It's really not that weird. It's the combination of foregoing Glyph of Life Tap as well as speccing Imp. Soul Leech. Since you don't have to Life Tap as often with the latter, you replace the Glyph (Imp is higher for my gear personally) and the result is that the value of spirit diminishes.

If it is any consolation, I think Shadowpriests have to weave dots and their AoE.

I agree that weaving Rain of Fire (RoF) when Empowered Imp procs, and Seed of Corruption (SoC) otherwise, would be best at high but attainable gear levels (specifically when the crit RoF dmg exceeds the crit SoC damage).

I'm still going to leave the mana considerations to another day, or if anyone else wants to pitch in.

To focus on a single AoE though, I'd like to calculate the effect of Empowered Imp on crit rate. As a very rough experiment, I was getting about 10 - 20% benefit from Simulationcraft for regular spells, and wasn't able to run Simcraft with just SoC/Rain of Fire.
Let C_p be the personal crit rate, and C_i be the imp crit rate.

Due to the cast time of the Imp Firebolt (2 seconds talented) and SoC/RoF (2 seconds), we have an upper bound of about 1 firebolt per SoC/RoF tick. This will occur regardless of the haste.
So, the expected new crit rate C_n will be the probability that you crit from your personal crit rate or the imp crits, which are independent events.

Recall that . We will apply this with A being the event you crit from your personal crit rate, and B being the event the pet crits.

Then,
.

As an example, with
(from the 2p t9), we have a new crit rate C_n of .58, or 58% as an approximate upper bound. Lancehead, in addition to the corrected Imp cast time, pointed out below that the imp doesn't always cast immediately, which lowers the value further. If anyone can help identify why Simcraft wouldn't run with only SoC/RoF as the spells, that would help, as well as anyone able to confirm at a target dummy.

Using a macro like
should keep the Imp casting and minimize lag effects, correct?

I've seen the debates on spell stone vs firestone and, like others, have always used the spell stone when afflic and the firestone destruction. So, after the servers went up today, I marched right down to the target dummy and tested both spec's out using the stones as above. I found with my afflic spec and spell stone, I did 4610 dps on the dummy then switched to destruction and forgot to switch my stone doing 4874 dps. I then noticed my mistake, switched to the fire stone and saw a dps decrease to 4698 dps. I used fel armor and the above stones as buffs and went about two to three minutes on each run. Has anyone better at testing this noticed this difference with the changes to conflag or is this just a fluke?

The cast time of Imp's Firebolt is 2.0 sec when talented and Imp casts it once every 2.1 - 2.2 seconds (average cast time in logs when testing).

I tested imp's Firebolt's average cast time on dummy with and without "Firebolt macro". Without macro my imp spent 714.5 seconds for 300 firebolts (2.382s per cast) and imp's dps was 482.7. When I used macro (spammed it like I would spam spells in Patchwerk-style fight), total time was reduced significantly. This time it took 646.1 seconds cast 300 firebolts (2.154s per cast) and imp's dps was 536.3.

So imp's dps increased by 11.1% and average cast time was reduced by 9.6%. It is still higher than simulated cast time and I think SimulationCraft gives too high dps for imp beacause of too fast cast time and this also means too many Empowered Imp procs. If I simulate 300 seconds Patchwerk-style fight, Firebolt's average cast time should be 1.85 seconds, but based on logs it was 1.96 seconds in my latest Festergut fight (that fight lasted 274 seconds so BL was up 14.6% of the time vs simulated 13.3%).

Hello, im not sure if this has been asked before, but the question i am pondering is not answered in the first post.

Question: If i can conflag 1.5-0.1 sec before immolate runs out, should i conflag and then reapply immolate,+CB+ incinerate. Or just conflag and CB, x2incinerate w/out immolate up.

Reapply immolate. CB and incinerate both gain 10% damage from Fire and Brimstone.

Just wanted to point out that the hit section is worded very strangely. Both column headers basically say "with talents / debuffs ". Should the first column say without?

Long time reader, first time poster. Just came back to WoW after a year off and I am not as up to date with theorycrafting as I would like to be.

I have a couple of questions that I dont find answers on in this thread. As I obtain ICC-gear I am getting less and less spirit on my gear and I currently only have ~400 spirit. I still use the LT-glyph, this may be wrong? My instinct says I should change it out to Glyph of Imp. Under the spirit section Krilz writes that dropping the glyph of LT inevitable. I am curious when the breaking point comes.
And that leads to another question. If I change to glyph of Imp, would it not be better to use a 0/18/53 specc instead of the suggested 0/13/58 specc that is supposed to haver the highest theoretical dps? I am not very good at math and have a hard time calculating on this but beeing a orc warlock my instinct tells me I would benefit more from 3 points in Unholy power and 2 in backlash than speccing Soul Leech. Please correct me if I am wrong. I know for a fact that there is much smarter warlocks than me at this forum so a response backed with some theory to my questions would be highly appreciated.

Calculating the breaking point of spirit when it comes to the Glyph of LT and other glyphs is pretty hard and too my knowledge hasn't been done by anyone. The best way to see for yourself is to make a profile at Wowhead with your current gear and throw it in Simcraft. Then replace the LT glyph with Imp (which is higher for me). That way you can see with one is the best exactly for you.

Also many people seem to disregard the benefits of Imp. SL. It's actually much better than you think. Without the LT glyph and with Imp. SL you will gain a huge amount of mana regen which makes LT:s unnecessary and you can put those GCD's to more DPS time. It really pays off in the long run, especially for nuke fights that doesn't last that long (Festergut comes to mind).

However, don't take my word for it. Most assumptions here are based on Simcraft and if in doubt, run it yourself with whatever gear/spec you think might be better.

I'm wondering if anyone who has been able to get on the 3.3.3 PTR has had a chance to test the change to Immolate. I wanted to do this test myself but haven't been able to log on.

First, how does it interact with Empowered Imp? It looks like the SimCraft implementation has the dot gaining 100% crit chance when cast with Emp Imp active--which would be nice, but it seems entirely possible that in the actual implementation the direct portion of Immolate will consume the buff and the dot will not benefit.

And if that turns out to be the case, I wonder what the effect will be on the Nevermelting Ice Crystal. Currently this trinket is horrible for Destro, but popping it before applying Immolate might be effective when the dot can crit. However, if the direct portion of the spell crits and removes a stack of the trinket's buff, does the dot benefit from 4 or the full 5 stacks? This is obviously much harder to test, and if nobody else feels like doing it then I will when the patch goes live (or if I manage to get on the PTR). But if the first question is answered favorably--i.e. Immolate dot benefits from Emp Imp--then it's probably safe to assume this one can be answered favorably as well, and the dot will get 5 stacks even if the direct damage crits.