First time poster, long time reader of multiple classes and forums.

I can see where this could be a *small* issue in timeline. As I understand, as of the hotfix, SB and Haunt both update the duration and the "Haste" of the tick. Lets just assume for a quick down and dirty calculation, base ticks happen every 3 seconds. We start with 10% haste, therefore ticks happen ever 2.7 seconds.

0 Sec - Corruption lands
2.7 sec - 1st corruption tick
5.4 sec - 2nd corruption tick
8.1 sec - 3rd corruption tick
10 sec - new SB lands with an added haste of 10% (Total 20%) due to trinket

Now, does the next tick happen at 10.8 seconds or will it get refreshed as of the landing time and happen at 12.4 seconds? Lets assume it ticks at 10.8 seconds and every 2.4 seconds thereafter.

10.8 sec - 4th corruption tick
13.2 sec - 5th corruption tick - Haste proc from trinket falls off
15.0 sec - another SB/haunt lands with 10% haste.

Now, again this has to be tested (I admit I haven't tested any of this, just scenarios) and find out, does the new spell haste (lowered) take affect after the next tick at 15.6 seconds (we'll assume yes to flow with the previous example) and further ticks come every 2.7 seconds after again.

15.6 sec - 6th corruption tick
19.3 sec - 7th corruption tick
22.0 sec - 8th corruption tick
24.7 sec - 9th corruption tick
27.4 sec - 10th corruption tick (This tick would've happened at 25.2seconds with the haste increase not overridden and corruption would have dropped off)
30.1 sec - 11th corruption tick
32.8 sec - 12th corruption tick

Now we have to find the opportunity cost analysis (and I leave this to someone who knows all the coefficients and takes all the math into affect) of casting Shadowbolt as a filler as it may impact a wonderfully hasted corruption. In the above calculation with just a 10% increase in haste, corruption would have been done 2.2 seconds sooner (7th tick because of how I dealt with tick assumptions). This allows a you to redo a corruption and start another spell in the time you saved only off a 10% haste increase.

Would another spell be a better spell to choose right after refreshing corruption at the end of BL and Haste pot times, so that Corruption wasn't refreshed but actually dropped off at the end of those times?

I figure by rough calculations if you get a 30% haste increase (total 40% in this example) and you time it just right at the end of the haste period (land an SB/Haunt right before haste wears off), you can get 6 ticks of damage 5.4 seconds faster. If you were to refresh with a 10% haste after the first tick, going from 40% to 10% haste, you lost 5 ticks of corruption damage over a base period (16.2 if we go back to 10% haste versus the 10.8 second with 40% haste). So, could you make the damage lost by refreshing your Corruption with the slower haste in the same time period? Or would it be better to find another filler spell just for that time period. This is situational, but it would also apply to procs like trinkets, etc.

If my logic is off, I apologize and I'm sure someone will tell me.

Thanks,

That could be a valid concern, with corruption being refreshed as often as it is now, there are times where you may not be able to take full advantage of a trinket, eradication, speed pot, bloodlust. As it stands this would serve the spec at least making it playable without having to worry about corruption falling off at all.

I figure once this issue is finally addressed and put to rest they can possibly looking towards the bigger picture of affliction and possibly buffing or addressing any of it's other shortcomings. For the time being let's just get the spec working.

I have seen arguments like these crop up all over the place lately, arguing that the increased refresh rate on corruption ends up being a de facto nerf since we benefit less from haste/sp proccs due to the fact that refreshing corruption often will make us overwrite a buffed corruption with an unbuffed one very quickly. In my opinion that reasoning is flawed though, since it seems to overlook that basically every relevant procc like a trinket or eradication is not a "on next cast" procc but stays active for a fixed duration, like 10sec in the case of eradication. For example in the quote above one should really specify when the trinket procc occurs. Sticking to the numbers above: You say the trinket procc falls off at 13.2sec. Now since the min duration of trinket proccs I know of is 10sec, I will just assume that the procc duration is 10sec. In that case the procc should have occured at 3.2 sec. If that however is the case than corruption should have either been refreshed to the procced version of it much earlier than at 10sec marker or you haven't been casting either SB or haunt for 6.8sec, which is either unrealistic or a not representative "worst case" scenario. In any case generally speaking it is true that once a procc has faded we will be much quicker to update corruption to the non buffed version. But people seem to overlook that the argument goes two ways, since with a higher refresh rate we will also be much faster in updating corruption to the buffed version. I am aware of the fact that most proccs having a 10sec duration and haunt having an 8sec cd potentially gave us the chance of getting two procced corruptions out of one procc, but for that case the stars really need to allign since the first haunts needs to land in the first 2 seconds of the procc and than you have to be able to cast haunt basically exactly when it comes off cd while still keeping your usual rotation up. The likelihood of this event should be that low that it doesnt influence our dps to an extent that makes it worth any consideration.

I've just tested on dummy and can confirm that each refresh of the Corruption will reset the haste factor of the Corruption ticks.

I applied Corruption and waited until Eradication procced, then popped Potion of Speed and re-applied Corruption. I then cast SB every few seconds for over 5 minutes. With Eradication + Potion of Haste and my gear at about 20% haste, I got ticks under 2 seconds. Later the ticks were 2.3 or 2.5 seconds depending on new Eradication procs as I refresh with new SBs. (assuming I read the logs correctly)

Logs: Wow Web Stats

One might assume the Crit effect will also reset.

Crit keeps rolling -- see notes one page back.

I fully agree that you will be able to refresh the hasted corruption a lot faster. In my example, it was basically a completely theoretical, not taking into account all the other spells you throw up. 6.8 seconds without an SB/Haunt landing is fairly unrealistic when I think of that. You would have to lifetap, CoA, UA and then haunt to really get towards that number realistically.

I would like to point out I'm not saying it is a nerf, but that it may change current play style to keep and maximize the dps. However, I was focusing on the end portion, and wondering, for those very attuned with the ticks/rotations - would it be best to 1) Not cast a refresher corruption when all your haste procs fall of and instead cast another dps spell (and/or lifetap) or should you refresh it. Looking at the start of haste procs, should you disregard your rotation at the start of any haste proc to get that corruption refreshed immediately? My guess is that, overall, it is going to be a very small dps increase, if any, but I'd like to have someone who can go through the numbers (someone who deals with the numbers and knows them much more than I do) determine that.

Drain Soul has now been hot fixed to refresh corruption.

Note that it is only refreshing on the cast, not on the ticks of Drain Soul. This should help immensely for the execute phases.

I expect this as well, and while it may "nerf" our damage slightly, it makes sense to me that attacks all scale equally with various procs; increased damage for a period equal to their uptime. The value of the nevermelting ice crystal, for affliction, is far higher than its proc would suggest for other specs, and while it's nothing like the several-thousand DPS increase of rolling haste, it's still the kind of thing I'd expect to see fixed - just not overpowered enough to warrant a hotfix going out.

Glad to hear it. That will help our execute phase out immensely. I saw the post where they said they couldn't give the date for the hotfix, but I have been very encouraged with how quickly Blizz has jumped on and squashed these bugs in addition to listening to concerns and implementing "common sense" changes

I dropped Glyph of Haunt a few days back and replaced it with Corruption to try it out.

So far it seems to be a DPS increase. With the amount Corruption ticks now, especially on multi targets, I seem to get Nightfall procs a ton.

This was our Patchwerk kill on Tuesday night for the weekly raid quest, this was using Corruption, Life Tap, and Decay Glyphs....

Was Attog, the other lock in the number 10 spot, another affliction lock or the demo lock? and if affliction, does he use the haunt glyph over corruption?

edit: never mind he is demo. He is in your party.

Yes it was actually only 1min 40sec fight. Since swapping to this Corruption Glyph, "most" fights now, my Shadow Bolts have been the bulk of my damage. Before this, Corruption was usually 30-35% of my total damage with Shadow Bolt in 2nd.

I'll have to swap both Glyphs around and test these out.

My concern about newest everlasting affliction;

ok there are 4 spells can refresh it. dumping DL, 3 left.

Haunt: Can only be cast as much as cd allows. In this scenario corruption drops off in raid environment.

SB: Function of talent remains with this change. But till %25.

DS: here is my concern. it supposed make everlasting affliction work without worrying about Corr drops except extreme situations. But to when we put mechanic of refreshing i see a flaw;

DS refreshes at spellhit like haunt, so when do we cast ds? when we interrupt it?

we interrupt DS for all dots and haunt. there are many scenarios can happen during execute but im going straight for worst one which is all dots are up and remaining times of those dots are longer than haunt CD. So that means we ARE NOT going to interrupt DS to recast ANY DOT till next haunt cd is ready. This scenario makes DS everlasting affliction refresh futile. Because the 2 spells in rotation refresh Corr isnt casted for duration of a haunt cd. that brings us back to haunt refresh only situation which was the problem in first place.

However there is a solution ofc. Recasting of DS while already channeling it. To be more precise for those using Fortewarlock with "show ticks" enabled, check your priorty list of spellcasting after 2nd tick of Drainsoul(if DS is still on top of spellcasting list recast it).
Solution from blizzards side: refresh at each tick. So Requirement of an addon to play will be removed.

Seems like a rather unlikely situation. Since DS hotfix I've played affliction for all TotGC and this weeks ICC I don't believe that was ever an issue. With Life Tap, UA, Haunt, CoA all being active while we drain the chance of corruption falling off is almost nil. I use Drain Soul Timer which makes a noise when DS ticks making it almost simple to stage your rotation efficiently.

Keep in mind that even in that "worst case scenario", corruption refreshes on the DS cast after your Haunt cast, effectively giving you another GCD's worth of uptime. Given the short CD on Haunt, that's a good step towards 100% corr uptime. And I agree - that situation is quite unlikely in the first place, and should be something we can deal with.

Yes it seems very extreme i could not able to make corruption fade with eradication+bl+hastepot+ 425 haste trinket use. however it was a save from less than 1 second being away from fading.

One thing that should be noted:

[Reign of the Dead] procs only from direct damage spell. Affliction stops using direct damage fillers at 25% and 25% until boss death we can only count on haunt to proc the trinket. At best we are looking at one proc every ~24-28s, where as you are getting full benefit from [Nevermelting Ice Crystal].

Yeah, I'm gonna put the numbers in the trinket discussion thread, but still, over a 200 dps increase? that's huge for losing 13 ilvls in the slot.

Does anyone have any insight as to what would be a bigger DPS increase between popping Potion of wild magic with nevermelting ice crystals BEFORE your initial corruption, or saving potion of wild magic for bloodlust?

I think popping a potion of wild magic just before you initialize combat, would still give plenty of time to get a corruption up on the target before it expires. Therefore you would still have your potion available for a heroism phase, if it is after the potion cd.

Assuming that you don't pre-pot. Popping the wild magic with nevermelting ice would be a far bigger DPS increase I'm sure because the crit from that potion will be rolled throughout the entire fight you keep that corruption rolling whereas if you pop it during bloodlust you're not going to get nearly as much out of it.

(First post on EJ please be gentle)
Bastetswarrior Post on Soc vs. Rain of fire (The Mage Equivalent: The Destruction Warlock)

Got me thinking, at what point does spamming SoC become higher dps than Tab doting targets(assuming the target live long enough that the dot gets off all it's ticks).

From Simcraft the DPET of UA is 17191 for 55/0/16 but that includes haunt and 2 stacks of shadow mastery on the target. So UA DPET on tab dotting would decrease by approximatly 43%. So UA would be about 9798.87 DPET with out Haunt and Shadow Embrace on the target.

Haste =23.61%
SoC Crit rate= 30.28% +5%= 35.28%
Spell Power =3743

Affliction talents
Improved Corruption: 5% crit on SoC
Shadow Mastery: 15% Shadow Damage
Contagion: 5% Soc Damage
Malediction: 3% Spell Damage

Assuming the plus damage talents are Mulitplictive(Please tell me if this is wrong) this should increase the Damage of SoC by x1.243725.

Ds= (Average Value + (Coefficent * Spell Power))*talents*CoE
Ds =((1897+1633)/2+.166*(3743))*1.243725*1.13 = 4013.5

using Bastetswarrior formula

N*4013.5*((0.3528*1.54)+(1-.3528))= N*4778.12

DPET = (N*4778.12)/(2*(1-.2361))= (N*4778.12) / 1.5278)
So with 5 targets like anub adds

DPET for Soc is (5*4778.12) / 1.5278 = 15637.256

With three target Soc becomes

(3*4778.12) / 1.5278 = 9382.35

I don't know the DPET of Corruption, I plan on calculating it later on today(and editing this post). Shadow Bolt has a DPET of 5248, but if it was used to refresh corruption wouldn't it's DPET become 5248 Plus DPET of Corruption mininus the percent of Corruption ticks still on the target multipluled by the DPET of Corruption.

let me try and make that more clear:
DPETC= DPET of Corruption
C% = percent of Corruption ticks still on the target
DPETSB= DPET of Shadow Bolt
DPETSBC = DPETSB + (DPETC -(DPETC*C%)

Thanks for your time, and i would apreciate any feedback on this. If their are any errors in my logic or math please point them out so i can fix this post.
EDIT: I Completly Forgot About Haunt and Shadow Mastery on the UA that the simcraft is calculating the DPET. So UA DPET on tab dotting would decrease buy approximatly 43%. So UA would be about 9798.87 DPET with out Haunt and Shadow Embrace on the target.

EDIT 2 Corruption: Corruption none crit damage(Dc) works out to be (1080 + 1.86*(3743))*1.1*1.15*1.03*1.13= 10250.356. It's total damage should be Dc*(.3928*2.09 +(1-.3928)) =14758.667. It DPET would be about 13722. It appears that if their are less than 5 adds tab doting is higher dps than SoC.

EDIT 3:Well this may not be correct, because We would have to compare the fillers. If three targets where to live for 30 seconds, This means you could Place corruption and UA on all three targets and then spam shadow bolt as a filler and to refresh Corruption. Your First Corruption would last for 28.9s, the seconds one would be 27.7 and the third would last 26.5. Lets assume you get 4 UA off(one target gets refreshed). The other ~22seconds r spent spaming shadow bolt. You get off ~14 shadow bolts. The shadow bolt would amount to ~105065 damage. the UA would total ~42156 damage. The Corruption would do 90650 damage. That totals 237871 damage done. In that time you could have gotten off ~20 SoC, which would have totaled 286687 damage. Looks like Even if their are only 3 targets spaming SoC is better.