Sidewalk

Which spreadsheet? 1.15? F9 and B18

CoE: You only put up CoE if the "extra" damage from the mages would be more then your CoA / CoD damage.

CoS: You only put up CoS if the "extra" damage from the sPriests/Locks would be more then your CoA / CoD damage.

 

PSGarak

As far as the spreadsheet goes, if you click on CoE you should get a dropdown menu allowing you to select which curse you are running.

The general wisdom is that if there are two people benefiting from a curse, it's better damage than CoA or CoD. If you have two frost or fire mages, you should run CoE. If you have yourself and another warlock, or shadowpriest, or boomkin, or arcane mage, you should have CoS. If do not have this number of casters (or the number drops below this due to deaths), or if those curses are taken care of, you may or may not want to run CoR, which involves its own barrel of research. In general, CoD > CoA not because of straight damage, but because it costs less global cooldowns to maintain. That is contingent on not getting knocked off, and running to completion; if your raid is tight on debuffs or the boss is low on health switch to CoA.

http://sig.gamerdna.com/quizzes/INFL...erks/Garak.png

 

Angyarr

On the topic of Nether Protection, I tested it this week and can say now that Illidan for sure switches targets if you are affected by Nether Protection. Of course /cancelaura works, but you still have a gap and the Shadow Blast is high enough damage to 2 shot anyone.

 

Fafhrd

To Konomi:

You need 5/5 Supression for best damage since you've ignored +hit gems for epeen +dmg ones.
Add Immolate to your rotation, and Dark Pact.

CoA/D is less damage than:
CoS, if the raid has 2 shadow damage players (unless you do 2000dps yourself)
CoE, if the raid has 2 elemental damage players
CoR, the raid has 3-4 physical damage dealers (most calculations adds ~5-6% to the dps each player, depending on boss armor)

CoR also help your tank maintaining aggro.
The buff part of CoR is negated if someone has 5/5 Improved Demo Shout.

 

Crepe

Awesome, this is pretty much what I was looking for. I was kinda pulling the shadow priest cast ratio out of my arse, since I didn't see cast frequency as part of the weighted average at first. However, now I see that it's simply a ratio of nukes/s compared to the raid's nukes/s. For a spriest at full burn with maxed out ImpMB, this would mean they're casting 1/5.5 + 1/12 nukes/s (MB and SWD's cooldowns, respectively). That's approximately 0.265 nukes/s with no chance of procing ISB. Thus, our result becomes (expanding out to make it a little more readable):

avg_lag = 100ms = 0.1s
spriest_nps = 1/5.5 + 1/12 = 0.265
affliction_nps = 0.4 * (1/2.6) = 0.154
destro_nps = 0.85 * (1/2.6) = 0.327
raid_nps = affliction_nps + 2*destro_nps + spriest_nps = 1.073 nps

spriest_crit = 0
affliction_crit = 0.15
destro_crit = 0.3

x = affliction_crit * affliction_nps / raid_nps + destro_crit * destro_nps / raid_nps

[top] (affliction_crit * affliction_nps + destro_crit * destro_nps) / raid_nps

(0.15 * 0.154 + 2 * 0.3 * 0.327) / 1.073

[top] 0.219 / 1.073

0.204

1 - (1-x)^4 = 1 - 0.4 = 60% uptime

 

Trickykid

Yep, that's how I model it -- I know there are some eccentricities about how ISB works, but as far as I can tell this is the safest way to go about it.

For the sake of gear-choice, I leave it in the component parts so that I can map haste rating into the equation more easily. Otherwise you'll need to have another equation for dNPS/dHaste.

Once you assign all of the different actors and their gear (crit and haste), it's relatively easy to get at least an expected increase in ISB uptime per stat point. Then throw in whatever the raid shadow DPS is and you can include that raid-benefit into "the next stat" table.

 

Gumibear

I'm not sure how big a difference this makes, but 5/5 Improved MB is every 7 seconds. 5.5 seconds to cooldown and then 1.5 to cast once the cooldown is finished. The cooldown doesn't start until the cast finishes.

 

Crepe

Oh, huh. I guess I hadn't paid attention to the fact that it had a cast time. Figured it was instant like SWD. I'll fiddle with that some.

Thanks for the help, Tricky. You should poke Leulier with something like this to add into The Spreadsheet.

Some other random discussion to keep things going in here: at what level of gear should you stop bothering with having the token affliction lock? Obviously, you can't have a full 0/21/40 raid since someone has to run an Imp for the tank (and if you're past that point, optimizing DPS doesn't matter). However, at what point will the raw scaling from 0/21/40 overtake the scaling of UA plus malediction? Is there ever a point?

 

LCN

On a totally unrelated note, and sorry if it has been discussed already, does [Syphon of the Nathrezim] proc use up ISB charges? I heard a rumor, and remember reading something about it in these forums, but as far as I know people never got into a conclusion whether its doing it in the current version. It seems stupid that propably the best Enhancement offhand (mainhand as well?) at the stage would have such a flaw, or as Blizzard would refer to it; an ability.
If it does, it seems cruel to deny the Shammies from loot, but seeing how much 3 warlocks and 2 shadow priests for example gain from ISB, it'd also be more beneficial dps wise to do so.

edit: Having worked my way around the stupid limitations of the search engine here, I found the Shammies in their thread stating that it would NOT eat up the charges. However, the conversation was a stub, and ended pretty much when one person said "lawl it does not!". So some sort of confirmation would be nice.

 

Arelenda

People have reported affliction to produce better results at low levels, others (myself included) have experienced 21/40 outperforming it on all gear levels. Some people have reported it to depend on the fight more than anything.

The compendium already answers this: it depends more on lag, raid setup, player skill and play style then on gear levels. Your mileage will vary.

Optimal raid setup has at least one affliction warlock because of malediction, shadow embrace and an imp, regardless of gear.


On the subject of scaling:
Note that 0/21/40 gets more benefits from upgrades to any stat (damage, hit rating, crit rating, haste rating) than affliction does.

 

sna

I believe, that for shadow bolt spamming (Had no time to do more math)
1% haste ~= 1% hit if hit<hitcap (Boss fight, three levels higher than level 70 player)

Input data:

• SBXI= 603 - damage of shadow bolt rank 11

• SC = 0.83 - spell damage coefficient

• BCT =2.5 - base cast time (SB with Bane, but calculations behave same way with 3 sec. cast time)

• FT = 600 - 10 min fight time (It's neutral factor, you will see it later)

• SDG = 1000 - spell damage gear(it looks like it doesn't have influence on calculations, neutral factor)

• SH - spell haste

• HR - hit rating

Formulas:

• MCT = BCT - (BCT * SH ) // Modified cast time
  DPS = (SBXI + (SDG*SC) ) * FT/MCT * HR / FT

First calculation:
16% haste rating, 0% hit rating

• SH = 0.16 (16%)
  HR =0.83 (83% chance to hit boss = 0% hit rating from gear)
  
  MCT = 2.5 - (2.5*0.16) = 2.1
  DPS
  

  [top] ( 603 + (1000 * 0.83) ) * 600/2.1 * 0.83 / 600

  
  566.376

Second calculation:
0% haste rating, 16% hit rating

• SH = 0
  HR = 0.99
  
  MCT = 2.5
  DPS
  

  [top] (603 + (1000 * 0.83 ) ) * 600/2.5 * 0.99 / 600

  
  567.468

But in DPS calculation given by formula DPS = (SBXI + (SDG*SC) ) * FT/MCT * HR / FT, there is one and only factor that has impact on result ( we still talk about comparison 1%hit to 1%haste in the meaning of DPS) the quotient HR/MCT.

I'll show you, that ratio HR/MCT (value that Shadow bolt damage is multiplied by) is about the same in n calculations for x Spell haste rating value and y Hit rating value, where x+y = z and x>=0, 0<=y<=16 (translation: ratio will be close to equal for n calculations if only you combine percentage x and y values that their sum always give z)

1)
16% spell haste, 0% hit rating, stats pool = 16
SH = 0.16
HR = 0.83
MCT = 2.5 - (2.5*0.16) = 2.1

Ratio1 = 0.83/2.1 = 0.3952

2)
0% spell haste, 16% hit rating, stats pool = 16
SH = 0
HR = 0.99
MCT = 2.5 - (2.5*0) = 2.5

Ratio2 = 0.99/2.5 = 0.396

3)
8% spell haste, 8%hit rating, stats pool = 16
(by doing this I show, that those values are equal in meaning of dps (rounding to 3 significant digits, rounding to 4 significant digits makes 1%hit = 1%spell haste + 0.001 )

SH = 0.08
HR = 0.91
MCT = 2.5 - (2.5*0.08) = 2.3

Ratio3 = 0.91/2.3 = 0.3956

So, ratio1 ~= ratio2 ~= ratio3.

I'm not saying, that this is a proof, as I said at the beginning, I believe, that those two values (Hit Rating and Spell haste) are equal (or close to) in therms of DPS and Shadow bolt spamming.

 

Spline

Setting SH to 1.00 ought to bring MCT down to 1.25. You can't make a spell instant cast via spell haste. Here's the correct formula:

MCT = BCT / (1 + SH)

Thus,

MCT = 2.5 / 1.16 = 2.155

Looking at your HR / MCT ratios:

0.83 / 2.155 = 0.385150812064965197215777262180

0.99 / 2.5 = 0.396

Looks like spell hit wins by a 3.7% margin. You're free to conclude that 1% hit and 1% haste are close to equal in terms of boltspam DPS. However, since 1% hit costs 12.6 item points while 1% haste costs 15.7, the economical choice is clear.

 

sna

I failed. Didn't noticed I have wrong MCT formula :/

Speaking of economical choice, items with spell haste tend to have slightly higher value of Spell haste than items with hit rating have value of hit rating.

I have listed items with Spellhaste using wowhead (epics only) and got 15 of them. On average they have 30.9 Spell haste rating which is 1.9696%.
There are 61 Epic items usable by warlocks, with Spell hit. On average they have 18.54 Spell hit which is 1.47%.

 

• frmorrison

Syphon's proc is basically a 6 second dot (that tics 6 times), so it will not eat imp SB charges, just like Corruption will not.

It is the best off-hand and maybe for Orcs getting two 5 expertise buffs may make it the best for MH, but if not Orc then Dragonstrike is best.

 

Silverstorm

Haste rating can be stacked until 100% haste is achieved, theoretically (though there isn't gear to support this). What you can't do is lower the GCD with haste. So MCT should be the following:

MCT = MIN(1.5, BCT * (1-SH))

His original calculations were correct for pure boltspam DPS.

Haste Explanation

You're assuming that 1 hit rating = 1 haste rating for itemization points. I haven't seen clear proof that this is the case yet. They do have different conversion ratios, that's been blue-clarified, but what I haven't seen is whether 1 hit = 1 haste on a piece of gear.

 

Trickykid

MAX, not MIN.
And while "pure SB spam" shows those numbers, there wasn't any inclusion of LT frequency, which favors hit.

 

• frmorrison

I calculated the item level of the items when spell haste started going on gear, and it was 1 hit = 1 haste = 1 crit.


Haste will be more of a benefit when caster items are made like a few Season 3 melee items (specifically the Rogue gear), putting spell hit, crit, haste, damage, stam and int and sockets all on one piece.

Currently the well-itemized haste items is the timed quest ring in ZA, T6 trash drop ring, and the Illidan Staff.

 

Kobal

You forgot the crafted bracers. Not that they are particularly well itemized (only dmg and haste), but since the alternatives suffer the same fate they are probably still the best caster bracers in the game, at least for nuke-heavy builds.

 

dakalro

To be honest, from my own experience, the difference between a token affli lock in a tank group and a destro in a support group (spriest + resto shaman + mana pots, if still needed) is so huge atm that it's not even worth having one just for malediction. When in need of an imp a destro lock (which could easilly have 1-3 pts in imp imp) can give up saccing for a fight and still maintain respectable dps, at least close to an affli one in the same position.
Removing Life Tap from a destro lock's life is simply the biggest dps upgrade and at this point it's quite easy, on some fights even possible to give up 1-2 pot cooldowns to destruction potions and still never need to Life Tap. That in turn translates to a huge dps boost that an affli lock can't get because of improved Life Tap and Dark Pact already getting them a lot closer to their max. And for my own experience it's normally 300+ dps diff, going up to 5-600 on a let's say Patchwerk fight where I got to a so so 2380 dps (he has no partial resists). And Malediction can't make up for that difference.

On another note, Tried Oomkin + Ele combo on a Gruul kill recently and would like to say I'd never give up batteries for crit, maybe just the resto shaman once spriests get another 2-300 dps boost. 8% lucky extra crit still doesn't come close to always getting 10% more dps time from Life Taps (average I'd say).

 

Katinsha

Reply to Dexterus (and our biggest destro promotor Arelenda):

I think you are overestimating the effect your talents have on your dps. I am affliction and my destro colleagues never trounce me on the dps meters, even while I am almost always in the MT group (so no help from others) and we run with an elemental shaman and two spriests most commonly. (At Archimonde at the moment, progression-wise).

Consensus is that one affliction warlock and rest 0/21/40 is best raid damage.

When looking at warlock damage, you should firstly ignore the damage done by CoD/CoA, since other locks will put up a non-damaging curse, skewing the damage done in favour of the former. Former is often a destro lock.

Secondly, destro locks receive more benefit from group buffs such as from the shaman (heroism, mana totem, wrath totem) and shadow priest (vampiric touch). Raid leaders know this and tinker caster groups with destro locks in it. Other locks often provide an imp to a tank group/are placed in a random spot outside the caster group. These buffs highly influence the ability and effectiveness of shadow bolt spamming. Not only for destro, also for affliction. You can't say: "oh but you have imp life tap and dark pact, you don't need it" since affliction actually requires more mana spend per second, seeing dots are mostly instant and bolts take 2.5 seconds.

Realise that destro locks in most circustances need (way) more healing to keep spamming shadow bolt. This may or may not be efficient for your raid group on a certain encounter. Like on Kaz'rogal and Naj'entus it seriously cripples you. Destro is just inferior on those healing-intensive fights.

Lastly, it simply takes more concentration (and skill! :P) to keep up DoTs and spam shadow bolt than it takes to spam shadow bolt alone. It is possible people get higher results due to the easier game-play destruction offers. When affliction locks seem to underperform, perhaps their dot uptime is low and/or their number of shadow bolts (which should not be lower than 80% of a destro lock).

In my opinion, most people see a jump in DPS because the game-play is that much simpler as Destro. Not because it is a hugely superior damage spec. It is better, I won't deny it, but not 400-500 dps with equal gear levels, except at the very high-end, where, you have to agree, dps is not a big issue anymore, since you farm the bosses at that stage. The relevant stages are when you still struggle on them (working on Vashj, Kael, Reliquary of Souls, insert other road-block).

Your own game-play (what spells to cast, keeping up most important dots, casting as many bolts as possible, when to life tap/dark pact, how to move, not to die, when to use what consumable, how to choose the best gear, how to tweak your build, etc.) is soo much more important than your talents, that generalizations like this:

are just false. Fresh destro locks with +6% to hit are absolutely terrible (on say, Gruul).

EDIT: I am aware that I have to change my name in the profile or people can't find me in WoW Armory. Recently changed it in-game.
EDIT: done

 

Dondarion

Has anyone noticed the shadow resistances of bosses in ZA is somewhat high?

 

Benafflock

I haven't noticed anything particulary significant, though I also don't have any WWS reports to back up such a claim.

 

dakalro

I actually still manage to get at least second on fight I live through even if I'm most of the time doing CoE. CoD just helps on pushing over to 1st

I never argued that you don't need spriest, as for ele shaman, I'd rather take a resto one anytime. Mana regen beats a possible 3% increase in dmg hands down, especially for me since each and every time I've tried to get some bigger numbers on real bosses I've just ended up on 18-25% crit, out of 32-33% buffed theoretical.


Absolutely FALSE. Since I don't Life Tap I don't overuse hp and Soul Leech is plenty healing on a 9k crit when I actually do Life Tap. With Healthstonex3 + Death Coil, it's enough.


Don't even try to go there, I've been Affli for ... almost 2y6m, kinda, and with quite good success rate even and as for DoT uptime, that's actually not even something that's worth mentioning, it's automatism at some point to keep DoTs up as much as humanly possible. As for myself, destro has one crappy part, a lot more bar watching intensity, opposed to watching DoTimer, latency on quartz bar is variable and one late (post 2.3)/missed (pre-2.3) SB costs dearly.

Have to agree on the first part, but that's actually only relevant for lower gear levels or poor Affliction warlocks.
The dps difference is there, especially at this stage since dps race is pretty much all that's left for fun, no wipes to learn stuff, no more need to grind or w/e. And when Affliction is lacking regen, I simply get your life tap time in extra dps by default. Add the possibility of using 1-2 destro pots a fight, on the side of 1-2 mana pots and it all adds up. At this point you simply relax, perform whatever you need to easilly since you've done it so many times and just roll over the dps meters.

Agreed here but don't like what you're implying and am nicely telling you shouldn't. It isn't that.


I'm sorry to say but you didn't read my post carefully ... I DON'T Life Tap, that's the whole point, the entire empty mana bar in 1 minute issue is simply gone, no more extra healing, no need to LT, DP, anything; and that gives a huge boost in dps that Affli already has part of it, through DP and Imp LT.

As for group buffs, Resto shaman and shadow priest are the best buffs, crit doesn't do much in terms of reliability of dps. It is my opinion that you should first increase your highest reliable dps then go for the lucky part with crit. I'd rather leave ele shaman and moonkin to mages, they actually have a lot more benefit from those than locks in terms of reliability of dps.

And when difference is 600 dps I somehow doubt it's CoD doing it. I've been affli myself and I love it but it just doesn't look in any way shape or form capable of even touching destro in terms of damage. I've been on the damage meters at start kinda behind the same lock that now is still affli. And even though he isn't quite affli geared (you know, better crit is still more ISB uptime and the SPriests even screamed at me for having fun with Incinerate on a trash mob ) I really doubt he's doing anything wrong in terms of maximizing his dps.
I've topped so far every fight but Anetheron and Azgalor, not every time but enough to know I can do it. And on those it's not gonna be locks topping them if melee has anything to say and casters have to kill infernals.

 

Katinsha

Realised it wasn't really a reply to your post, as more in general against the many people simply calling affliction as dead on these boards. So, don't feel attacked...

I do however like to check out the WWS from you before switching compared to after you switched. 600 DPS difference is huge, I like to see how this was made possible.

 

Trickykid

No one is saying it's dead -- you can do respectable DPS as many specs if you plan the fights around your build and know how to execute. At 3/4 tk 5/6 ssc I am able to push similar levels of DPS as affliction, demonology and destruction. The reason why I choose destro is that there are fewer conditions for or restrictions on my success and I am able to take better advantage of raid/group buffs without gimping debuff slots.

As gear improves, 21/40 continues to separate from the others. I am on the low-end of gear for locks on this forum, so for most that difference is more pronounced. While you can still say affliction CAN do respectable DPS if played correctly, you're on a board with people who are looking for ways to squeeze an extra 10 dps out of their lock. So if a build is even only slightly inferior, it won't garner a lot of support.