Nibelung is behaving like Bryntroll in that it benefits greatly from talented side-effects on spells. I did a few hours of testing on target dummies, starting with a talent wipe and adding talents one by one. Untalented blizzard didn't seem to proc the Val'kyr at all (it might have had a 1% chance to proc on cast, but I never saw that). Putting 1 point in Winter's Chill gave a 0.3% proc rate per blizzard tick (so each chill had a 1% proc chance). Adding 1 point to improved blizzard adds 1% proc chance. I'm usually seeing a 2.3% proc chance per tick on target dummies, but I don't have a good explanation for the 0.3% (1% is from winter's chill and another 1% is from improved blizzard).

Analyzing data from our Anub25H kill, the number of procs matches a scenario where frostbolt and fireball spell hits have a 1% chance to proc and blizzard ticks have a 2% chance to proc (per target). This encounter and the talent wipe testing were last week, so before the hotfixes. Winter's chill does not seem to increase the chance of frostbolt to proc the staff.

Given that the increased proc rate on Bryntroll for paladins and death knights is now considered a bug, I assume that the increased proc rate on Nibelung will also eventually considered a bug. Right now it is a good weapon for AOE encounters and frost spec, but there's no way to tell where it will end up. If they "fix" the side effect procs, but don't compensate with a proc chance on untalented blizzard, it could end up being totally useless for AOE.

Open questions:

- Does Blizzard want the proc chance to increase linearly with the number of targets that AOE hits? (Possibly not, but this is currently a major part of the fun factor.)
- Does Blizzard want the staff to proc on DOT ticks? (Apparently not, since DOT ticks are reported not to generate procs.)

If the damage contribution is intended to be about 3.5% of overall DPS, a single target proc rate of 1% is not sufficient. 1% chance to proc 31k damage equals 310 damage per cast on average. Using a 2 second per cast average, that's 155 DPS, which is under 2% if the overall DPS is 8000.

These "fun proc" weapons will be much harder to balance for Blizzard than the more boring "stat sticks". However, they add a lot more flavor to the game, so if Blizzard learns how to code these items properly, it should be good for the game. I like the fact that I'm now carrying one weapon for single target use (heroic Tarasque) and another for AOE (Nibelung) and can even switch mid-combat if desired.

My modeling shows a 4.1% (387) dps increase with the proc.

Without the proc RAWR shows my dps as 9014. Editing Nibelung to 1% chance on spell hit to do 26,000 holy damage; increases my dps to 9401, still behind Mag'hari but better than anything else on live with my current gear.

This contradicts what I saw in Rawr regarding the staff and I'm curious what we're doing differently.

I put it in as a 1% chance on hit to do 30k damage. I based this on full raid buffs and assuming the val'kyrs survive to do full dps for their full duration.

I had the staff come out around the same as the 245 ToC25/ToGC10 staves (ie Cold Convergence). This seems lackluster for a staff 1.5 tiers above the others of similar performance.

Are you setting it for one stack?

When I enter everything but don't set it at one stack then it doesn't change the dps on the weapon at all.

I used the following settings:

Trigger: Spell Hit
Duration: 0
Cooldown: 0
Chance: 1%
Stacks: 1
30000 Fire damage

Adding the proc changes the weapon DPS from 1140 to 1368, an increase of ~230 dps. This was done being hit capped not using a weapon.

As comparison, these are some of the weapons around it:

Enlightenment (245): 1394
Mag'hari (251): 1393
Cold Convergence (245): 1330

-- Edit

Someone pointed out that it's shows as much better for Arcane spec vs Fire spec (which I'm currently using). This would make sense assuming that each missile in Arcane Missiles can proc the Val'kyr. More events give more chances compared with fire which is a slower cast event spec.

Im fairly certain if you set it like that rawr will think every wave of arcane missiles can proc the valkyr, hence giving it a lot higher dps value than it really has. I couldn't manage a proc from am but the initial cast on dummies so far. I used 30k dmg per proc as well and it doesn't shine as good for fireball than it does for arcane for example.

Also tried it with imp bliz+fbite+wc spamming it on the 6 target dummies in theramore and getting up 5-6 valkyrs at once is really common. Only fight i found it useful was gunship so far where you can rocketjump to a place where rockets dont kill them 5 seconds after they spawn, they tend to fly over the enemy ship tho and get killed in seconds. The staff needs a lot more changes to be close to viable for dd at the moment in my opinion.

I've found that, in order for the expected damage per hit to be corret, I had to input a proc value below the actual damage a Valkyr might do. The default handling for a "Holy Damage" proc seems to include talents and crit chance. If I want my proc to actually do 30k, I need to input 20,5k as the proc value. Or should I select another type of effect?

There are also a few reports of it proccing from Living Bomb ticks - presumably, only those that crit, as they "cast" an Ignite, although it would be nice if someone could test (I don't have the staff, or I would). Would Fireball also get a double chance at proccing when it crits, and Scorch, a triple (1x Scorch, 1x Improved Scorch, 1x Ignite)?

That is correct, you should enter the lower value (and yea I forgot to mention that the value entered should be noncrit; crit, talents and debuffs are all applied on top of what you enter).

Good point Kavan. I was being overly optimistic on my estimate of total damage done to try to get an upper bound on how good the staff could be. Reducing the base damage will only make the staff worse in comparison to the other weapons available.

Since my goal with my rough estimate was to verify that I don't want to replace my current weapon, having a rough upper bound still below my current level was all the information I needed.

Nibelung:
To account for Rawr triggering a proc chance 5 times per AM cast (instead of 1), and to account for holy damage procs being influenced by talents/buffs, I had to use the following settings to simulate 27k dmg/proc, 1% procrate:

Equipment Effects / Holy Damage: 18300
Triger: DamageSpellHit
Duration: 0
Cooldown: 0
Chance: 0.58%
Stacks: 1

This is only to give a general idea for Arcane. Each person will need to set a different damage value and change %, depending on their gear/buffs/spec.

It would be nice to see Nibelung modeled in Rawr correctly though. All of this damage guessing is really giving everyone different results, especially since people are entering the proc itself differently. When I tested Nibelung on a target dummy after the latest hotfixes, it was getting 16 casts (same as before) for ~1650 damage non-crit. Its crit rate is exceedingly low -- something like 5%. There is a very informative post on wowhead regarding it, and his comments definitely mirror the results I've seen using the weapon, and the results TigaFin has seen.

min / max damage is 1591 / 1785 (old 1061 / 1189), crit multiplier is 1.5

That would give an average of 1688 non-crit damage, against a target without debuffs (IE, CoE). That means input into Rawr should be between 26,400 and 27,008 (depending if you want to use a flat 1650 or his 1688). Questions arise though... what crit rate is Rawr using for the proc, if any? What crit multiplier is it using? Is it applying buffs to the damage -- because Valkyr Smite benefits from things like CoE, but not auras like Sanctified Retribution (or its counter-parts, Ferocious Inspiration and Arcane Empowerment). How is the proc handling things like Living Bomb? I know that Valkyrs cannot proc off the Living Bomb DoT -- only the explosion.

Also, his post is indicating that they do not proc off each Arcane Missile -- only the initial cast. Has anyone verified this? In other words, has anyone started a cast of Arcane Missiles and actually seen Valkyr spawn after, say, the fifth missile hits?

If you model Nibelung for a single target, you have to remember that the damage is done over 30 seconds, so you could model it as a DOT effect. Any procs during the last 30 seconds of the fight will be less powerful because you do not benefit from the full duration.

If you have multiple targets, the Val'kyr will switch to a new target when their original one dies, so the DOT model isn't accurate. There and the single spike model is better.

Some information about [Dislodged Foreign Object];

* ICD is 45 seconds
* When it procs you gain 2 buffs the first is Surge of Power - Spell - World of Warcraft and the second Surging Power - Spell - World of Warcraft, Surging Power is the one that stacks (to 10).
* It procs of of spell casts; for that matter the beginning of the spell cast is the trigger, each AM tick cast can trigger it.

I'm not sure about the proc rate, since the above information is from parsing worldoflogs, but it's reasonable to assume 10%. In general, the uptime of the buff during a fight seems to be around 40%.

To model it in rawr you can edit the trinket and add a special effect of (105 + 1050) / 2 = 577.5 spellpower for 20 seconds 10% proc chance and 45 seconds ICD. It is around 30% better then normal Reign for Arcane.

--Rawr's evaluation agrees with 40% uptime since the numbers match almost exactly the value of 155 haste + static 230.8 spellpower (577.5 * .4).

Uptime for [Dislodged Foreign Object] should be 20/(45+1.75*(1/10%)) = 32%. (Proc Mechanics)

With my Arcane gear and the Rawr settings you've entered that's 10.6% 14.1% better than heroic Reign, and 28.1% better than normal Reign.

Edit: Kavan, you're right, the proc formula above is overly simplistic. Simcraft gives me 40% uptime over a 5-minute fight, which agrees with Maje's testing, and with my t10 head/shoulders in Rawr, the 264 Object is a 174 and 98 dps upgrade for the normal and Heroic Reign respectively.

1. The uptime is closer to 40%, that's from real logs, saying it's 32% is pointless.
2. Either the proc rate is higher then 10%.
3. You didn't take into account the fact that each AM tick can trigger it, which significantly lowers the average cast time.
4. There are various other haste effects that you haven't taken into consideration.

* It is only around 20% better then then heroic version (mixed up normal and heroic reign's since I am using both) and 30% then the normal reign, again the gearset I used uses 2t10 and has around 200 spellpower and haste more then yours. But the exact value would change for each gearset, the certain thing is that it's better then previous tear trinkets.

EDIT: I took a log of around 5 minutes of DPS for arcane mage, it lasted for 302 seconds and had 275 spell cast events, which gives a value of 1.1 for [v] and and an uptime of 37% (assuming the same 10% proc chance), which is much closer to what the logs show.

EDIT2: Just to make sure I ran simcraft with the same gear used in rawr; swapping in/out normal reign with DFO; the uptime for it's buff is 40.6%, the average time between procs is 50.6 seconds. The trinket's value is 40% better then the normal reign.

This uptime calculation is very simplistic and only gives you the long term trend. When you calculate uptime on a limited time duration (which is what Rawr does) you'll get a higher uptime since you start off cooldown.

As far as I can tell this is because Rawr applies 10 Spyglass stacks on the first damaging spell hit. It appears to work the same with [Eye of the Broodmother].

Edit: Meaning they're modeled the same in Rawr... not that they behave the same in game.

Useless post, delete please.

Actually the two trinkets are worded quite differently. Eye is listed as "cast" a damaging spell, which allowed the user to stack it up with Arcane Explosion, prior to a fight. Spyglass says on damaging spell hit. They likely should be modeled differently, unless the spyglass is behaving LIKE the eye, and stacking on spell casts. Hell, they fixed the eye to get a stack with blizzard, and now it seems to proc off casting Slow Fall still.

I have the 10man Spyglass and it is proccing on Spell damage hit, which can get awkward at times with spells with travel time; you loose the stack, although the spell has already left your hands.

I can confirm that they behave different. The Spyglass gains a stack when your damaging spell hits the target. Several hits, even simultaneous ones, will gain several stacks.

Just swinging by from the Moonkin forums:

1688 mean damage before CoE is what I've seen after some testing. I get just about 1900 per hit on a level 80 dummy with Earth and Moon (1688*1.13 = 1907). It it not buffed by Sanctified Retribution or Moonkin Aura, but it does gain the Totem of Wrath crit effect and/or Imp. Scorch (at least, something is making it crit on bosses, even though it doesn't on a target dummy).

It is affected by level-based partial resists (how much is that again? 6%?).

I get a 1% proc rate on Wrath as Resto spec, but over 2% as Moonkin spec, and over 3% as Moonkin spec with Starfire. So various odd things do proc it. In our case, I think it might be Earth and Moon, and maybe Glyph of Starfire.