CoD vs. CoA in Instances and RaidsFollow

Hello all.

I am fairly new to warlocks and mine is currently at 69. Switched my spec quite a bit, but right now I am 42/18/0. I plan to go something like 42/8/11 at 70 to focus on raiding and PVP.

http://armory.worldofwarcraft.com/character-sheet.xml?r=Fenris&n=deyanna

My question is in instances (5 Man, 10 man, and 25 man) fighting bosses, which is better a single Curse of Doom or multiple Curse of Agony. This assumes that the boss will live longer than 1 minute for CoD to proc.

Lets say for arguments sake my amp curse is not on cool down and I use it.

Is 1 Amp'ed CoD > (1 Amp'ed CoA) + 1.5 CoA ?

CoA is a 24 second DoT, so you can get 2 1/2 CoA in a minute. Keeping in mind that the 1/2 CoA is not 1/2 the damage as CoA scales over the 24 seconds.

BARE WITH ME AS I TRY AND WORK THROUGH THIS....

Not factoring in +damage from gear:

Base CoA r7 = 1356
2/2 Imp CoA = 136
5/5 Shadow Mastery = 136
5/5 Contagion = 68
------------------------
total = 1696
AMP'ed total = 2544 (1696*1.5)

So a CoA rotation would yield:
1 AMP'ed COA = 2544
1 Reg CoA = 1696
1/2 Reg CoA = 424 ( I estimated that 1/2 a CoA time would yield 25% of the total)
-------------------
total = 4644

Base CoD = 4200
5/5 Shadow Mastery = 420
-------------------
total = 4620
AMP'ed total = 6930 (4620 * 1.5)

So if this is accurate then a CoD is MUCH more effective. I am sure I am missing something.

Some follow up questions:
a) Can the damage of the CoD be resisted? That would suck to wait a full min to see a resist.

b) Does your +damage from gear affect either one more?

Interesting thought...

If my gear gives me +1000 damage.

Then:

1 AMP'ed CoA = 2544 + 1000 = 3544
1 Reg CoA = 1696 + 1000 = 2696
1/2 Reg CoA = 424 + 250 = 674 ( I estimated that 1/2 a CoA time would yield 25% of the total)
-------------------
total = 6914

Base CoD = 4200 + 1000
5/5 Shadow Mastery = 420
-------------------
total = 5620
AMP'ed total = 8430 (5620 * 1.5)

Still looks like 1 AMP'ed CoD is better.

*****************************************************

If my gear gives me +2000 damage.

Then:

1 AMP'ed COA = 2544 + 2000 = 4544
1 Reg CoA = 1696 + 2000 = 3696
1/2 Reg CoA = 424 + 500 = 924 ( I estimated that 1/2 a CoA time would yield 25% of the total)
-------------------
total = 9164

Base CoD = 4200 + 2000 = 6200
5/5 Shadow Mastery = 420
-------------------
total = 6620
AMP'ed total = 9930 (6620 * 1.5)

Closer, but CoD still wins.

Any thoughts? Is my math close?

What do you all do?

Jenova, you should do the math.

Also, Imp. CoA has been fixed for a long time, it affects the full package now. I said it before, you were going to test it, remember? I suggest doing it. You'll be pleased.


1 minute of CoA, perfect up time = 2.5 CoA, w/ .5 CoA actually being equal to 1/3 of the total damage.

1 minute of CoD = 1 CoD



Thus @ 1500 shadow power:

CoA = 1.25x[1356+(1.2x1500)] = 3945

Total CoA = 2.33x3945 = 9205

CoD = 1.0x[4200+(2.0x1500)] = 7200

The major diffence maker is Amp Curse. Amp. Curse does only affect the initial damage of the spell, adding very little to CoA while adding a very large amount of damage to CoD (678 vs 2100 respectively).


The real answer to maximizing your DPS is to Amp Curse+CoD on cooldown and CoA otherwise.


Jenova, you're using information about CoD and CoA from almost 6 months ago. Things have changed and the theory craft with it. Even opportunity cost is a tough justification for CoD given the highly mobile nature of most TBC boss encounters.

Yes, CoA will do more damage than CoD. If we compare them directly to each other. However, let's take a 2 minute segment. Going by your math above, 2 CoDs will cause 14400 and 2 mintes of CoAs 18410.

CoA CoD
18410 - 14400 = 4010 more damage going with CoA right? Wrong.

Now, you had to refresh CoA 5 times in that 2 minutes. How many SBs could you have fired off instead of those CoA's? Just one critting and you made up the difference. Therefore, CoD is better as it allows more SBs.

Personally, I never use either, I'm always on CoS Duty.

Thanks all.

I never thought about the global cooldown of the CoAs and the mana cost.

One last question...

Does the CoD ever get fully resisted?

Average damage on a 1500 shadow power SB is just under 2k (2043). The opportunity cost of casting CoA in a 2 minute turn around (i.e. 7.5 seconds vs. 3 seconds) covers the cost of 1.8 SB (meaning .9 SB/minute). .9 SB in D/CT is 1838 damage. That's less than the damage differential between CoA and CoD.


Also, evidence... since I have to do all the work.

My own toon, 1140 Shadow Power (this is out in SMV) with Imp. CoA, Contagion and Shadow Mastery.

If Jenova is correct then the average damage of my CoA will be 3269 damage, meaning it breaks down into 136-137 damage for the first 4 ticks, 272-273 for the next 4 and 408-409 for the next 4.

If I am correct the average damage is 3405. 141-142 for initials, 283-284 for middle, 425-426 for late.

Well testing empirically we find 2 very large problems. First, the damage done by my CoA is 3408, not 3405. Small differene but worth noting. Also, the damage is distributed much more evenly across the 12 ticks. The first 4 ticks go for 213 damage (and 1 for 214 actually) each. The next 4 put out 284 damage each, on the nose. The next 4 are 355 (and 1 354). The first 1/2 of the DoT, therefore, is actually 41% of the total DoT damage. Granted it's only another 300 or so damage in favor of CoA, but like I said, worth noting.

Of course, screen shots or it didn't happen:
http://i27.photobucket.com/albums/c173/Lokiliesmith/WoWScrnShot_031308_112122.jpg

http://i27.photobucket.com/albums/c173/Lokiliesmith/WoWScrnShot_031308_112156.jpg

So how much difference in dmg we got in CoA vs CoD after 1 min, 2mins, etc. etc.?

presuming imp.CoA works. presuming use of amp.curse (so afliction lock)

CoA. 1356 base damage
1.2 coefficient
CoD 4200 base damage
2 coefficient

universal multipliers
CoS 1.1
SM 1.1
SW 1.1
Misery 1.05
presuming 60% ISB (low, but easier for here). 1.12
for CoA
imp.CoA 1.1
cotagnation 1.05

presuming 1000 +dmg. 25% crit with destruction spells. hit cap.

CoA.
(1356+(1.2*1000))*(1.1*1.1*1.05*1.1*1.05*1.1*1.12) = 4620

imp.CoA
((1356*1.5)+(1.2*1000))*(1.1*1.1*1.05*1.1*1.05*1.1*1.12) = 5846

CoD.
(4200+(2*1000)*(1.1*1.1*1.1*1.05*1.12) = 9704

amp.CoD
((4200*1.5)+(2*1000)*(1.1*1.1*1.1*1.05*1.12) = 12992


counting in amplify curse. 1 use over 3 minutes (180 seconds)

180/24 = 6.5 CoA, 1amp.CoA = 11.25 seconds of GCD
2 CoD, 1amp.CoD = 4.5 seconds of GCD
GCD variance of 6.75 seconds

CoA damage over 3minutes
(4620*6.5)+5846 = 35876

CoD damage over 3minutes
(9704*2)+12992 = 32400

damage variance of 3476.

shadowbolt damage.
541-603 damage (572 avg)
0.8571 coefficient

(572+(0.8571*1000))*(1.1*1.1*1.1*1.05*1.12) = 2237 avg hit.

SB avg damage inc.crits = 3076.

SB cast time 2.5 seconds

6.75/2.5 = 2.7 additional SBs can be cast with CoD over CoA.

3076*2.7 = 8305.

total CoA damage over 3minutes = 35876
total CoD damage over 3minutes = 32400+8305 = 40705

total variance. 4829 damage in favour of CoD.


there you go loki. as i said, i was right. I've done these maths a million times. it's always come out this way and it's true for all specs of warlock.

as you increase spell damage, CoD gets closer and closer to CoA, SB slips a bit further behind. end result is little to no variation. end rankings won't change.





Edited, Mar 13th 2008 12:44pm by Jenovaomega

Well actually i was just building a case here jenova^^

nm. what the outcome would be, that in a perfect enviroment CoA might outdamage CoD abit, there is no such thing as a perfect setup in a raidenviroment.
Here the 3 seconds come into play wich can be used to time your SB-Nightfallprocc

Alot better or keep an even tighter DoT-rotation.
Because if you add up the secons (read ticks) where CoA isn´t up due to you casting a spell, gcd is triggered or you are not in range for some reason (Archimonde, anyone?) the CoD-Caster will gain ground and mostly likely
easylie surpass the CoA-caster...

If the math jenova made is right, or close than it´s even alot worse for the CoA-caster.

Theorycrafting (as the name implies*hint*hint*) unfortunatly is only viable on forums or in a bar, but too often the game is too dynamic, too keep up perfect cycles, rotations and whatnot and in this particular case when you cast CoD (fire and forget) this proofes true, contrary to CoA where you have one more DoT to keep an ultra-tight cycle on in order for it to perform.

Having 1 25% crit rate does not add over 25% of your spells average damage to your average damage. That is absurd.

For future reference, average spell damage yield overall on a direct damage attack is:

[(chance to crit) × (crit damage)] + [(chance to hit and not crit) × (average damage)] + [(chance to miss) × 0]

So, a 2237 damage with a 25% crit rate yields

.25 * 1.5(2237) + (.99-.25)2237 + (1-.99)0

838.75 + 1655.38 + 0

That's 2494.13 Average damage. Where you got 3076 is beyond me.

Edited, Mar 13th 2008 1:28pm by Loki

I hope that no one is bored enough to calculate the lifetap requirements of 2 different 10 minute spell rotations but the basic theory behind it follows that if the mana cost of CoAx6 < 2CoD + 2(.9SB), then the lifetap requirements are lower for the mixed rotation than a pure one. The reasoning? The divergent part of the spell casting sequence allows for more damage by CoA. If this is done by spending less mana then the sequence is obviously superior over all.


CoA = 275 mana

CoD = 380

SB = 420


CoA x 6 = 1650

CoD x 2 = 760

.9SB x2= 756


CoA produces the damage at a cost of 144 mana/3 minute rotation. Given the max boss fight time of 10 minutes, 480 more mana is spent in the CoA rotation. Technically it's less of a difference (438 mana spent) because you cannot spend .9SB worth of mana on a SB, either you cast it or you don't. For more realistic fight lengths (i.e. Not Vashj, Kael, and Illidan) of 6 minutes (even that's kind of long, imo, but I've been told my guild kills things ungodly fast), it's only 288 mana, which isn't enough to cast anything but another CoA... oh sweet irony.

Why don't you do some math? It's your theory, you do the legwork. Saying "one should crit and the crit justifies the use by raid wide damage" doesn't mean anything. First, we're talking about personal DPS. 2nd, you've got no information, just an untested hypothesis.

The preliminary math is showing mixing and matching to be "optimal" but close enough that it won't be noticeable, while pure CoD > pure CoA.

i did factor in ISB. 60% uptime leading to an average 12% damage increase to self. yes technically the CoD rotation will have a slightely higher ISB uptime... but that's to much to factor in :P