This confirms 4 years of observation whilst playing, plus a deafening silence on all other forums, and as the data itself is coming out of WoW DBC files, I'm inclined to believe it.

After a few hours at a training dummy, it turns out I was incorrect, the combat log reports identical times for the missile impacts. However, I would often witness a visual lag in their artwork, and one missile would reach the target slower than the other and which missile it was (Fireball of Pyroblast) seemed to be random. As far as Ignite munching is concerned, range to the boss should not matter, and I retract my previous statements. Thanks for the DBC research.

Is it worth keeping up the 3x ffb debuff through hardcasting as a frost spec to maximize dps? I hope the question can be understood.

Please delete

Mage friends, I come to you hat in hand. I seem to be doing something silly irt my fire rotation. I've forged and geared as simc and rawr suggested, however, I'm having a heck of a time getting good combusts. More often than not, when my combustion is off cd, I either have no ignite or a shoddy one up, and no pyroblast dot up either. I've had a few ignite-less combusts this week because I was waiting until the last possible second and clipped too late. I know I don't want to be doing that.

Any input / advice is, as always, appreciated.

WoL log and armory link! Note on oddness in WoL--near the end of Hagara attempts, I went arcane to try to up my dps a little.

I was wondering if anyone would be able to take a look at this log and help me out at all. I'm pretty new to playing mage, so any hints or tips would be great.

Log: World of Logs - Real Time Raid Analysis
Armory: Wortortle @ Blackhand - Game - World of Warcraft

Thanks for any replies.

No, it is almost never worth cancelling a spell for another once you have started casting. Just starting to cast is going to cost you at least a global cooldown which ends up being completely wasted time that deals no damage if you cancel the cast before it finishes. You could think of that time being added to the spell you are switching to and suddenly you are casting a 3 second 'instant' pyroblast. You are better off waiting until the end of the cast and then using hot streak while the fireball is in flight.

I'm a strong advocate that crit should be our best stat overall....


its the stat that is consistent in providing us high dps.

I understand haste caps - additional ticks on combustion/living bomb and pyro - I don't think it should be value as high as some make it out to be. Yes, if the stars align it should present you with amazing results - but more than often it falls short along with you DPS.


Crit never lets you down - the more the better - when you get a string of crits to build a powerful ignite... pure bliss.

Crit actually lets you down more often than any other stat. The stars need to align for crit, not haste or mastery. haste provides you with extra ticks regardless of procs, unless you count your t13 2pc and thats not rng off your own stats it's rng off the 2pc.

Crit has the widest range of RNG compared to any other stat for fire mages. Yes I agree it is generally our best stat, but it is far from consistant. If you look at the bell curve distribution for fire mages you can see how wide the gap is, 30ish-70ish. That gap is so wide because of our crit distribution. The higher your crit rating and the more other stats you give up for it, the more you're exposed to RNG. The last statement in your post is a perfect example, you need to be extremely RNG dependent to string together a powerful ignite.

What I'm more interested in is seeing the standard deviation and 95% CI of this bell curve. How do those values compare to an arcane spec'd toon of equal gear? Just saying the curve runs from 30-70K doesn't tell me anything about what to expect 95-99% of the time. Yes, RNG is causing the results at both extremes but the shape of the curve will tell you what to expect.

This is the bell curve for fire. Obviously RNG is causing the wide gap between min (37558) and max (54437), but what is more interesting, and unforunate, is how wide the range is towards the center of bell curve. You can't really expect any given value 95-99% of the time. With 95% confidence, your range is still something like 40k-50k, thats a pretty wide range, which is caused by the RNG inherent in Fire Mage play. The bell curve is also slightly skewed to the left, meaning you're probably going to see numbers like 40-45 more often than 46-50.

You can't delay ignite ticks, you can only spread out ignite damage. Also delay between crit and application of ignite also plays a role. What would happen is as follows (based on my understanding of it). You're also ignoring flashburn, which I will also, but just wanted to point it out.

EXAMPLE 1:

0:00.0 Crit for 100k (ignite bank = 40k, 2 ticks)
0:00.5 Ignite applied (ignite bank = 40k, 2 ticks)
0:01.0 Crit for 100k (ignite bank = 80k, 3 ticks)
0:01.5 Ignite refreshed (ignite bank = 80k, 3 ticks)
0:02.5 Ignite ticks for 26667 (ignite bank = 53333, 2 ticks)
0:04.5 Ignite ticks for 26667 (ignite bank = 26666, 1 tick)
0:06.5 Ignite ticks for 26666 (ignite bank = 0)

EXAMPLE 2:

0:00.0 Crit for 100k (ignite Bank = 40k, 2 ticks)
0:00.5 Ignite applied (ignite Bank = 40k, 2 ticks)
0:02.5 Ignite ticks for 20k (ignite bank = 20k, 1 tick)
0:03.0 Crit for 100k (ignite Bank = 60k, 3 ticks)
0:03.5 Ignite refreshed (ignite Bank = 60k, 3 ticks)
0:04.5 Ignite ticks for 20k (ignite bank = 40k, 2 ticks)
0:06.5 Ignite ticks for 20k (ignite bank = 20k, 1 tick)
0:08.5 Ignite ticks for 20k (ignite bank = 0)

EXAMPLE 3:

0:00.0 Crit for 50k (ignite bank = 20k, 2 ticks)
0:00.5 Ignite applied (ignite bank = 20k, 2 ticks)
0:01.0 Crit for 50k (ignite bank = 40k, 3 ticks)
0:01.5 Ignite refreshed (ignite bank = 40k, 3 ticks)
0:02.0 Crit for 50k (ignite bank = 60k, 3 ticks)
0:02.5 Ignite refreshed (ignite bank = 60k, 3 ticks)
0:02.5 Ignite ticks for 20k (ignite bank = 40k, 2 ticks)
0:03.0 Crit for 50k (ignite bank = 60k, 3 ticks)
0:03.5 Ignite refreshed (ignite bank = 60k, 3 ticks)
0:04.0 Crit for 50k (ignite bank = 80k, 3 ticks)
0:04.5 Ignite refreshed (ignite bank = 80k, 3 ticks)
0:04.5 Ignite ticks for 26667 (ignite bank = 53333, 2 ticks)

Depending on exact timing the last example could also result in free ticks as follows (ignite bank comments here are a bit different because there's actually several different banks):

0:00.0 Crit for 50k
0:00.5 Ignite applied (ignite bank = 20k, 2 ticks)
0:01.0 Crit for 50k (ignite bank = 20k, 2 ticks)
0:01.5 Ignite refreshed (ignite bank = 40k, 3 ticks)
0:02.0 Crit for 50k (ignite bank = 40k, 3 ticks)
0:02.5 Ignite ticks for 13333 (ignite bank = 26667, 2 ticks)
0:02.5 Ignite refreshed (ignite bank = 60k, 3 ticks)
0:03.0 Crit for 50k (ignite bank = 60k, 3 ticks)
0:03.5 Ignite refreshed (ignite bank = 80k, 3 ticks)
0:04.0 Crit for 50k (ignite bank = 80k, 3 ticks)
0:04.5 Ignite ticks for 26667 (ignite bank = 53333k, 2 ticks)
0:04.5 Ignite refreshed (ignite bank = 100k, 3 ticks)

There's some Ignite behavior that I've seen that hasn't been captured by the examples in previous posts. When there is no Ignite on the target and you get a crit, the Ignite debuff duration is 4 seconds. If there already is an Ignite on the target, sometimes the duration can be up to 6 seconds. I wonder if, after a certain amount of time, a pending Ignite tick can't be modified, and the remaining Ignite bank gets the new crit's damage added to it, creating 2 new ticks.

00.0 Crit for 100k, apply a 40k Ignite (2x20k)
01.5 Crit for 100k, apply a 40k Ignite (not to first tick, but add to 2nd tick and roll 2 new ticks, 20k + 2x30k)
02.0 Ignite ticks for 20k (first tick of first crit)
04.0 Ignite ticks for 30k (2nd tick of first crit + damage from 2nd crit, divided into 2 ticks)
06.0 Ignite ticks for 30k

I don't think that is possible. If there's any refreshes then the last 3 ticks should always be the same unless the target died. If you have an actual log example that would really help.

I have a question about Combustion. Lets say I have Living Bomb up, I have just procced Pyroblast! from two Fireball crits, and landed said Pyroblast!. I look at Combustion Helper and its predicting a sizable ignite. So I spam my Combustion to toss it up for big damage.

Now, does Combustion compute its damage from all dots applied to my target even if I haven't given any of them time to tick or does it only compute damage for dots that have ticked at least once?

Combustion computes it's total damage from all your DoTs currently applied to the target, regardless of whether they've had an opportunity to tick yet.

Of course, it will not add in the DoT from a Pyroblast! that has not yet landed on the target. You do need to wait for the Pyroblast! to land before you activate your Combustion to get all three DoTs counted.

Since a lot of players are keeping fire for AOE fights, is it a benefit to optimize for Arcane and keep fire on a high Mastery setup?

I suspect that it may be beneficial to optimize for Arcane, but with a high crit/low haste setup instead of a high mastery/low haste setup.

Having an insanely short Arcane Power CD with 4-piece T13, I often find myself unable to conserve/regen all the way up to a high amount of mana to take full advantage of arcane's mastery before I need to pop Arcane Power. Because of this, I feel like the value of mastery may be a bit inferior to crit for a majority of the time. I haven't checked in a while, but I do remember that during T12 content, crit wasn't *too* far behind mastery.

Also, having a high amount of crit as opposed to mastery will be great whenever you need to respec to fire.

If you're not able to maintain high mana that is mainly a playstyle issue. There is nothing preventing you from using a mana neutral mix during AP.

It's definitely not a playstyle issue, since I do use a mana-neutral rotation when I can (even under AP).

I should have mentioned that I was playing Arcane for H Blackhorn (my guild was having issues killing the drakes before they flew away whenever we first started learning the fight), which requires several small, consecutive burn phases where there is little room for a mana-neutral rotation. I used a high crit/low haste build for my arcane spec during the time that my guild was first learning the fight, which allowed me to spec fire for other fights without sacrificing crit. It worked out well.

So, I guess it depends on what fight you're going arcane for. Reforging strategies can be different from mage to mage.

Kavan, what is your suggestion for fateswarm's question?

I don't have a suggestion because the question is very subjective. It all depends on which fights are more important for you. Optimize for whatever you have the most problems with.

Does anyone know if Cunning of the Cruel or the Rag trinket proc spellweave on Deathwing: Madness?

Also in regards to Fateswam's question, I'm inclined to agree with Kavan. You should optimize for the fight you are progressing on. For instance, I recently optimized for Arcane on Spine, but now that we will be working on Madness I am switching to a more Fire optimized spec.

Reforging all your gear only costs a hundred or so gold, it shouldn't be too big of a deal to change it once or twice a week.