Takes some gall to call LB useless.

Firstly, it'll work in conjunction with World In Flames. Secondly, it requires nothing to maintain, thus it works -while you're casting something else too-. Combine Flamestrike's floor DoT effect with Living Bomb, an Ignite from BW and AE spam, I'm not suprised it's not doing much DPS of it's own. It's not supposed to. It adds to AoE DPS and that's that, it's not supposed to be something you switch on and make everyone's FPS drop to 3 from damage overload.

Yes, I raided as frost for a bit. From our first Illidan attempts until Sunwell came out.
And I hated that my water buddy got one-shot by just about every AoE out there

Improved Water Elemental - the screenies put it at +5 seconds / +10% health/mana per point.
He used to be around 10% of my damage on boss fights from WWS reports, rarely less than 9%, and only sometimes above 12%/13%.
Performance on trash was lower, around 7% on trash maybe, but I can't claim to be good at micromanaging pets.


How does that scale our talents?
Water Elemental - 10% for 1 point.
Improved Water Elemental - 1% per point if it's 5s/point, 2% per point if it's 10s/point. And better pet survival.
Molten Fury - according to Rawr is only a bit more than 4% total DPS (6 minute fight, 15% MF time).


What specs could there be then?
0/34/37 - cookie-cutter MF/WG Frostfire Bolt spec that's been mentioned/disussed.
0/30/41 - skip 4 points in MF/Comb/Pyro to get the WE. Lose 6%/7% to gain 10%.
0/27/44 - dropping points in Fire Power hurt and should be considered if Imp.WE increases uptime by 10s/point.
If it's only 5s/point, it will most likely be a net loss. The increased survival is a great thing, but if pet survival is really that much of an issue, it should be better to just go back to 0/34/37.


Notice that the liches Ras Frostwhisper (Scholomance!), Kel'thuzad and Rage Winterchill had high frost resistance and the frost wyrm Sapphiron was immune.
If Blizzard remains true to their interesting and exciting mechanics, speccing WE may not be worth it.
If frost resistance is not relevant (Rage Winterchill is hard!) and crowd control becomes important in 10-mans and maybe 25-mans, the extra novae might turn the tides in favour.

Also, bear in mind that the WE has a high base damage and low scaling, and from the numbers I've seen the base damage of our spells become less and less important.

Mind you that shadow priests and moonkin are given an AoE spell. If AoE becomes important and AE remains our best AoE, it might be necessary to drop another 2 poins for arcane threat reduction.


So, what does that mean?
If Frostfire Bolt with an elemental spec turns out to be highly competitive and if the Water Elemental can survive okay'ish (i.e. at least as well as it is now), speccing into WE looks like a damage increase.
Speccing imp. WE is only worth it if it gives 10s uptime per point.
And it all depends on WE and spell scaling, and which synergies of Frostfire Bolt remain after its reviews

This was really not often a problem. You had to take an uptime hit at times to time deployment so that it fell between ability timers, but there weren't many fights where I couldn't get most of the 45-sec duration out of him. He's immune to a surprising number of effects.

That said, I still think pets should get the 50% aoe damage reduction that Hunter pets get from their Avoidance skill (and hunter pets should, of course, get it for free). Losing pets to these abilities -- whether the problem is real or perceived or partway between -- is the single biggest deterrent to people actually using their pets in raids.

From 50 to 70, his base damage increased by 16.3 per level, so we can expect the average base damage to increase from 522 to 685 from level 70 to 80, if this remains consistent. Beyond that, assuming a raid debuffed target, he'll gain about:

2.5 / 3.5 * 0.4 * 0.83 * 1.075 * 1.1 * 1.05 * 0.949 / 2.5 = 0.1118 DPS per point of +Damage

I'm finding that with most fight durations, you can expect about 50% uptime (barring early deaths) with Cold Snap, Ice Floes, and Improved Water Elemental, up from 37.5%, the most common uptime without Imp. WE. So Imp. WE increases the DPS scaling we get per point of damage from 0.0419 to 0.0559.

Well, that depends somewhat on how important controllable burst damage is -- sometimes, it can be of considerably more value to increase your DPS by 40% for 2 minutes straight (plus, of course, using it in conjunction with other cooldowns, heroism, totems, drums, etc) than the sustained DPS increase gained over a 10-minute fight might imply.

The real question is, is FFB not being affected by arcane talents a side effect from private servers, a bug, or just intended ?

Very nice post rowyn, thats just what I was after with regards to the WE.

A few points to address on the subject:

WE is less threat - This is true, but with a potential 81% threat on FFb it may not matter (although WG+Shatter might result is some godawful agro spikes).
WE Scalling - we may see that WE makes a better entry level spec, with MF pulling ahead at higher progression (depending on gear).
WE mana - if the WE's bolt stay at a reasonable cost, with buffed up JoW in raids as high as it is the WE might last out its mana bar longer (plus, we havn't seen elem shammy tallents yet - they might get some group mana buff).

At best, the WE seems to me to be pulling ahead by only a few %, and given how situational it can be (ONOES AOE!) i'm not sure that's going to be enough to justify it over MF.

With regards to AoE: I think with AoE being thrown around more liberally and being made more 'tank and spank' viable, that mage AoE won't matter as much. between the ~6tank-capable classes in your raid (those who can throw on tank gear and pick up 2-3 adds each, regardless of spec), combined with the new super-hurricane for AoE damage suppresion, I think burst damage in AoE pulls will become less of a priority over a standard tank and spank setup, even without prot pallies (although they'd still be better at 1-man tanking large packs). I can live happilly staggering novas with the other mages in the raid and blowing CoC to fuel my AoE on shatter alone. The DPS might not be as good as AE spam, but the DPS won't matter as much. On that not, could there be a possibility of long-duration AoE pulls where mana efficiency on your AoE becomes a serious concern? A 5-minute long AoE Boss(es) encounter would be rather orriginal, and opens up a huge amount of options for making an interesting fight.

I found almost nothing worked properly on private servers I wouldn't give too much weight to this.

Well, I whipped up this dramatically oversimplified comparison for my own edification:

Base Stats
1800 damage
25% crit before talents
99% hit
30 mana/sec regen while casting
15000 mana
All calculations assume that Burnout works like Spell Power, which is the worse of two possibilities

DPS Summary
2148.55: Fireball (Deep Fire Spec)
2063.75: Frostfire Bolt (Elementalist Spec, vs. non-fire-immune target)
1875.30: Frostbolt (Frost Spec)
1529.82: Frostfire Bolt (Elementalist Spec. vs fire-immune target)
1378.50: Frostfire Bolt (Frost Spec, vs. frost-immune target)
1201.91: Frostfire Bolt (Fire spec, vs. fire-immune target)

DPM Summary
21.92: Frostfire Bolt (Elementalist Spec, vs. non-fire-immune target)
16.25: Frostfire Bolt (Elementalist Spec. vs fire-immune target)
13.75: Frostbolt (Frost Spec)
12.66: Fireball (Fire Spec)
11.92: Frostfire Bolt (Frost Spec, vs. frost-immune target)
8.89: Frostfire Bolt (Fire spec, vs. fire-immune target)

Now, it's important to remember, these are just the bolts, not full rotations.

- Frost Mage: Winter's Grasp will increase his Frostbolt crit rate from 40% to about 51% and every 10th spell will be an Ice Lance with an 85% crit rate. His Ice Lance will do 5% more damage due to CttB. And of course he also has a Water Elemental to factor in.

- Elementalist Mage: Winter's Grasp will allow him a Shatter Combo, one Frostfire Bolt out of 9 at 91% crit (increasing his average crit rate from 41% to 46.5%), and once Ice Lance at 85% crit rate.

- Fire Mage: Very simplified total crit estimate of 37%.

My conclusion is that, yes, if in fact FFB is supposed to benefit from all these Frost and Fire talents, it is indeed quite possible to build a competitive elementalist spec.

Math for the bolts is below.


Frostbolt
Base: 830
Gear: (3 / 3.5 * 0.95 + 0.1) * 1800 = 1646
Avg: (830 + 1646) * 0.99 * 1.4 * 1.06 * 1.05 * 1.05 * 1.1 * 1.05 = 4688.25
Avg mana: 505 * 0.85 * 0.97 - 2.5 * 30 = 341
DPS: 4688.25 / 2.5 = 1875.3
DPM: 4688.25 / 341 = 13.75

Frostfire Bolt (Frost)
Base: 780
Gear: (3 / 3.5 * 0.95) * 1800 = 1466
Avg: ((780 + 1466) * 0.99 * 1.25 * 1.06 * 1.05 + 20) * 1.15 * 1.1 * 1.05 = 4135.51
Avg mana: 530 * 0.85 * 0.97 - 3 * 30 = 347
DPS: 4135.51 / 3 = 1378.50
DPM: 4135.51 / 347 = 11.92

Fireball
Base: 1010 + 29
Gear: 1.15 * 1800 = 2070
Avg: ((1010 + 2070) * 0.99 * (1 + 0.37 * 1.275) * 1.1 * 1.03 + 29) * 1.15 * 1.1 * 1.05 * 0.949 = 6445.64
Avg mana: (655 * 0.97 * 0.97) - (655 * 0.99 * 0.37 * 0.3) + (150 * 0.99 * 0.37) - (3 * 30) = 508.97
DPS: 6445.64 / 3 = 2148.55
DPM: 6445.64 / 508.97 = 12.66

Frostfire Bolt (Fire, vs Fire-immune target)
Base: 780
Gear: (3 / 3.5 * 0.95) * 1800 = 1466
Avg: ((780 + 1466) * 0.99 * (1 + 0.37 * 0.625) * 1.1 * 1.03 + 20) * 1.1 * 1.05 = 3605.74
Avg mana: (530 * 0.97 * 0.97) - (530 * 0.99 * 0.37 * 0.3) + (150 * 0.99 * 0.37) - (3 * 30) = 405.70
DPS: 3605.74 / 3 = 1201.91
DPM: 3605.74 / 405.70 = 8.89

Frostfire Bolt (0/34/37 Elementalist spec, non-Fire-immune target)
Base: 780
Gear: (3 / 3.5 * 0.95) * 1800 = 1466
Avg: ((780 + 1466) * 0.99 * (1 + 0.41 * 1.8) * 1.1 * 1.03 * 1.06 + 20) * 1.15 * 1.1 * 1.05 = 6191.24
Avg mana: (530 * 0.85 * 0.97) - (530 * 0.99 * 0.41 * 0.3) - (3 * 30) = 282.46
DPS: 6191.24 / 3 = 2063.75
DPM: 6191.24 / 282.46 = 21.92

Frostfire Bolt (0/34/37 Elementalist spec, Fire-immune target)
Base: 780
Gear: (3 / 3.5 * 0.95) * 1800 = 1466
Avg: ((780 + 1466) * 0.99 * (1 + 0.41 * 1.0) * 1.1 * 1.03 * 1.06 * 1.05 + 20) * * 1.1 * 1.05 = 4589.47
Avg mana: (530 * 0.85 * 0.97) - (530 * 0.99 * 0.41 * 0.3) - (3 * 30) = 282.46
DPS: 4589.47 / 3 = 1529.82
DPM: 4589.47 / 282.46 = 16.25

Yes, that is precisely my own interpretation of it, hence a bit why I am pointing it out, since we're doing TC on a mechanic that I doubt its how it works on live servers.

I can't imagine a good reason why arcane would not affect FFB, at all. It would simply be inconsistent with every other mechanic in the game.

-----
To add to lhivera's calc, the interesting thing to note is FFB DPM -- nearly the double of fireball (!), and almost the same DPS.

1800 spellpower is perhaps conservative, Lhivera. My sense is that at very high gear levels for WotLK, FFB will simply be outscaled. You can hit 1800 raid buffed on live or near enough.

Lets go out on a limb and say that end game raiding in the expac is in the neighborhood of 3000 spellpower.

Yeah, there's a big flashing "IF" over all this. Not working with Arcane talents sounds guaranteed to be a bug. Frankly, applying Winter's Chill while doing Fire damage also sounds like a bug. And of course we still don't really know how Burnout works (it may not even be working -- apparently Eternal Flames simply doesn't do anything at present). But, it's fun to speculate.

You can if you're Arcane specced. Can you if you're not? I'm not so sure of that. 3000 sounds way, way high to me. +Damage on gear increased from around 600-700 to around 1200-1400 (non-arcane specs) from 60 to 70, IIRC. Would we expect it to double again, to 2400-2800? Or would we expect another +600-700 damage added? Impossible to say at present.

Remember that the more damage you add, the more expensive it becomes. Yes, item budgets are increasing, but not as fast as the cost per point of damage is increasing. I would more expect to see the +damage increase slow down considerably, with gear carrying increasing amounts of crit, haste, and base stats instead.

One interesting thing to note: it looks like the rating required for a 1% increase to a rated stat is going to nearly double (42.9 for 1% crit, for example; 24.5 for 1% hit). Meanwhile, the amount of Intellect per 1% crit is only increasing from 80 to 90. So in relative terms, the DPS value of Intellect is going up considerably compared with the DPS value of Crit Rating.

Now: 1 int = 0.0125% crit; 1 rating = 0.0453% crit; 1 rating = 3.6232 int for crit increase
Lvl 80: 1 int = 0.0111% crit; 1 rating = 0.0233% crit; 1 rating = 2.0979 int for crit increase

The DPS comes from extreme crit scaling with 200% crits + Ignite + Shatter/WG.

As fro DPM - deep fire only has MoE and loses it via burnout. FFB has -15% from a talent and MoE with a shatter build.
And only costs 530 mana compared to FB's 655.

Skill list: Wrath of Lich King Stuff (was 'Blizzcon' thread)

Lhivera:
Could you add the calculations for a 33/38/0 arcane/fire spec for Fireball spam?
It should outscale deep fire significantly and offer vastly better DPM.

Sure, link the spec you like so I make sure I get the point placements right.

It's basically "Arcane Power + Spell Power + best fillers" in arcane, and "Best you can get for 38 points" in fire.

33/38/0
War Pirate :: Talent tree Mage

The MoE/PWF/Pyromaniac filler point spread can be argued.
You can maybe 1% crit from the extra int (arcane mind), and will lose ~1.5% crit because you have to regear/regem for elemental precision.

Thanks for your quick replies!

Fireball (33/38/0)
Crit: ~39%
Hit: -3% (no elemental precision)
Base: 1010 + 29
Gear: 1.09 * 1800 = 1962
Avg: ((1010 + 1962) * 0.96 * (1 + 0.39 * 1.45) * 1.1 * 1.02 * 1.03 + 29) * 1.15 * 1.1 * 1.05 * 0.949 = 6543.08
Avg mana: (655 * 0.98) - (655 * 0.96 * 0.39 * 0.2) - (3 * 45) = 457.95
DPS: 6543.08 / 3 = 2181.03
DPM: 6543.08 / 457.95 = 14.29

Of course, if Burnout works the good way, deep Fire Fireball spam looks very different:

Fireball
Base: 1010 + 29
Gear: 1.15 * 1800 = 2070
Avg: ((1010 + 2070) * 0.99 * (1 + 0.37 * 1.625) * 1.1 * 1.03 + 29) * 1.15 * 1.1 * 1.05 * 0.949 = 7009.58
Avg mana: (655 * 0.97 * 0.97) - (655 * 0.99 * 0.37 * 0.3) + (150 * 0.99 * 0.37) - (3 * 30) = 508.97
DPS: 7009.58 / 3 = 2336.53
DPM: 7009.58 / 508.97 = 13.77

Ah, found it:

http://i32.tinypic.com/2rrm15l.jpg

To give folks here some idea of how gear is going to scale upwards. Pity they didn't include more caster loot in there.

Burnout is actually a 25% mod to the entire crit as tested and verified on the real alpha realms, so the tooltip is wrong. Not sure what that means.

It doesn't, though. Remember, we're only looking at pure bolt spam here, which is a number on which Fire has an advantage. When you start mixing in the cooldowns and procs (water elemental, winter's grasp, icy veins/cold snap, etc), the other specs narrow the gap considerably.

Using the same starting stats, I came up with the following about a week ago:

Rotation DPS (assuming 7-min fight for cooldown uptimes)
2480.67: 9x Fireball / 1x Scorch (if Burnout rocks)
2479.38: 12.33x Arc Blast / 6.67x Arc Barrage / 1x Fireball on NP Proc
2357.50: 7x Frostbolt / 2x Frostbolt Shatter / 1x Lance Shatter / 50% WE Uptime
2281.15: 9x Fireball / 1x Scorch (if Burnout sucks)

Rotation DPM (7 minutes again, remember, spirit regen not yet accounted for)
15.37: 7x Frostbolt / 2x Frostbolt Shatter / 1x Lance Shatter / 50% WE Uptime
13.88: 9x Fireball / 1x Scorch (if Burnout rocks)
12.67: 9x Fireball / 1x Scorch (if Burnout sucks)
5.14: 12.33x Arc Blast / 6.67x Arc Barrage / 1x Fireball on NP Proc

If you're interested, I can paste in all the rest of the math. This only accounts for highly abstracted cooldown use (averaged percentage increases), it's not a detailed analysis, but it shouldn't be too far off.

Huzzah! Do you have any info on the source of the testing? Is there more to read? Does this mean Eternal Flames is likewise resulting in 230% crits? Any info on exactly how the mana cost works?

Honestly, I still believe this is the only way it makes sense. With CttB providing a straight 5% DPS increase and a positive utility effect, it made no sense for Burnout to provide a 4.something% DPS increase with a negative DPM effect.

Would you mind running those numbers for Arcane with a rotation of Arcane Missiles - Arcane Barrage - Arcane Missiles with Arcane Blast being used on NP procs (use the 5% per volley model for NP procs with Arcane Missiles)?

I really think that is the rotation that the developers are trying to base Deep Arcane on, especially if they don't give us another set with a bonus like 2T5.

The reason why I think AB would be better on NP procs is for mana conservation since the damage would be similar to Fireball but the mana costs are much much less.

Does anyone have a coefficient for Arcane Barrage? At 1.5/3.5 it's pretty bad.

I'll put it on my list, but it probably won't be today (I'm getting backed on up real life stuff).

Actually, fully-talented AB vs. totally untalented deep-Arcane Fireball, Fireball damage comes out way ahead without 2xT5 (3739.43 avg dmg for AB, vs. 5030.68 avg dmg for Fireball).

One thing I'd like to do before calculating any new spells or rotations is get some kind of consensus estimate for reasonable level 80 raiding stats (Int/Spi/Damage/Haste and pre-talent Crit). Deedre may be right that I'm being way, way too conservative.