I don't quite understand your comment. Are you implying that adding 'only' 10% to Frost Mages DPS via this single talent point spell - is something to shake a stick at?

Considering that's the best case scenario by using a very high damage estimate of 2.5 times FB damage, not including latency and using low enough haste not to be GCD haste capped and the fact that Frost is lacking way more DPS to be competitive than the remaining increase and the water elemental change the whole spec still seems lacking.
The 10% is also only relative to the damage the Mage himself does, not including the water elemental damage.

Comparing single talent points isn't really a good idea since Blizzard balances specs as a whole, not each talent point on its own.

Yes but put the current situation into context. Do you think Blizzard has a "Ok we need to up Frost DPS, Increase its burst potential, fix AOE, de-bloat the tree" list they are rushing to get through simultaneously?

Based on recent changes, we're getting methodical tweaking to things intended to push up the PVE DPS side of frost. Things like the aforementioned points can come afterwards. They are the ultimate goals of the spec, but for now we just really want to see it do comparable DPS. After we reach that point, then we can start worrying about things like AOE and Burst Potential (Which are still valid things to worry about, but just need to be prioritised).

Frankly, i'm surprised that we're actually getting any of these changes - let alone moving on to thinking about a bigger list of things the spec needs to address.

For the record, the estimate is about 2.0 times Frostbolt damage, not 2.5 times. And here's the math, using raid-buffed stats from the SimulationCraft Frost profile (which was generated originally from Rawr):

Frostbolt
Crit multiplier: 1 + 0.5083 * 1.09 = 1.554047
Avg damage: 4185.5 * 1.3743324 * 1.554047 ~= 8939.30
DPS: 8939.30 / 1.762 = 5073.38

Frostbolt (Shatter)
Crit multiplier: 1 + 1.0 * 1.09 = 2.09
Avg damage: 4185.5 * 1.3743324 * 2.09 ~= 12022.24
DPS: 12022.24 / 1.762 = 6823.06

Deep Freeze (Shatter)
Multipliers: 1.06 * 1.05 = 1.113
Crit multiplier: 1 + 0.9783 * 1.09 = 2.066347
Base damage: (2636 + 2923) / 2 = 2779.5
Gear damage: 7.5 / 3.5 * 3665 = 7854
Avg damage: (2636 + 7854) * 1.113 * 2.066347 ~= 24125.37
DPS: 24125.37 / 1.149 = 20996.84

You're going to cast it about once every 42 seconds.

Frostbolt Spam
- 1.85 Shatter Frostbolts (0.15 FoF charges lost to overlapping procs)
- 4.82 regular Frostbolts
Total damage: 1.85 * 12022.24 + 4.82 * 8939.30 ~= 65328.57
Total time: 6.67 * 1.762 = 11.75254
DPS: 65328.57 / 11.75254 = 5558.68

Using Deep Freeze
- ~24 Frostbolts per Deep Freeze
- 6.66 FoF Frostbolts
- 17.34 regular Frostbolts
- 1 Deep Freeze
Total damage: 6.66 * 12022.24 + 17.34 * 8939.30 + 24125.37 = 259200.95
Total time: 24 * 1.762 + 1.149 = 43.44
DPS: 259200.95 / 43.44 = 5966.87

Increase: 5966.87 / 5558.68 = 1.0734 = 7.34%

Water Elemental is about 12% of damage, so actual increase closer to 0.88 * 7.34 ~= 6.46%

Now, this is assuming that Shatter and Ice Shards are intended to work on Deep Freeze, which they currently do not on the PTR. If they're not intended to work with Deep Freeze, then we have a problem.

The problem is that many players think Frost should be behind other specs' DPS because of its survivability. However I just can't see that bit of survivability make up the loss of sustained single target DPS at the same time as giving up a lot of AoE DPS as well as burst DPS.

Additionally, looking at the above post, I even forgot that you need a FoF proc to use DF, meaning my estimate is way too high.

At Lhivera:
What do you mean by "assuming that Shatter and Ice Lance are intended to work on Deep Freeze"?
Going by patch notes and implementation Deep Freeze is a pure single target damage spell against raid bosses meaning you don't get its debuff's bonuses.

I meant "Shatter and Ice Shards," and have corrected the post. Currently, the damage from Deep Freeze does not seem to be gaining the +50% crit chance from Shatter or the +100% crit damage bonus from Ice Shards.

Also, Haste is really not an issue here. The spell averages twice as much damage as a Shattered Frostbolt, in 65% of the cast time, giving it about triple the damage per execute time of Frostbolt. If you had 130% haste, reducing them both to the 1-second GCD floor, it would still have double the damage per execute time of Frostbolt. That cuts into the advantage for those portions of the fight when you're over 50% haste, but it's still a very solid increase.

And did anyone already observed how will this translate for Fingers of Frost?
Usually we use the ghost charge for Ice Lance, but now replacing FoF with Deep Freeze, since DP is instant and has no travel time, what happens with the second frostbolt? Does it still benefit from FoF?
Can anyone actually try this on PTR since I'm at work atm.

You are correct but the fix for this is coming in 4.0. GC stated previously (can't find the reference at the moment) that he is currently stuck with specs like Frost and Subtlety. If you have a spec that has tons of utility but poor damage, no one picks it for raiding. If you give that spec equal damage, then EVERYONE picks it for raiding. His targeted solution is to make sure that in 4.0 no one spec of a pure dps class has all the utility. This is going to be accomplished by gutting the trees as they have already planned. Remember, all the simple +dam, +hit, etc talents are going, so all that is left are the fun, utility talents.

In the context of 3.3, frost mages getting within 5% of fire/arcane is totally acceptable. In the context of 4.0, totally unacceptable. The mindset has to change for the long suffering frost mages. If you accept -5% damage you will get -5% damage. I don't know about you, but in my raids a mage doing 5% less because of spec would be asked to respec or lose their raid spot. And I say this as a raid leader who desperately wants to raid as frost.

Having only Deep Freeze not benefit from Shatter wouldn't really prop up the value of crit rating. You'd still consider 50% crit on Frostbolt (97% crit on a Shattered Deep Freeze) to be the soft cap.

Yes, but I don't think to an extraordinary degree. Nothing terribly significantly more than a Fire Mage faces with Hot Streak.

Well, it is 5% less damage theoretically, it could be more or less in practice. 5% puts it in the kind of range where you're likely to find one spec having an advantage in some fights and the other specs having the advantage in other fights, at which point asking someone to respec because of what the spreadsheet says may be excessive. Now, if the spreadsheet says "you'll be 20% behind" that's a whole different story.

But I don't lead a 25, so my opinion may be invalid.

No it's not?

It's a big buff any way you slice it. Scoff at "only 10%" all you want, differences of this degree are the REASON people don't want to play frost right now.

If it's within 5% then that's more than enough to justify use in progression raiding. Top progression raids have had ice mages all through TBC and Classic simply for the huge leeway the spec grants in fucking things up. Unless you're a perfect player (you're not), it's a non-trivial benefit that neither for or arcane bring. Frost's survivability is demonstrably superior, it just needs a bump in damage so it can still contribute meaningfully to kills. It does not need to be exactly on par with the other specs.

Magegraf updated to have DF but not frozen core:

http://www.magegraf.com/3.3/index.ph...6cb7e5ffcd7490

Correct. And in Cataclysm/4.0 frost (if done correctly) may be the one spec that does more damage at particular tier levels.

This is correct in 3.3. In 4.0 the idea is that frost doesn't have any more special utility than fire or arcane. For example, frost may have ice barrier for increased survivability. Fire may come with a talent that decreases spell damage taken that provides a similar benefit via a different mechanism. And so on. Yes, if the current paradigm that fire is pure damage and frost is survivability stays, then frost won't have a place in high-end raiding. But that is not the goal as GC has stated it.

Anyway, this is has gotten off topic from a discussion of the 3.3 PTR changes. The current PTR changes need more testing and questions about shatter and DF need to be answered by Blizzard. Overall the announced and suggested changes are very encouraging.

This is why I was thinking that they would have to add a glyph for Deep Freeze along the lines of "Your Deep Freeze cooldown is reset if it deals damage". Then the damage does not need to be obscene in order to boost frost DPS in any meaningful way. Not to mention it would just be cumbersome to have it on a 30 second cooldown AND be dependent on FoF procs. Of course, this would make Ice Lance useless especially if Frozen Core is reverted, since you'd want to be using as many FoFs as possible on Deep Freeze.

You cannot save FoF procs for the Deep Freeze cooldown, unless you stop casting Frostbolt. This would have no impact on how useful Ice Lance is, since you would see many FoF procs while DF was still on cooldown.

I think what they're looking to do here is add one spell to rotation in a shatter-combo-like fashion. They tried the Frozen Core change first because (a) it's the most like regular shatter combos, and (b) it doesn't create a PvE/PvP split in a spell. They're moving on to Deep Freeze and will probably remove Frozen Core because (a) Frozen Core just doesn't work very well, and (b) it's pretty easy (as Vontre demonstrated on the official forums) to come up with a simple and lore-friendly explanation for the different behavior.

Thanks for posting numbers, everyone; so far, the widest base damage range I can figure (this is factoring out the piercing ice and arctic winds multipliers) is 2368 - 2635, and everyone's data seems to be supporting a 7.5 / 3.5 coefficient (not the 8.3 / 3.5 I mistakenly posted last night that didn't account for removing those multipliers).

I've sent code for adding Deep Freeze to SimulationCraft to dedmonwakeen. Hopefully I did it right.

As far as preparing for your next AE while the present is casting (or channeling) you are able to do so however you need to wait till half of your channeled spell has completed. IE I usually do PoM Flamestrike(rank9) and when it half or more finished I press the button for Flamestrike (rank8) and get my circle ready to go where I want it then just before it goes off due to latency I trigger that and halfway through its cast I will get Blizzard ready and repeat for another blizzard if neccesary. Not sure why it works this way /shrug

Sorry I am late, had to wait a day to register =/

Also of the opinion that with these changes going through we will see a lot more AE scenarios in Icecrown but I imagine we will have to wait and see.

The one exception to that is ghost charge ice lances; though I know it's not the recommended move (and I know some people are conceptually against them working at all) more than a few people seem to make ghost charge ILs part of their normal routine. Deep Freeze will eat some ghost charge opportunities, since as an instant cast spell it won't be possible to get a ghost charge IL off after a Deep Freeze.

Still, at most it would be missing out on one ghost charge every 30 seconds, and since they're not terribly beneficial to begin with it shouldn't matter.

Another issue would be whether Deep Freeze should be on the second FoF charge or on the ghost charge (I assume it will never be on the first FoF since that would usually require interrupting a FrB cast). It seems likely that trying for a ghost charge DF would be Bad, both because you want to get your DF cooldowns rolling ASAP, and because missing the ghost charge due to latency would be terrible.

What you're waiting for in that situation is not half the channel, specifically, but rather the Global Cooldown triggered by the cast. You may not set up a new targetting circle while on the global cooldown.

While I'm on the subject of AoE, the new AoE cap is something worth discussing. The old AoE cap rarely capped it at or before 10 mobs. For many spells, it was quite a bit more. Making it strictly limited to 10 is going to have a serious impact on mage AoE rotations in certain situations. As a current example, let's take the whelps on Onxyia. Even if you're not doing the achievement, you get something like 30 whelps in phase 2, 15 from each side (roughly). This means that, assuming you tank them together in the middle, you are capped at 1/3 of the total mobs you are AoEing.

In that situation, the value of Flamestrike skyrockets because it's Ground Dot Ticks are NOT affected by the AoE cap.

At a certain level of mobs, a pure flamestrike (alternating ranks so as to keep multiple DoTS on the ground) probably becomes superior to any other rotation. I have no idea how many mobs this is because there aren't many particularly good tools for modeling AoE rotations (though I think Kavan said he was working on that for Rawr). This state of affairs is not *entirely* new, but being at or above the cap on your number of targets is sure to happen much more frequently now that the cap of 10 is going to be enforced (some spells were previously capped so high that even with BiS gear you could hit 15-20 mobs before reaching the cap) so now it might be worthwhile to actually take the time to get a proper sense of how the value of Flamestrike and its lesser ranks increases as you pass the cap, and at what number of mobs each rank passes Blizzard in terms of effectiveness.

Just some random semi-related food for thought.

I have about 300 ms latency to the PTR server and I was testing on Dr. Boom, so I was near maximum range. There may have also been some server lag at the time. In any case, because FoF procs on spell impact and because I was getting the proc notifications with significant latency, my deep freezes were actually usually on the ghost charge. The time margin for switching from a frostbolt cast to deep freeze wasn't enough, because the FoF proc showed up only a tiny fraction of a second before I could buffer in the frostbolt.

In a stable situation (constant range, constant latency, decent frame rates), the timing becomes reasonably predictable. On Dr. Boom on the PTR, I knew I had to buffer in the deep freeze as soon as the buff showed up on Power Auras or it would be too late. It's likely that with over 300 ms latency, the only way to get deep freezes off at absolute max range is to use the ghost charges.

Similarly, my brain freeze fireballs were frequently firing off just after the FoF proc showed up, so I ended up using several FoF procs on simple frostbolt+fireball combos, which isn't a disaster, but isn't all that great either. It's probably something that the simulators and analysis tools do not model, because they usually assume perfect play.

I think FoF procs should go back to happening on spell cast, so that players with high latency have a slightly easier time reacting to the procs. When things get laggy, it becomes much harder to determine which spell procced FoF. Because FoF doesn't show the number of remaining charges on the buff, you need to be aware of the latency in order to properly use ghost charges.

A tester on the general forum provided a list of a hundred trials of Deep Freeze against Dr. Boom. The results correlated very nicely with the results I'd put together from several smaller samples both here and on the official forums, matching the max base number I had exactly and missing the min base number by one (on the high side). So I feel pretty confident that a base range of 2368 - 2635 and a coefficient of 7.5 / 3.5 is correct.

I just noticed today, the patch notes for the latest build on the PTR. The Frozen Core deal now affects (grants the 1 second cast time reduction) to Frostfire Bolt as well, which I hadn't heard of before. Is this new or am I just slow? How do the numbers add up for this, since one wouldn't have to deal with GCD cap with FFB?