Muphrid

I'm going to assume that AToI can refresh itself, which makes the math particularly hairy--in short, by increasing your own haste, it increases the effective uptime, as you have more chances to proc it. However, I can use some of my old work on Quag's Eye modeling to try to figure this one out.

Assume that, with AToI equipped, one has an average casting time t (which we don't know, because again, having AToI increases your casting speed, which increases AToI's uptime, which increases...you get the point). Given this, however, we can say that, for any given cast, there are n = ceil(5.5/t) casts that could have procced AToI (on average) in the last 5.5 seconds. Thus, the chance that AToI is active, for a crit chance c, is p = 1-(1-1/2*c)^n--1 minus the chance that every trial or cast failed to proc.

From here, we calculate the average haste rating as 145*p and can easily convert this to equivalent +damage or real added DPS or whatever.

The task, then, becomes figuring out the average casting time t. I imagine this could actually be done in an iterative process: beginning with the non-proc average casting time, we can calculate typical uptime, then recalculate with the new uptime, and so on. This should produce a converging underestimate. Indeed, due to the discrete nature of the problem, the average cast time converges quite rapidly...

For example, consider a 2.5 second cast Frostbolt with a 30% crit rate.

t1 = 2.5
n1 = ceil(5.5/2.5) = 3
p1 = 1-(1-(1/2*.3))^3 = .3859

t2 = 2.5*(.3859/(1+145/1576)+(1-.3859)) = 2.419
n2 = ceil(5.5/2.419) = 3
p2 = (1-(1-(1/2*.3))^3 = .3859

t3 = 2.5*(.3859/(1+145/1576)+(1-.3859)) = 2.419
...

Such a calculation could be carried out for any given initial casting time, bearing in mind any initial haste rating, of course.

Also, this was rather whipped up on the fly; I wouldn't be surprised if there were some aspect of the model that I've neglected.

 

Vhad

Yes, but that is if you consider the 5 aoe targets. Remove that and you'll find what I got too.

What!?

 

Roywyn

You forgot to include the proc duration. It lasts 5 seconds, which is enough for 3 Frostbolts or 2 Fireballs. Maybe 3 Fireballs if you have a lot of passive haste and very low latency/cast delay.

If it affects 3 spells, and has a 20% proc chance per cast for you, you get ~50% uptime, which is around ~90 damage average. Pretty close to what you got I'd say

 

Vhad

Yeah, I knew something wasn't entirely true. I guess another way of looking at it would be a 20% chance to proc 181 dmg for 5 seconds, something that's alot easier to not be partial about.

What!?

 

Roywyn

It's not for 5 seconds, that's the whole thing about haste procs.
For a spell, the cast time is determined at the start, while the damage is determined at the end.

With a 5 second damage proc, you get the bonus damage on 1 fireball or 2 frostbolts.
With a 5 second haste proc, you get get the bonus haste for 2 fireballs or 3 frostbolts.

So, the haste proc will vitually last ~7 seconds for frostbolts, which makes the trinket good for frost, or fire with some passive haste and low latency.

 

Vhad

That's a good way of looking at it, so to revise it would be: 20% chance to proc 181 spell dmg for 2 fireballs, again much easier to accept as good.

What!?

 

Cortney

http://www.wowhead.com/?talent=oroVZxgz0rcf0gRxVhb0o

This is what our class lead has posted for fire to run come the patch. I am writing a comparison up between that build, 2/47/11 +1, and 10/47/3 +1. I think I can explain the 1% damage loss, 1% crit loss and mana reduction to fire, 2% crit loss to scorch and fire blast, and the mana lost due to crit.

Could someone help me understand how 2/3 scorch will affect the fight or point me in the right direction to research?

 

Copernicus

Actual math will be tough for that, but I'd stress the important of keeping a Fire Vulnerability stack going and the downsides of a 7 Fireball to 1 Scorch ratio vs an 8 fireball to 1 scorch ratio. It'd really hurt for trash and any fights with lengthy transitions or time spent changing targets.

I think the larger thing would be to point out how weak Clearcasting is. The new mana gems give about as much mana as Clearcasting gives. If the mage is using gems now, they can switch over to a 2/47/11 build without losing anything. If they aren't using mana gems, than clearcasting wasn't all that important in the first place.


It's also possible to get a 7/43/11 build (ditching Dragon's Breath, Blast Wave, and Incinerate... I used the garbage point in Improved Flamestrike). I don't know what the next points to shave off of Fire would be if mana problems were still an issue or important- probably Playing with Fire, as it's the next weakest talent.

 

Doroteasenjk

It means that it will be harder to guarantee reapplying Improved Scorch debuff to the boss. With a one-third chance to not apply it, you have to start refreshing the debuff earlier and monitor the Resist messages. Since you have an 11% chance to resist two of them in a row and have the stack fall off, you need to start casting Scorch at the 25 second mark in your rotation and keep trying until it is applied.

This decreases benefits from haste, since Scorch is at the GCD already, and decreases benefits from Fireballs about one-third of the time. At a guess, you would see a 1% - 1.2% decrease in damage compared to the current optimal rotation.

 

Roywyn

What does your 10/40/11 build achieve?
You get Icy Veins. Good. But you loose essential fire talents. Not good.

Icy veins is a 2-2.5% gain, you lose 1% damage and 1% crit and 1% mana efficiency (PWF/Pyro). I'd rather drop the second point of Incinerate and go for full 3/3 Imp. Scorch.
It's pretty hard to model, but it means that a 15% damage buff will randomly fall off, or you spend too much resources keeping it up.

In any build with Clearcasting and Icy Veins, you lose more from the points in fire than you'll gain from Icy Veins. you'll always lose 1 point in PWF and 1 in PWF or Pyromaniac, so you lose what you gain.

 

Cortney

You really hadn't read what I wrote, and I might have been a bit unclear as well.
My class leader put that build up for are fire mages to run come the patch. When I posted earlier I understood how to explain everything why not to use that build except for knowing exactly how 2/3 scorch worked.

 

Pintofbrew

If only how 2/3 scorch works is your concern, suffice to say it's an exquisitely bad idea. While it's debatable whether 3/5 Winter's Chill will work well enough, imp. Scorch mandates you use a sub-par spell instead of your normal nuke. Given that you'll drop anything from 1-1.5% by having to increase the frequency of scorch and still risk having it drop off, netting you at least 3sec worth of -15% damage and at least 7.5sec re-scorch time, the potential for loss is much greater than the gain of IV, which can't realistically be assumed to be more than 2.5-3% at the best of scenaria.

Prior research has showed that Incinerate is not crucial to overall DPS, given it'll only affect (ideally) 1/16th of the DPS cycle and the worst part of it at that. your Wowhead link is broken for me, and I can't see what you propose but I assume it's a 10/40/11 as that's what Roywyn commented on.

Commenting on 10/40/11? Don't do it. Commenting on 2/3 scorch? Don't do that either.

 

Roywyn

"It's pretty hard to model, but it means that a 15% damage buff will randomly fall off, or you spend too much resources keeping it up."

That was my reply to it. I know, not much you wouldn't have known before I guess. As for "exactly", I can try to make a model at least.
Assume 1.5k fireball, 1.1k Scorch DPS fully debuffed, and just a single mage.

If you cast a scorch when the debuff runs out and that scorch fails to proc, you have to rescorch. For 5 stacks, you need 5 procs at 2/3 proc chance, i.e. 7.5 casts and you have 2 stack of scorchs up on average.
Scorch at 2 stacks is 1.1k*(1.06)/(1.15) = 1.014DPS, casting that 7.5 times at 1.5s cast time each is a DPS loss of 7.5*1.5*(1.500-1.014) ~= 5.500 damage lost.

So, everytime the debuff falls off, you lose 5.5k damage.

If you only scorch at the end of the debuff, you have a 1/3 chance to have it fall off, i.e. a 1/3 chance that you lose 5.5k damage every ~30 seconds, a 60 DPS loss (5.500*/30s*1/3), that's (4%).
If you rescorch at 24 seconds, then cast a Fireball at 25.5, you can check if your scorch failed at the start of the fireball cast, and if it failed, rescorch at 28.5 to get a second chance at keeping the debuff up when 30s ticks.
That means you have a 1/3*1/3 chance to let it fall off, which would then be a 20 DPS loss.
But you'd also rescorch more often, every 24 instead of 28.5 seconds, which means one more scorch every 152 seconds, which is a 1.5s*(1.500-1.100)/152 ~= 4 DPS. You also have an 1/3 chance that the first scorch fails and you have to cast a new scorch at 28.5s, which means an additional scorch every 3*28.5s = 85.5s, which makes you lose another 400*1.5s/85.5s = 7 DPS.
So, that's a ~30 DPS loss total, about 2%.

From these numbers, I'd estimate that if you try to rescorch even earlier, you'd lose more damage from the early scorches than you'd gain from prevented stack drops.

This is assuming 1 mage and perfect reaction on stack drops. In reality, you'd have at least one fireball cast per mage hit without a debuff, so a lost stack is additional 15%*3s*1.500DPS = 675 damage loss per mage.
Add reaction time and the fact that you can't easily cooridnate scorch stacks that belong to another mage, and you easily reach 11k damage lost for each stack that drops.

So, you'll lose ~8% without pre-scorching, or ~4% with pre-scorching.

 

Praen

Hi...I dont know whether this's the right thread to ask this question or not but i wonder if i there's any theorycraft about mage's water elemental which i can find on this web site or at another. Because i had an issue last nite at A'lar fight, my water elemental spells' were resisted by almost 80% It could barely damage A'lar. So how can i learn hit/resist chances of mage pet?

 

xiaoxin21

For a 2.3.2 mage build,

I am more leaning towards a

3/47/11 or a 4/46/11( -1 in incineration) build than the suggested 0/50/11 build suggested. Taking 2 subtlety and 1 point or 2 points in focus will require you to aoe with AE with considerably less threat and the 1 or 2 points in focus will help in boss fights where CS is used without swapping in more +hit gear(like council).

Will the buff to flamestrike be significant enough to spec 0/50/11?

 

Dustwhisper

I'll gladly take the two points in arcane for IAE in Hyjal etc, never raid with a palidantank anyways. In most cases I feel confident I can output more dps with IAE than flamestrike, especially with erratic mob-gathering.

 

Pintofbrew

Praen: Your water elemental will have the usual 17% miss versus bosses. This issue of pet's non-scaling with hit, haste, crit and resilience has been noted by the warlocks and seems to be in que to be resolved at WOTLC. Being grouped with an elemental shaman will help, as the poor thing will benefit from ToW and lower it's miss to 14%.

Your high miss rate is odd; perhaps a combination of fluke-statistics with premature death. Particularly on Al'ar who shouldn't really have any frost resistance... Odd.

The WE, I understand, gains 25% of your spellpower and an un-defined amount of life/mana from your sta/int but none of your hit or crit. It is affected by Winter's Chill. It will also be affected by any temporary buffs that are applied after it's been summoned, ie. Bloodlust, Drums of War/Battle.

When the next patch is out (jesus where are you already IV?) it'll probably be optimal if you could summon him, make him attack target (you can make a macro for this effect) and when the GCD of the summon is over, pop a macro with Trinket(+manstone for Serpent Coil Braid)/IV/Coldnap/Frostbolt. This will compact your cooldowns and make you get the maximum amount of love from your trink as IV will increase it's value by 20% and WE by 25%. As soon as IV is over, re-IV and as soon as the WE is out, re-WE. With any luck (and fights over 8m) you'll get two coldsnaps in and plenty elementals.

Unfortunately, high pet-resist is something we have to live with. Just like there was no % hit before Zul Grub so long ago, now there is none for pets. Watch this space, I guess.

Xiaoxin21: It depends heavily on what AoE strategy your guild uses; guilds with a palatank seem to be very happy spamming flamestrike but others seem to think AE is the way to go. It's very guild-centric. As for -1 incineration, it'll make so little difference it won't matter, so don't worry about considering it a direct-damage drawback.

 

Rugrud

Just a question about +damage trinkets and WE. We know that WE only take Bloodlust for instance if it is casted after the summoning of the WE. But for + damage, which value is taken? Il it the value at the time the WE is summoned? checked at each cast (at the beginning of the cast, at the end?). I'm wondering if i have to pop the trinket in the same macro than the WE, or if i summon the WE, trinket the last 20 seconds of it and still gain the same damage?

 

Pintofbrew

I'm almost adamant the WE will check your spellpower on initiation of cast, just like you do. Perhaps a warlock is more suited to check this.

The evidence I offer is the much bigger average waterbolt I observe when popping Icon and Braid while it's out. I never use them before he's out already, as that is a waste of 1.5sec worth of frostbolt time under trinkets and also because sometimes he delays picking up the "petattack" command or getting in position (I'm a big fan of pulling him out on the move, particularly on fights where proximity might get him killed like Eagle Boss or Vashj P1 or P3).

 

Kewangeder

The only remote benefit I can think of to a 2/3 scorch, that a player might be inclined to think of when considering it, is a mage trying to bank on the notion that multiple fire mages in a raid will tend to do an 8xFB,1xS rotation, making all but one of the scorches redundant.

But suppose there are three mages. If they're all 2/3 scorch, then for the debuff to drop, all three have to fail to renew, which happens 1/9 of the time. For 2 mages, it's 1/4. For that rare raid with 4 mages, it's 1/16.

This assumes there is no prior agreement between the mages to have only one of them be the designated debuff maintainer. If there is, then all but one of the mages are simply spamming FB. Indeed, if they're all disciplined to agree to take 2/3 scorch, then they're all disciplined enough to have a scorch policy (say, whoever lays the first scorch down will be the maintainer). And in that case, the debuff drops only 1/99 of the time, assuming the mages are hitcapped.

So it looks quite bad for 2/3 scorch on that front, too. Given this, I'm inclined to believe a 2/3 scorch mage is sort of the analog to an arcane/frost spec who's trying to piggyback on the deep frost mage with WC.

 

Kir

I vaguely remember reading about frostbite being buggy after 2.3 hit, but I wasn't frost at the time, and didn't pay much attention.
Our guild is now working on Illidan, so I've specced frost. I was solo farming Black Morass while on standby for Hyjal last night, and frostbite did not proc ONCE in over 250 frostbolts on those mobs. They weren't freeze immune, as I was frost novaing them just fine, and did not see 'immune' messages anyway. After that, I realized I hadn't seen it proc at all when I was solo farming elsewhere, just hadn't paid attention.

Frostbite procced right away on the first blizzard I did (just to test), and it was proccing in a duel I did right outside the zone line. I asked the other mages, and they said they weren't noticing an issue with it.

Anyone have any info on this? I have a feeling it's related to how many talent points you have in frostbite/permafrost/imp blizzard in some way. I have 1/3 imp blizzard and 3/3 in both permafrost and frostbite.

 

Cwealm

Sorry if I missed something, but is there a change to flamestrike in the works?

 

Plankel

As far as I understand it the bug involves frostbite not proccing at all from frostbolt, but I didnt bother to research it completely, I do know it is confirmed though and should be fixed soon.

WoW Forums -> [Bug] Frostbold chill not procing frostbite


Not that I know of, unless I missed something I think you are confused by