 arcane talents who's only/primary function is spell damage. If it were possible to get down to ABar on 51 points and a fully talented AB was competitive with 20points of fire fireball, or 20 points frost frostbolt, then the arcane mage would find themselves able to choose a preferred nuke between AB, FB or FrB - and pick up some of the chosen trees flavour (pure arcane getting lots of fun toys in IA and PC, fire having access to PoM/Pyro and some other buffed fire abilities, frost getting IV/permafrost/shatter/whatever else you want from there).

Another problem for arcane mages trying to build around this idea is the feeling of obligation to get improved scorch. If the talent ends up defining the class for 5/10 mans then any mage without it will be branded impotent and laughed out of instances.

This doesn't address where AB is going in terms of the ramping debuff though... I'll think more on that subject when blizzard make their minds up about it.

As far as the crit debuff goes, a deep frost spec will probably serve the group much better in 5/10 man situations. From my previous 10 man experience, the bosses are rarely about min/max and pumping out every last possible dps, but rather about strategy. The strategy/survivability/sustainability aspect make Frost a clear favorite for 5/10 mans.

However, in 25 mans, survivability and sustainability seem to be less important, and overall DPS is of a much greater importance.

It's also useful when soloing or in 5 mans where you want to drop a chill effect on something without having to close range or lose a lot of DPS like frostbolt requires. It's a decent "opener" that lets you get in another fireball while the monster waddles toward you. For frost, useful when they're running the frost glyph and lose the chill effect.

It's utility isn't entirely limited to immunity fights. But the secondary uses aren't valued much by raiders - they'll just make life more pleasant when doing solo work or small groups.

Except of course that in the past the WE's weren't survivable and are as significant a portion of frost DPS and timing as molten fure is to fire. Plus rooting monsters isn't desirable sometimes and lots of things seem immune to chill.

I leveled frost to 60 and played fire after that, and did nearly all the 5-10 man content in BC as fire, both in a raid spec and in a "PVP" spec. The former got me better numbers in the damage meter, the latter let me do more aggressive AOE or work with tanks that I significantly outgeared. I can count the number of times where I wished for a second ice block on one hand. If I'd had ice barrier, I probably would have used it but I don't think there would have been many fights where it would make much difference. (as frost, I used IB mostly as pushback resistance, something burning soul does adequately).

I have a friend who raided arcane most of the time, except for gimmick things like when we took four fire mages to Kharazan. He also had a great deal of success, and never missed the fire or frost talents. Instead he made good use of what he had. His damage numbers were similar to frost given similar gear level.

The 5-10 man content reward things mages just plain do. Polymorph, counterspell, decurse, blink out of stuns or to get back to range when the boss pulls you toward him or to avoid hazards, AOE whatever your favorite way is, ice block in an emergency, invis to dump threat if you get ahead. Any half decent spec was an asset in most fights. Arcane mages had advantages with counterspell/invis, frost with ice block, fire with AOE. If a player worked to the strengths, he was seen as an asset.

I think though ZA should be seen as the future of 10 mans...a 10 man that is harder (on speed runs anyway) than most of the 25 man content in BC. 10 mans will have their own progression, and they'll have Patchwerk/Brutallis type fights scaled for 10 man raid output. There will be fights where DPS is king, fights where raid coordination is king and fights where survivability is king. Probably also fights where AOE is king (like Hyjal trash waves) and whatever else they've come up with for 25 mans scaled to 10 man raid compositions.

If you play your spec well, you'll likely feel it gave you advantages when you raid with it. If the DPS isn't so far apart that you get depressed at the damage meters, the rest of the spec is accentuating one or another aspect of the mage class. If you like that aspect, you'll take the spec that does it better.

Or that's the hope anyway.

Min Maxers will certainly respec for any tough fight to optimize their dps. But they probably won't bother for farmed bosses.

My objection to Alar wasn't so much on first kills, but when on farm. I was willing to respec frost for Alar, but it was a pain to run around with a gimpy spec for the rest of the instance. Mages either had to be fire specced and use gimpy frost spells on Alar, or had to spec frost on the entire instance, or had to respec to frost for Alar and respec back to fire when done and get summoned back.

Frostfire offers a way out. Spec optimally for the first few kills, but once the boss is a safe kill, just use your preferred instance spec plus frostfire.

My worry is that Blizzard usually takes with the other hand when they give with one, particularly when it comes to spell school mechanics. They keep touting mage's spell school choice as an advantage, when it fact all it does is give Blizzard an excuse to create mobs and bosses which are immune to our schools - something they don't do for classes that have less school choice (shadow is the obvious one). The only time our spell school variety has been useful rather than a respec pain is on school rotational fights like Chromag.

As such, my fear is that giving us frostfire bolt just indicates that Blizzard is going to be drowning us in fire and frost immune bosses and mobs now, and just say "use frostfire" when mages bitch that this kills their dps.

In my experience, I've seen far more mages sitting out rather than respec. Not necessarily out of the mage choice, but raid leaders choice. And were I a raid leader I'd probably do the same.

As a raid leader, and a mage, it's far easier to have them sit out, sitting is typically the fastest way...since most mages take a while to change their spec and UI, though I managed to get it down to <3 min for changing my spec, and my UI from Fire-->Frost and Frost-->Fire.

Also pushing this will be that the Arcane mage can maintain the scorch debuff without dps loss (aside from the chance to proc MB).

As for dedmon's sim, I went through the arcane calculations fairly carefully to get them to match up with my spreadsheet. I didn't check the fire/frost portions as carefully because I don't have those specs on my spreadsheet. For those asking about rotations other than ABr->FB, here's a sim run from my setup (which is based on the naxx 10 spreadsheet numbers we've been using and have most of the changes sondag put in with the exception of pom+pyro):

3638 Mage_Arcane_Fire
3589 Mage_Arcane_Frost
3227 Mage_Arcane_AB

Fireball has a 1.5% miss rate here and everything else is capped, so fire is slightly further ahead if you hit cap. All in all, you can take your pick for whether you want Fire or Frost. I wouldn't trust the sim numbers to be authoritative for differences under 3%. Also, using FFB with a 53/10/8 build produces dps below the frost numbers here and you don't get any flavor talents from the fire/frost tree, so I don't see a point to speccing that way.

Any arcane rotations discussed here are 60-70% arcane damage at minimum. I've always considered the chromatic spell buffs to be part of the arcane tree identity and those lead directly to viable off tree nukes. Tweaking rotations to allow AB as a viable filler isn't hard (I like the ABr gives a short duration AB buff solution because it leaves stacked AB unaffected) but keeping AB in line as gear scales is difficult due to Ice Shards and Ignite. The crit bonus is actually a significant fraction of the difference showing up in the above numbers. Without the huge amount of crit raid buffs provide, the numbers are much closer.

Grey, thanks for reposting those numbers. For standalone arcane, as you've said it needs better crit scalling. The problem is that such crit scalling needs to be arcane-specific to avoid further pushing arcane/fire ahead of the crowd. I don't expect for a second that blizzard will look into that problem untill long into wrath, once T8+ gear is around and the crit difference starts to really show disparity between deep fire and arcane.

The difference between arcane/fire and arcane/frost is close enough that random factors can throw it one way or another on any given fight, so thats negligable.

What specs are you using for the above results? I would assume 53/18 (in fire or frost) or 68/3/0 for pure arcane.

The pom+pyro part was the kicker that made my arcane_fire the definite best choice in tests.. would it be hard to calculate in your spreadsheets too? i'm interested in how "my" rotation is performing.

I don't see why I would bind MI and fireball together; to me, the most interesting part of MI is what it does to your threat, not the actual dps increase.