I can see three methods to proceed, when hit-capped:
Either you try to write a whole cycle, and then it's enough to compute the expected damage of each spell, depending on haste / sp / crit / etc.
The main trouble are the length of the cycle, and the haste caps that will change this cycle. These shall be treated specifically (but haste caps are always tricky to deal with).

The other solution would be to ignore cd conflicts, and consider that each spell is cast on cooldown (or for dots, that we can assure 100% up-time). That will be an approximation, and I'm not sure it would lead to a good dps approximation. But I suspect that the scaling factor that we could derive (by taking the Taylor expansion) would be (up to a common scaling factor) quite similar to the real ones. Once again, the main trouble with such a model is haste, which will change the impact of conflicts...

Last methods would be a first order approximation of conflicts. Consider that you can deal only one spell at the same time, but that you can "interrupt and save" a spell to cast another more urgent spell, and then resume the first spell. That would be a fluid time limit, that you can imagine as "divide each spells as N spells, with damage and cast time divided by N". It just removes the "discrete times / spells / gcd" effect. Such an approach is a trade-off between both previous approaches : it predicts better results as the "ignore conflicts" approach (because it takes the expected "total time of conflict" into account, at a first order of approximation), but the cycle would be easier to compute, and be shorter (at least, I guess, in most cases).

Such methods could be used in BC for hunters, as long as haste gear was not available, with very good results. People went on after Hyjal / BT, fixing haste problems with different ad-hoc tricks (and with less precision in prediction). Hunters were pretty easy to model because you could ignore hit (next cast didn't depend on whether previous cast hit or not), and because cycles were fixed (before haste). Haste made it difficult when it was changing the cycle used.

Just saw Nevermelting Ice Crystal on mmo-champion. Looks exceedingly good as a trinket for fights where SW:P doesn't drop off.

Out of curiosity, is it known yet whether or not the upfront damage from Improved Devouring Plague would be counted as non-period damage by this trinket?

It's treated as non-periodic for reign so I assume it would count for this trinket as well.

It's PP value has been calculated at 237.97 assuming you cast SW:P on the first tick and keep it up (math on sp.com).

Considering it is available from a 5man instance, it will put this trinket at #6 (not counting the heroic version of the 2 new raid trinket since that will take a while). So yes, it is a very nice and easy to obtain upgrade for those still not using Reign of the Unliving until you can get Muradin's spyglass or the dislodged foreign object.

Nice replacement for all those sundials/illustrations out there!


#01 Dislodged Foreign Object (264) (Rotface25)

330 = 155 + 175

#02 Reign of the Unliving(258) / Reign of the Dead(258)

ToC: Anubarak / Heroic-25 / Value based on Thangrol's work
296.75 = 297

#03 Muradin's Spyglass (251) (Gunship10)

295.28 = 115.28 + 180

#04 Reign of the Unliving(245) / Reign of the Dead(245)

ToC: Anubarak / Normal-25 / Value based on Thangrol's work
264.46 = 264

#05 Flare of the Heavens

Ulduar: Vezax / Raid-25 Hardmode
260.15 = 155+120*0.88

#06 Nevermelting ice crystal

237.97

I will prepare a new OP sometime this week - if you have suggestions on what to include, please post here before I lock it/start the new thread!

It does 1.1k per hit, afaik.

One thing that is worth including in the new thread is telling people to turn on enemy name plates. The fact that they are visible up to 40 yards away now is incredible. It made multidotting during any fight with lots of adds incredibly easy.

To go with this tun off Overlapping Name Plates or you cant use the plates to select specific mobs.

I got a Nibelung and played with it for a short while before going to sleep:

Based on very preliminary testing (5 minutes with a dummy) it seems like it has a very very low proc rate, and no ICD. To see the real proc rate and to get a good uptime estimate I'll use it tonight in ToGTC 25 and post some updated information later.

Ok, I want to argue the need for Shadow Affinity in a Post 3.3 world.

With the nature of general ICC trash and Boss adds (even H Anub), I've found that with the significant DPS buff and Imp VE that I'm pulling 6.5k tps single target in 5 mans. I'm spiking 8k tps in 25 Boss encounters too.

I'm currently switching my 'general purpose' raid spec to Talent Calculator - World of Warcraft. I can't seem to hit Fade enough, and I'm hoping 8% will be enough to offset this BC like threat I'm putting out.

I realize this is tied into a tanking issue as well, but Fade in it's current form only allows for time for a taunt to be applied as it has no lasting threat drop.

At this point in my gear I'm only dropping 1/3 FM or 1/1 IF, but I believe the trade off to be worth it at this juncture of the patch.

Two things I'm most interested in hearing about is the interplay between Haste and SP and some info about Nibelung. I realize Nibelung might not be the right topic for the OP, but some facts/numbers regarding proc rate and expected longterm dps gain from the proc would be very interesting as I acquired the staff this week and the cool factor is through the roof!

It would appear as though the value of haste depends on how much haste you currently have, as indicated in this post/thread page.

Today (or in an upcoming hotfix), corruption haste-lock-in was fixed. No word yet on whether it affects crit/damage and SW:P in those regards. If it does, Nevermelting will be a below-average trinket and our opening rotations will finally be simplified.

I suspect SW:P => MB => MF => VT => DP will be the best starter because you get full shadow weaving and a higher incidence of SP procs (depending on your profs and trinkets) on your opening dots.

Something about the upfront damage of Improved Devouring Plague is bugged again, I have gotten multiple misses with hit cap (and the dot itself is still applied) since patch.

I've been unable to reproduce this on testing on the dummy.

I have 3/3 Shadow Focus (3% hit) (the likely culprit if there's a bug).
I have 299 hit rating (11.4% hit).
I made sure Misery was kept up (3% hit).
With all 3 combined that's 17.4% hit or 0.4% over the hit cap.

From 100 Imp. DP's I saw 0 misses. With Misery not on the target obviously I started to see Misses.

Well, I can give some initial theorycraft on that:

I got a Nibelung on Wednesday on our second ICC run, then used it on Thursday in TotGC:

WoW Meters Online - Combatlog Replay

Here is my damage breakdown, a few details about the proc:

My damage sucked on Jaraxxaus because I got a phonecall from a friend in the middle of the fight, sorry!

- The valk summon DOES NOT have an internal cooldown, and you can have multiple at the same time.
- The valk seems to stay up for 30 seconds or so.
- The valk can crit and miss. It misses a lot even with misery on the target.
- The smite spell is instant cast and hits for 1100.
- The valk does not have avoidance and dies almost instantly to any AoE.
- Overall, it did 2% of my DPS, or slightly more than the Anub trinket proc.

A better question will be, now that modifier lock in no longer exists will SWP ever be affected by haste? I know that we are in an alright spot right now, but it seems only fair to have all of our dots scale with haste now that a long standing bug has been fixed.

Well they already said that SW:P was removed from haste not due to the bug, but due to it when not using tricks to increase and keep rolling haste/crit, it was still providing too much DPS. And I believe a few of us when testing on the PTR came to the conclusion that the first implementation with SW:P hastened was just a bit too good.

Now that the first rounds of fixes/changes for the outliers are in, we can probably look in earnest where we stand, but it doesn't appear to be too bad right now. Of course I gladly take any boost, but I think we're mostly fine at this point (and excellent compared with last expansion's final raid)

That's interesting. Here are the DPS numbers for the Valkyr on various fights in that log:

Beasts: 135 DPS
Jarraxus: 171 DPS
Faction: 102 DPS
Twin Val'kyrs: 158 DPS
Anub'Arak: 87 DPS

The Beasts fight has enough downtime that the number is surely a bit low. Anub'arak has enough AoE and downtime that the value there also seems a bit suspect. Faction champs involves a lot of non-offensive spell casts, so it is similarly low. But it seems like this proc is a solid 150+ DPS on average if you can just stand and nuke. (And to be fair, all your spell power items are less useful in situations where you can't stand and nuke.)

So a static 150 DPS increase means the proc is worth roughly 100 spell power. Value accordingly, I guess.

Shadowfiend now will autocast shadowcrawl every cooldown (where as before only if it was oor of something or you clicked it manually), as well it seems to remember its behavior settings between summons. Even with the locking fix I can't see adding SW:P to haste scaling, it feels like it might be a little to much but who knows.

How is that a better question? That's up to the devs. SW:P haste scaling was nerfed because our damage was too high, not because of the bug.

I think it would be safe to assume that it's a possibility for Cata or a theoretical in-between patch, but right now we don't need it.