Yeah, and it also doesn't even matter if the hasted frostbolt has a lower average crit chance, it doesn't change anything. That frostbolt would have a lower crit chance no matter what. It's a haste buff, not damage, so you're still getting the same amount of extra frostbolt casts due to the saved time. In other words it doesn't even matter.

The change to Water Elemental is very welcome, and I'm happy to see frost get a nice buff. A spectacular buff really, considering the changes to pet scaling as well.

However, you cannot in good conscience double the uptime of the WE when doing theorycrafting. The WE is up, in a typical fight, more like 60-70% of the time, due to double uptime in the first two minutes. In a four minute fight, the uptime for the WE is 75%... in a six minute fight, it's down to 4 out of 6 minutes uptime. But then ColdSnap returns, and the WE will be up two minutes in a row again.

This ends up as about a 1/3 buff to WE uptime, and the freedom to use Cold Snap as a defensive or offensive cooldown at a more appropriate time (rather than just after summoning the first WE).

But I'm unimpressed with the Ice Lance -> Frostbolt change. It looks to be a minor change at best, and that's if you are great about catching ghost charges. Is it a DPS loss if you don't use ghost charges?

I'm pretty sure using Brain Freeze is going to reduce the benefit of the new Frozen Core effect. Example -- imagine 30% passive haste, and a FoF proc every 6.67 Frostbolts.

If you're using BFF's, your average sequence is going to include 6.67 Frostbolts, 1 BFF, 1 Ice Lance:
(6.67 * 2.3 + 2 * 1.5) / 1.3 = 14.11

If you're not using BFF's:
(6.67 * 2.3 + 1.5) / 1.3 = 12.95

The FC effect is going to knock one of those Frostbolts down:
2.3 / 1.3 - 1.3 / 1.3 = 0.77 seconds

14.11 - 0.77 = 13.34
(14.11 / 13.33 - 1) * 100 = 5.85% haste

12.95 - 0.77 = 12.18
(12.95 / 12.18 - 1) * 100 = 6.32% haste


MyrddinE: EDIT: Response was product of backwards-think. Removed for reconsideration.

Is it possible to get 2 ghost charges off of FoF with enough distance?

i.e.
Spam FrB till FoF proc
Cast 2 more FrBs
Brain Freeze FB (w/ ghost charge)
Ice lance (w/ ghost charge)


Now this assumes that the BF FB leaves your hands at the same time as the 2nd FrB. This leaves 1s - 1.5s for the IL to leave your hands before the 2nd FrB (which left your hands at the same time as the BF FB) hits the target.

I have not played frost in some time, just curious.

I had a quick vision of using the new Frozen Core mechanic to squeeze in a BF FB just before the Ice Lance so that both benefit from FoF. And if the IL crits it gives you a hasted FrB.

Any reason why no one is considering FFB spam instead of Frostbolt spam? It might be obvious and I don't know since I haven't been in a frost based spec in a very long time.

Regarding the above linked spec. Wouldn't Empowered Frostbolt represent a larger DPS gain that Fire power at our current gear levels? it may not be the case when accounting for Ignite.

FFB spam would need to be more than 12% higher DPS than FrB spam to make up for the loss of TtW. However, with crit chances in the 75+ range, it's quite possible that a 0/20/51 build might outdo a 18/0/53 build. FFB would get full benefit from IL+shatter procs due to it's longer base cast time, and it's crit scaling of 2.8, in such a high crit environment, may make up for its shitty base damage.

Or not. I'm not a theorycrafter. But when Rawr is updated, I'll probably be plugging in a deep frost/FFB build to see how the numbers work out.

Empowered Frostbolt doesn't affect Frostfire Bolt, you probably confused that with Empowered Fire which does since a few patches. Unless they're going to change that as well - and cause hell when having to balance it for 0/38/37 specs at level 85.

Fire Power beats Playing with Fire though, so the the previously linked basic 0/30/41 template would become Link: 0/30/41 FFB+WE Template. Also, dropping Scorch and Permafrost (filler I guess) for Winter's Chill allows for 1 more point in PWF. Depending on fight, mechanics and personal fine-tuning as always.


Hot does this compare to 0/51/20 Frostfire Templates?
You gain Frozen Core (~7%), Arctic Winds (5%), Fingers of Frost (10%), Water Elemental (~12%).
You lose Hot Streak (10%), Living Bomb (7%), Burnout (12%), Empowered Fire (8%), Molten Fury (4%), (Pyromaniac (3%).

Those are Rawr values with napkin math guesstimates to changes, being generous towards the Frost centered build. Also ignoring Combustion (not sure if it's working correctly now) and how to juggle PWF with filler talents and Scorch vs. Winter's Chill vs. Warlock Debuffer. Also, the values of talents will vary greatly when changing other talents.


Will those 0/30/41 be viable for PvE DPS?
We'll know when the sims are updated. But honestly, it's not looking good. Fire is a bottom-heavy spec as you see, you gain about 45% DPS in the 31-51 point zone. The corresponding 21-41 point zone in Frost is more a 35% increase even with those buffs.

So 0/30/41 is likely 5-10% behind 0/51/20 specs, which already are behind Fire and Arcane specs. The major cause of those balancing issues is that the Fire tree from 20 to 50 point is simply loaded with 2% talents, while Frost has a lot of 1% talents and some stronger ones which aren't strong enough to carry the burden of the weaker ones.


So far the napkin math preview as first estimate. On another note, does [Glyph of Eternal Water] replace the old -30s summing cooldown glyph, or is it a new glyph?

I believe it's a new, alternate glyph, but apparently glyph vendors aren't available yet or something, so nobody knows for certain. But since they gave it a new name, where the old one would have been entirely appropriate (unlike Glyph of Scorch) for the new effect, it seems most likely to me that it's an addition.

I've done a lot of napkin math here:
World of Warcraft - English (NA) Forums -> Frost 3.3 Analysis (PvE)

Changes (excepting Water Elemental mana modeling) are in Simcraft but I can't quite get them to do what I want yet.

One thing that may be of interest is that it does look like Glyph of Ice Lance becomes superior to Glyph of Molten Armor with these changes.

Just looking at the numbers, is the Frozen Core buff that much of an increase to PvE DPS? At 30% haste, that Ice Lance would have to do at least 34.8% of the damage of a Frostbolt just to break even on damage, right? It's going to make that next Frostbolt GCD capped, so the only actual time you gain are the difference between your Frostbolt cast time and your GCD, which at 30% buffed haste 1.77s and 1.154s, respectively. So that Ice Lance cast (1.154s) needs to be worth the difference between the base Frostbolt cast and the GCD (0.616s). Is it really worth that much?

Do calculations from live exclude buffs on the elemental while calculations for PTR include full buffs?

We will have to wait and find out what the new pet scaling coefficients are, then see just how much dps the elemental is worth and wether it beats out LB, burnout and hot streak, which would nudge it towards the "nerf bat" pretty quickly. Intresting thought though, i would like to see a frost heavy FFB spec become viable as well.

Facinating prospects. On paper I guess this must increase the value of Crit while devaluing Haste a little bit. I am very intrigued by the “intangibles”.
Critting Icelances becomes a massively important part of maintaining rhythm in your rotation. An intangible Crit scaling factor that in practice might be more important than it is possible to calculate.
Cooldown management becomes easier, WE is no longer part of Cold Snap decisions, this is very significant.
Except for movement situations, Brainfreeze seems to be moot, if not downright penalising since it carries no rotational benefit (FrB -> FoF, IL -> FrB speedup).
Enduring Winter becomes a 1 pointer at max.
Winters Chill is rendered obsolete unless there are no other Crit debuffers. Scorch/Imp SB seems vastly superior, raidwide.
This pushes a 20/0/51 build with SotM and Spell Impact maxed surely? Even PoM is in range.

I've just stood up so I may be quite confused but I don't know why Winter's Chill should be rendered obsolete (it's still 3% crit for your FB and the highest Mage and Warlock DPS specs currently don't include the crit debuff).

Neither do I know how crit would become any better through the changes, if they don't get to change crit to go to the pet in 3.3 it's most likely to become worse than now. The only time you use Ice Lance is when you have FoF up, meaning you have +50% crit anyway. Even in my Arcane gear I'd currently be at >50% crit with Frost spells raidbuffed, so I'm at 100% for IL anyway, making any further crit useless for IL and the 3.3 proc.

A few fire pvp buffs in the latest build and a change to Frozen Core

I like the change to Frozen Core and the change to Water Elemental will wash away most of our need to spend two global cooldowns to dismiss our water elemental and to resummon it once it's mana is out.

Very nice changes to those fire talents to possibly make them worth taking in AoE situations. Might change the build and rotation for Heroic Anub25 a bit to have those flamestrikes from firestarter be FREE as well as instant, and blast wave/dragon's breath be significantly cheaper. The biggest argument against firestarter at the moment for that fight is mana drain. Of course, these are changes for 3.3, so ToC won't matter anymore, but if any ICC encounters rely on AoE the way Anub does, it'll be good to take these into consideration.

Nobody ever answered this, but no, it's not. It's not based on spell travel time or any intended game mechanic, but on buff fade lag time. That's what makes the ghost charge a sort of gray area, difficult for people with certain types of lag (especially variable latencies) to take advantage of, and NOT the same at all as shatter combos. You're only able to get the ghost charge because the buff doesn't fade exactly instantly when the second frostbolt leaves your hands, but a few dozen milliseconds later--late enough for the third cast in most cases, but nowhere near long enough for a spell full second later to gain any benefit, regardless of what has or hasn't reached the target.

Mana is not an issue as frostfire, the problem is and was that the flamestrike target area is very small and the fact that you need to target it after an instant cast so you waste time. The changes are more PvP related then PvE.

I've done some analysis of optimal frost cycles in 3.3. For now I'm just assuming infinite duration on Frozen Core effect. If someone has information on that let me know. This was run with full dps talents and Frostbolt and Ice Lance glyph, no fire talents, full spell impact.

Cycle Code Legend: 0 = FrB, 1 = IL, 2 = FB
State Descriptions: BFx+-,FOFy+-(z),FC
x = remaining time on Brain Freeze
+ = Brain Freeze proc visible
- = Brain Freeze proc not visible
y = visible count on Fingers of Frost
+ = ghost Fingers of Frost charge for instant available
- = ghost Fingers of Frost charge for instant not available
z = actual count on Fingers of Frost
FC = Frozen Core effect active

0: BF0-,FOF0-(0)
1: BF0-,FOF0-(0),FC
2: BF13.84889+,FOF0-(0)
3: BF0.03562331+,FOF0-(0),FC
4: BF0-,FOF1+(1),FC
5: BF0-,FOF0+(0)
6: BF13.84889+,FOF1+(1)
7: BF12.15053+,FOF0+(0)
8: BF0.09221625+,FOF1+(1),FC
9: BF0-,FOF1+(1)

0022010100: 5788.341 dps, 34.29562 mps

There is only one optimal cycle in this situation, and this is using Brain Freeze Fireball when you don't have Fingers of Frost active and using Ice Lance on ghost charges.

Disabling use of ghost charges we get:

0: BF0-,FOF0-(0)
1: BF0-,FOF0-(0),FC
2: BF13.84889+,FOF0-(0)
3: BF0.03562331+,FOF0-(0),FC
4: BF0-,FOF1-(1),FC
5: BF0-,FOF1-(1)
6: BF13.84889+,FOF1-(1)
7: BF13.84889+,FOF1-(1),FC

00220110: 5665.476 dps, 29.94508 mps

That is Brain Freeze Fireball when you don't have Fingers of Frost active and using Ice Lance on 2nd FOF charge.

I'll be checking in the changes so others can test out other scenarios. I'll probably add 2T10 support sometime over the weekend.

Last time I checked Frozen Core buff had a 10 second duration. This was 2 patches ago on the PTR.

You cannot cast any spells during the GCD. Bringing up a targeting circle for a spell requires "casting" it(but not finalizing the target), so no, you can't. This means that after any sort of instant spell, retargeting and beginning a Blizzard/Flamestrike cast wastes time. It's one of the things that makes Firestarter very awkward to use.

I just assumed you could as you can aim your next target area for blizzard/flamestrike while channeling a spell but you are indeed correct. Although you can still pseudo-aim it during the gcd as the center of the target area will be your cursor's position though that requires hotkey + click instead of just click, I don't know how significant a time less that would be though.

It's not that hard to hover your mouse over there area you wish your Flamestrike targetting circle to appear over.

On the AoE perspective, but speaking from an arcane point of view (with frost warding rather than ice shards) I've been noticing recently that Arcane Explosion is outstripping Blizzard spam, or Flamestrike opener / Blizzard spam. Whether it's because Illustration stacks with Arcane Explosion while Blizzard doesn't, AE can be used on the move to follow mobs or that AE has threat reduction and can generally be used earlier, I don't know, but for me personally I've been finding Arcane Explosion to be a more effective AoE. Both without and without cooldowns (ie arcane power / icy veins) I've been showing both higher dps and higher damage lately by spamming Arcane Explosion.

Has anyone else had similar experiences? I'm finding it hard to believe Illustration alone can account for this given Arcane Explosion has long been regarded the weakest AoE we can throw about.

I find it hard to believe that AE could outdps a skilled mage who can get LB up on targets (assuming explosion will go off), and then do rank 9/8 of FS->blizz. If you're talking just AE vs Blizz then maybe so? I don't have the numbers to run that though.

And while you could say that AE can start up earlier because of threat reduction, depending on how you and your guild plays, threat shouldn't ever be too much of an issue.

For example on Anub, I LB the mobs as they're coming in, and even start casting my flamestrike when I see they're about 10yds from being grouped up, then do rank 8, blizz spam, etc. Really can't see AE outdps'ing all of that assuming AE will start at the same time as my FS goes off, and I don't usually pull off the tank unless it's during heroism (and even then everyone is dps'ing faster so they die before my threat is a problem).

Does someone have strict numbers on different AoE rotations?