Back when AoE caps were first applied Faxmonkey released a list of approximate caps, since I was rather bored this morning I decided to expand the list with all our AoE (except blizzard, haven't done that one yet) and all ranks of the spells.

For any who don't know what the AoE caps are, lets assume for a moment that Arcane Explosion does 200 damage a target. The total damage done by the spell is easy to calculate, just multiply the targets it hits by its damage done, so vs one target:

200 damage * 1 target = 200 damage total

Vs two:

200 damage * 2 targets = 400 damage total

Vs 10:

200 damage * 10 targets = 2000 damage total


Too easy right? Now lets assume that you apply an artificial cap to these spells, where the total damage output of the spell can't exceed X amount. lets say for our 200 damage arcane explosion, the damage cap is 2000 points of damage, so for 1-10 targets we havent exceeded that amount of damage, so we deal full damage. What happens though, when we hit 11 targets?

200 damage * 11 targets = 2200 damage total

That exceeds the cap. What appears to happen in game, is that when you exceed the damage cap for a given spell, the game evenly distributes your damage cap over all the targets, like so:

2000 damage cap / 11 targets = 181.1818 damage per target


There are several really easy places to observe this behavior, low end instances are an easy place, the actual "best" place is the quest in Terrokar "The Dread Relic" (http://www.thottbot.com/q10877). When you open the chest and take the quest item from it, 29 level 60 zombies with around 600 hitpoints each spawn around you, they hit for about 6 damage a hit on cloth, and make for excellent cap testing. After you take the quest item you can continue to open the chest over and over and over again summoning the Zombies (as I did, for 10-15 testpoints per rank per spell, this has to be by far the most fun testing I've done yet).


Things I observed beyond the caps themselves while doing this testing:

1) Resists/immunes don't count towards the damage cap. If a target resists, its share of the damage is given to the rest of them.
2) Talents and gear do not change the damage caps, this means that the damage cap is exactly the same if you have 0 +damage as if you had 1000 +damage. Effects such as Molten Fury do not add to your damage done to the targets.
3) Effects such as Curse of Elements *DO* add to your damage done to that particular target, I cannot confirm at this time whether or not it comes out of the cap or not, I observed this while writing the addon I used to gather the data on this while a warlock was nuking Dr Boom and I was flamestriking his bomblets.
4) Extra damage that comes from critting does not count towards the cap.
5) The game is terrible at reporting large numbers of AoE damage events at once. Vs 29 targets about every three data points or so I'd lose one or two damage events, lending to a bit of chaos in the data.


Below are the tables I came up with for caps, along with spell damage ranges (sourced from http://wow.allakhazam.com/db/spelllist.html?class=Mage&sort=name), and how many targets its set up for based on min, mean, and max damage.


BIG HUGE ZOMG DISCLAIMER:

While doing this testing I got a lot of funky readings, caps would hold strong for 10 datapoints, then one would be 600 damage off. I was able to attribute most of this to missing damage events, but sometimes it just plain didn't add up. The numbers shifted plus or minus about 50 damage (probably due to rounding), so all these numbers are rounded averages. I took an average of the data collected, and rounded it to the nearest 10. These numbers are by *NO MEANS* absolute caps, but are guidelines of around how much ceiling you have to work with with a given spell. This mechanic is hard to feel out with any degree of accuracy worthy of decimal places, so my usual four places worth of accuracy has been forgone for this test.

Columns are Rank, Minimum Damage, Average Damage, Max Damage, the observed damage cap, then that cap divided by the Min, Mean, and Max damages. The last three columns represent how many targets you can hit for full damage and be right on the cap.

Flamestrike				Targets By	

Rank	Min	Mean	Max	Capped	Min	Mean	Max

1	52	60	68	980	18.85	16.33	14.41

2	96	109	122	1780	18.54	16.33	14.59

3	154	173	192	2670	17.34	15.43	13.91

4	220	246	272	3660	16.64	14.88	13.46

5	291	325	359	4700	16.15	14.46	13.09

6	375	417	459	5940	15.84	14.24	12.94

7	471	523	575	7830	16.62	14.97	13.62

------------------------------------------------------------------------------------------

Blastwave				Targets By	

Rank	Min	Mean	Max	Capped	Min	Mean	Max

1	154	170	186	2470	16.04	14.53	13.28

2	201	221	241	3130	15.57	14.16	12.99

3	277	303	329	4140	14.95	13.66	12.58

4	365	399	433	5250	14.38	13.16	12.12

5	462	503	544	6900	14.94	13.72	12.68

6	533	580	627	8120	15.23	14	12.95

7	616	670	724	9440	15.32	14.09	13.04

------------------------------------------------------------------------------------------

Arcane Explosion			Targets By	

Rank	Min	Mean	Max	Capped	Min	Mean	Max

1	32	34	36	610	19.06	17.94	16.94

2	57	60	63	1180	20.7	19.67	18.73

3	97	101	105	1800	18.56	17.82	17.14

4	139	145	151	2470	17.77	17.03	16.36

5	186	194	202	3180	17.1	16.39	15.74

6	243	253	263	4000	16.46	15.81	15.21

7	306	318	330	5300	17.32	16.67	16.06

8	377	392	407	6730	17.85	17.17	16.54

------------------------------------------------------------------------------------------

Dragon's Breath				Targets By	

Rank	Min	Mean	Max	Capped	Min	Mean	Max

1	370	400	430	5370	14.51	13.43	12.49

2	454	490	526	6470	14.25	13.2	12.3

3	574	620	666	8470	14.76	13.66	12.72

4	680	735	790	10100	14.85	13.74	12.78

------------------------------------------------------------------------------------------

Cone of Cold				Targets By	

Rank	Min	Mean	Max	Capped	Min	Mean	Max

1	98	103	108	1650	16.84	16.02	15.28

2	146	153	160	2350	16.1	15.36	14.69

3	203	213	223	3160	15.57	14.84	14.17

4	264	277	290	4000	15.15	14.44	13.79

5	335	350	365	5120	15.28	14.63	14.03

6	410	429	448	6500	15.85	15.15	14.51

And as is customary, for the TL:DR crew:

Theres a bunch of numbers up there you don't care about because the post was too long.

So Dragon's Breath is our best AoE attack? /cry

No, the cap itself isnt really important, how many targets you can hit with it is. DB may be able to do the most damage at once, but AE can hit the most targets for full damage at once. Its those last three columns that truly matter.

I guess I did forget a big crayon explanation of the figures and what they actually mean.


The cap itself just represents the ceiling of the spell, without relating it to the base damage of the spell its not very useful at all. Dividing the cap by the base damage gives us the number of targets we can hit at once (which is the true measure of AoE spells in relation to their caps). The ranking of 'best' to 'worst' using that metric to judge them by goes like this:

Arcane Explosion 17.17
Cone of Cold 15.15
Flamestrike 14.97
Blast Wave 14.09
Dragon's Breath 13.74

So wait... the target list - Are those numbers the calculated 'number of targets you have to hit to reach the cap', or is that the number of targets you hit in your tests to calculate the cap?

Those columns are Cap / Min, Cap / Mean, Cap / Max, the only measured values in the tables are the caps themselves, the rest are calculated or sourced from datamining sites.

Ahhh, thought you were doing it the other way around - ie. Looking at the average hit and multiplying it by the avg. target number to determine the cap.

Gotcha

Clarified the table post.

Basically all the addon did was say "Woa, you cast an AoE spell, and I saw X sucess events for it, Y were resists/immunes, check all the damage events, they're all within 2 damage of eachother so you're capped out, so multiply that damage number by (x-y) and we have our total non-crit damage done". Which is why the missing events screwed the observed caps up. If I really hit 29 targets, and I've only got 28 damage events registered my calculation will be fubar.

K i gotta ask... why are they doing this? I don't see any real good reason for this damage cap. Sure in reality the more objects in a blast radius would take slightly less damage but this is not reality.
Just another thing that blizz didnt see as "working as intended" or whatever.
Sure the change will be hardly noticable the majority of the time but one has to ask why they're making "important fixes" like this when there are many other things much of the WoW population is way more worried about.
Last I checked there weren't 50 threads complaining about how fast a mage can run a lowbie through an instance...
So many *Helpful* things could be done meanwhile they're doing this. I just dont get it.

Btw awesome job figuring this stuff out Z, you're amaze me :P

They did this to limit mages from simply running around AOEing everything to level and make money ;\

I think this data seems fairly accurate one point I have noticed myself however is bizzare damage behavior under some circumstances while AOEing..

I could be using AE, in a large group, deal 1700 to every target but a handful would only take 400 and mash as much as I might, that number would not increase whatsoever (These were level 60 mobs in Gahrons). It usually occured if I gathered more than 16 mobs and was VERY noticeable at the Troll Village cemetary in EPL. When 30 zombies spawn I could barely scratch them, yet they were still hitting me as hard as they did at level 60.

Sometimes the damage cap still seemed to be active even after I had stopped aoeing a large group and was starting to aoe another group (but with less mobs) several moments later

Also note: Dead bodies sometimes still count against the damage cap for no apparent reason.

---

Q u o t e:
K i gotta ask... why are they doing this? I don't see any real good reason for this damage cap. Sure in reality the more objects in a blast radius would take slightly less damage but this is not reality.

---

I think they first applied the caps because of people like Faxmonkey during the scourge invasion taking entire nodes at once. I think the idea of a cap itself is fine, but the fact that it doesnt scale with spec or gear makes it not smurfy. Spec yourself for Imp CoC and get some fair gear going on and you'll hit that cap really quickly.

---

Q u o t e:
Btw awesome job figuring this stuff out Z, you're amaze me :P

---

Bored geeks do crazy crazy things, heh.

---

Q u o t e:
I could be using AE, in a large group, deal 1700 to every target but a handful would only take 400 and mash as much as I might, that number would not increase whatsoever (These were level 60 mobs in Gahrons). It usually occured if I gathered more than 16 mobs and was VERY noticeable at the Troll Village cemetary in EPL. When 30 zombies spawn I could barely scratch them, yet they were still hitting me as hard as they did at level 60.

---

Interesting, Might be worth going back and checking to see if its partial resists causing that, I haven't seen what they can do to the mechanic. Easy place to test that is strat, the citizens have nasty resists that you cant penetrate through easily.

---

Q u o t e:
Sometimes the damage cap still seemed to be active even after I had stopped aoeing a large group and was starting to aoe another group (but with less mobs) several moments later

---

Hmm, havent seen that one yet. So you hit group 1 with X mobs for Y damage each, then hit group 2 with < X mobs for Y damage each again?

---

Q u o t e:
Also note: Dead bodies sometimes still count against the damage cap for no apparent reason.

---

Hmm, hadn't thought to check that, all my testing was done with targets stacked on corpses, that could explain some of the weirdness if its true.

Don't bother testing blizzard as you indicated you may in your original post; It is not modified with an aoe cap.

But if the math is correct on everything else, this is a good post to bookmark for theorycraft later on.

Check the PTR patch notes, its getting one in the next patch. I havent bothered doing it yet because I'd have to teach my addon to collect its periodic damage (since I believe its a different set of events that it throws), and dont care enough right now to do it, once the patch is finalized and in place I'll feel it out.

---

Q u o t e:
Mages


* Arcane Missiles: Some procs were not triggered by this spell and now will be (such as Eye of Magtheridon). The tooltip for this spell will now update correctly when a Mage is affected by spell haste.
* Blizzard: This spell is now affected by area damage caps. Its bonus damage coefficients have also been increased.
* Dragon's Breath: The movement-slowing component of this spell is now properly considered a Snare. Immunity and clearing effects will now work properly with it.
* Frost Nova: This spell is now affected correctly by area damage caps.
* Molten Armor: This armor can now be triggered while the Mage is sitting.
* Polymorph duration against PvP targets has been reduced to 10 seconds.
* Slow: It is no longer sometimes possible to Slow two targets at the same time in the outdoor world.
* Spell Steal now receives the proper amount of points from a stacked aura.

---

http://forums.worldofwarcraft.com/thread.html?topicId=174505927&sid=1

Just as a thought, wouldn't testing this over by Doc Boom in Netherstorm be a bit easier?

He endlessly spawns a great number of little bombs, that appear to have a short range and life. I typically see 10-15 of them up all at once. Typically, I head down there to test Chain Lightning out.

It might be a better, and less situational, place to test these things out, as they're all fairly tightly packed and make excellent AoE targets.

It works there, the best place in the game is that quest in Terrokar though, unlimited supply of 29 zombies at once that barely hit for any amount of damage.

The problem with Dr Boom testing is to test point-blank AoEs you get within his range and he blows you up.

Hmm.

Didn't consider that one...

Now I may have to go give him a hug, out of curiosity.

So when i clicked this thread, All i wanted to know was the new blizzard +damage coefficient.

Instead i got a bunch of useless tables, and nothing about blizzard.

This thread is full of fail.