Latency has very small, yet existant, effect on the effectiveness of haste. If you figure out how much time you lose for every spellcast, then figure out the "effective haste" ((old time between casts / new time between casts) - 1, after adding latency), you'll see that you get quite close to the same benefit somoene with 0 latency would be getting, although slightly slightly less.

With latency being X, H being the haste (0.01=1%) and T being the base casting time:

haste gain with latency = (X + T) / (X + T / (1+H)) - 1

Note that X is the time lost and not your actual latency, so even with 1000ms ping if you can stopcast and only lose 0.4s on average between casts then X=0.4 and not 1. Basically if X isn't << T then your DPS sucks so bad no raiding guild will take you anyway, which should be enough to justify my X<<T assumption.

With X << T:
haste gain with latency = H - X*H*(1+H)/T + (a bunch of less significant stuff)

The "bunch of less significant stuff" is less significant because it includes (X/T)^2, ^3 etc which are significantly smaller than X/T if X << T.

As you can see your loss of haste dut to latency is very small and is estimated as roughly X*H*(1+H)/T.

For common haste ratings, H is also << 1 in which case you can estimate the loss of latency as H*X/T, so with 1% haste and 0.2 delay between casts on a 2.5s cast spell you lose approximately 0.01*0.2/2.5=0.0008 or 0.08% hatse, and are left with 0.92% haste where a player with 0 latency would get 1%.

Feel free to copy-paste this explanation wherever is relevant (assuming I made no errors).

You are using an old version of the spreadsheet, but there was only one fix made recently that you can do yourself if you don't want to download the sheet again.

Change the value of spell haste from the old value of 20ish to the new value of 15.77, and you will see haste beat damage for SB/Curse spam.

me and a frend are debating what produces more dps for a affliction warlock. does 3 points in malediction do more dmg than 2 points in CoA, considering i use sifon life, corruption unstable affliction and SB. i'm thinking using CoA is better but he dpes bring a good point in that CoS buffs all my shadow spells, so if anyone can help me out i would greatly apreciate it.

If you're the only shadow-DPS in the raid, CoA/CoD will do more damage than CoS. Check your local warlock spreadsheet for details. The point of CoS is for raiding, and being able to increase DPS for all shadow/arcane DPS.

ty for the info

Enough that there are 2 shadow users in the raid, CoS is generally worth using, which means as an affliction warlock you'll be expected to cast talented CoS which makes points in CoA pretty much wasted. TBH you'll never run with less shadow users in a 25-man, maybe in a 5/10-man and even then usually you'll at least have a shadow priest too.
And don't forget CoD>talented CoA on most bosses so there isn't much point speccing imp CoA anyway.

Again, you can check the spreadsheet to get numbers for whatever you want.

Which Blessings do warlocks prefer? The first Blessing is probably Salvation, but do you prefer Kings or Wisdom as the second Blessing (assume only 2 paladins).

Salv is obviously the first choice. Beyond that it somewhat depends on the situation; when standing and nuking the 50 mp/5 from wisdom means fewer lifetaps, but most fights have enough movement to allow for comfortable lifetapping and wisdom's contribution isn't exactly *massive*, amounting to one less lifetap every three minutes. Kings gives more health which is incredibly important in T6 encounters, as wel as boosting mana and crit (albeit by a small amount). I personally prefer kings for most things, but arguments can be made for wisdom due to it having a more direct impact on your DPS.

In order of importance:

1. Salvation. Only one that matters, really.

2. Kings gives you a higher health/mana pool and a bit more crit chance because of added intellect. Even for non-demonologists this one is usually the best.

3. Wisdom gives you additional mana, reducing the need for Life Tap/Dark Pact slightly. You get 50mp5 talented or about a Life Tap's worth per 2-3 minutes.

4. Blessing of Light improves healing done on you by paladins slightly.


Blessings are typically more important to other classes. Salvation is still very convenient, allowing almost full dps at all times when combined with Soul Shatter.

Can somebody shoot me a list of the t6 bosses where it's not safe to go above 100% of MT threat? I know you don't pull aggro on a tank n spank from range until you're at 130, but are there bosses where if you go above 100 you mess something up? I'm envisioning the tank missing a fear on Archimonde and him going after me and sticking, or something like that. Are there any bosses where secondary abilities mess up if you go above 100?

Fears after 2.3 should no longer cause this sort of things (although now I'm pretty sure they do although I hadn't done archimonde himself, on nightbane and even nef/ony back in the day it was almost obvious although not thoroughly tested). However I've noticed some boss (forgot where it was, wasn't in a raid) where he stunned the tank and went to the highest on threat (me) without me doing any damage between the stun and him switching targets nor me going anywhere near 130% (but was above 100%).

Overall a list of those bosses would definitely be very useful if any guild actually dared to test.

Your tank eat fear and you loot Archimonde anyway?

You can go "all out" at:
- Supremus, Kite Phase
- Akama, Shade Phase and on the mobs upstairs
- RoS, Phase One
- Gurtogg Bloodboil, during Felrage

And you can do some crazy amount of dps with very decent tanks (1200+ tps) on the common tank&spank bosses, too. An early Soulshatter should be enough for the whole time on these fights.

My first EJ post, though I have read for a long time.

I was wondering if we could delve into the diminishing return factor for haste. I am switching from 41/0/20 to 0/21/40 and I was building a gear set it. The aforementioned benefit from haste is most certainly something that you would want to include into and 0/21/40 set, but I don't want to stack too much and waste itemization points on something that I am receiving extremely diminished returns from. Especially since the majority of haste gear has nothing besides stats, haste, and damage. So how much is too much? As I said this decision was a recent one and still haven't played around too much with Leulier's spreadsheet regarding 0/21/40, figured I had planned on bringing it up anyway so might as well make a post early as I wanted some outside opinion as well.

This might be something that should be mentioned in the Compendium once some conclusive evidence comes about concerning haste's diminishing returns as it pertains to 0/21/40.

Again this could also be something that has been brought up in the other thread that I tend to do a lot more skimming in rather then attentive reading. If that is the case, my most sincere apology for being a waste of space. <3

ShadowSeer as linked in the compendium does this already. At my gear level it tells me

100 haste rating is about 5-6% extra dps output
100 crit rating is about 4-5%
100 damage is about 4-5%.

This is with a 0/21/40 build, 1300ish spellpower, 12% hit, 24ish% crit, 5% haste (and a Draenei Elemental Shaman)

I'm guessing that at around 10-15% haste you'll see equal gains from each. However, since the differences are so small, latency is probably a lot larger factor than any low amount of haste vs other stat tradeoff. Since casting mechanics are changing in 2.3 things might be different there.

Your mileage may vary.

The following questions seem to come up constantly and it may be useful to have answers or links to answers for them in the compendium:

Boss armor is provided in a link, but when it's ok to use Recklessness due to cleaves/etc (bosses and phases of bosses) would be helpful.

Debuff priority for all classes so affliction warlocks know when to kick immolate or kick another class member in the ass for using a stupid debuff would also be useful.

I'll try to search for the posts but they're mentioned/asked about so many times it gets tough.


Edit: Found a recklessness thread: CoR/amp/dampen magic

Raid debuff thread:
Raid Debuffs

Theron can cast doom on the MT if you go over 100%. This will give the entire raid about 50s of blaming you, before the inevitable wipe. (Of course, in the last minute of the fight, you can go over).

This is assuming Theron follows the same mechanic with Doom as Prince with Enfeeble. I didn't risk testing this, for obvious reasons.

Well, better him than me :P

I don't believe this is true, we've never had the MT get ghosted. Unless someone can show proof this occurs?

On Gorefiend, I did pull aggro once by going over 110%, I don't recall how it happened, but I think it has something to do with his casting of the Shadow Damage spell (where he changes targets) and when he rechecks aggro list going back to the MT.


There is no way primary aggro can get ghosted, although if someone pulled and then the ghost debuff's timer was up, it could happen.

Also if you go too far ahead, you can get ghost then accidentally clip his melee range for the threat cap when running off his platform to the ghost spot. This causes him to 1shot you then ghosts squirt out into the raid.

I guess this depends on your tank/raid position but i'm pretty sure everyone uses the same universal spot.

Here is me: The World of Warcraft Armory

The spreadsheet says I should be doing 1150 dps roughly, but I don't find myself doing that DPS in any of the fights I'm doing (except AOE stuff where I come close, but I'm pretty cautious on AOE so I could probably be higher on those).

Here are some sample WWS's from different kinds of fights (I don't get to raid much so I'm at the mercy of the schedule on the kinds of fights I get to go to):

Wow Web Stats - Solarian death (that's the kind of AOE fight I was talking about)
Loading... - Lurker death
Wow Web Stats - Magtheridon death (I think the guy at the top of the meters was on a cube right next to Mag, so he could DPS almost the whole time...my cube was at the opposite end of the room)
Wow Web Stats - Void Reaver death (note I was stupid, died once to orbs, once to aggro...very stupid)
Wow Web Stats - Alar death (this is me going nuts in phase 1, and returning to normal phase 2)

So there ya go, I was hoping to give a nice selection.

In summation, I get it that if I make less mistakes (like dying on orbs *cry*) my dps will go up, but is there anything you can tell from these that I could improve on? I just want to kick ass on the few days I get to raid.

Sheets only tell you your dps in fights like Al'ar P1 at most ... anything with heavy movement, target switching will lower your dps. Though if you compare spreadsheet with a fight like Teron or Rage, Mag P2 if you don't click, you get quite close values but you rarely reach them due to various reasons.

There's also getting a decent estimation of LT/DP frequency and regen from spriest, if you have one. In a fight where I don't pot/have batteries I can go up to 15 LTs, huge dps loss. If in best regen group and potting I pretty much never LT, max 1-2x depending on pot values and batteries' quality.

So if you want to compare, find a log/meter with a fight that's as close to stand still and nuke as possible, calculate a bit how much time you spent LTing then fill up the spreadsheet and see how you did.