PSGarak

No, it doesn't.

We still don't have a quantitative analysis of it. We know that your crit chance lowers, but we don't know the exactly mechanism behind it. There are at least three possibilities: subtractive penalty (-5%), multaplicative penalty (your crit rate goes down 1/10), and a revaluing of crit rating (and possibly int->crit as well) by the level 73 conversions instead of level 70. Note that the last two may actually resolve to be the same, minus an additive constant. No testing has been done, and the testing required to get a quantitative analysis would be prohibitively enourmous, compared to the qualitative testing to show that something was happening.

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

 

Fimotik

Following some controvery regarding Immolate in a guild that I raid an alt with, I had a look through a couple of months worth of WWS on my main (this lock). Theorycrafting is great, but it can be funny how things work out in practice.

By way of background, my guild is in the first half of BT, so I had access to WWS for all TK/SSC bosses, 4/5 Hyjal and 3/9 BT bosses. This toon is UA Affliction, so the results relate only to this build.


What I found was a huge variance in the DPS output of both Shadow Bolt and Immolate. Our raid composition is fairly consistent, 3-4 locks, 3-4 mages (all fire or frost), 1-2 Shadow Priests (mostly 1, and I excluded those raids where no SPriest was present).

Things that had a big impact: Miss rate variances between Immolate & SB, partial resist rate variances between Immolate & SB and crit rate variance between the two. Sometimes Shadow Bolt would crit at a rate 10% or more than Immolate, and the DPS increase in SB was huge (due to ISB procs). Conversely sometimes Immolate would crit much more than Shadow Bolt, making its damage per cast time value much higher than the SB average hit.

Where miss, partial resist and crit rates did not differ signficantly, Immolate had competite damage per cast time.

What I learnt from a fairly cursory glance through WWS was that whether to cast Immolate or not for an Affliction lock is a very complex issue, being greatly affected by one's own gear (crit in particular), the composition of the raid (number of warlocks, their specs and gear; presence and number of Shadow Priests; presence of a fire mage; plus other raid/party buffers like Moonkin, ele shammies etc.

For example, with several locks with low crit rates, ISB uptime is going to be low. Add a couple of Shadow Priests and ISB uptime falls even further. Assuming a fire mage is in the raid, Immolate would have to be very appealing. Start changing the composition, have 2 locks with high crit rates, 1 SPriest, a Moonkin in their party and no fire mage, then Shadow Bolt really gets boosted. Unfortunately, no simple, one-size-fits-all answer to this debate it seems.

 

Botvinnik

I've used shadow seer but that just tells you about your on SB, I'm looking for something that measures raid ISB uptime. Any idea?

 

Arelenda

Shadowseer tells you everyone's damage increase by ISB and how much damage they contributed through their ISB debuffs. I don't see what more you could possibly want.

If you just want a simple number that says %, you can look up the total raid shadow damage. It'll have (x% boosted by ISB) behind it. Divide x by 20% and there you have your uptime. For example, 10% damage boosted by ISB means 50% (=10/20) uptime.

 

Bolche

No it does not. If you know for sure that your crit rate is 3% lower against a boss, you can juste decrease it in the crit cell.

As Arelenda said, ShadowSeer is a very nice addon to measure you raid ISB effect.

As for more thorical approach, you can check the "Raid ISB" tab of my spreadsheet, it tries to factor SPs & warlocks cycles, and I beleave the results are pretty accurate.

Lastly, you can try my ISB uptime simulator.

 

Arelenda

There you have it. Either you get real data from shadowseer (which can be flawed due to the sillyness that is the Blizzard Combat Events System) or you get theorethical data from Leulier (which can be flawed if people suck, or on movementy fights).

I'm kind of hoping that 2.4 will allow me to rewrite ShadowSeer so that it posts 100% accurate data all the time. Not only when fighting enemies with non-identical names in your 30 yard range.

 

Eph

Are my numbers off some how or is [Brooch of Nature's Mercy] the best neck for the hit capped destro lock?

 

Krazen

Err, from the same instance, you get [Loop of Cursed Bones].

If you are hit capped, it is a tossup between that, [The Sun King's Talisman], and [Translucent Spellthread Necklace]

I plug in my stats, and the Nature's Mercy neck is about equal to those 3, but 2 of those have stamina.

With my gear, they are all between +55 and +57 dps.

 

Eph

On my sheet I've got:

[Brooch of Nature's Mercy] is 65.07 (counting the spirit as 3.8dmg)
[Loop of Cursed Bones] is 61.45
[The Sun King's Talisman] is 60.36

The stamina loss is really a non issue since I'm not using any gear with low or no stamina (Belt of Blasting, FSW, etc).

 

Krazen

Well, you have a rather curious gear selection if you are hitting 12% hit (with ele shaman) without those items.

My raids rarely run with imp divine spirit, but I thought it was a 10% return, not a 20%? But if it is you might be correct.

FWIW, though, I think the increased ISB uptime from Kael's or the RoS neck would put it close to on par with the Brooch. Plus with some regemming you could make use of the +hit on the latter.

 

Necrostar

Could it get better than this as a Shadow Destro Lock ?

70 Undead Warlock

Pop trinket for 2 sec bolts. I'm curious to know if BL would even reduce it under 2 sec ? Does BL and Haste stack ? Even if it doesn't I just can't see a better set of gear for 0/21/40.

 

Aliah_KT

Has anyone looked into the accuracy of WOW Webstats and/or the Warcraft character sheet's accuracy on critical strike rating? I've noticed that my critical strike rating is almost always significantly less on WWS sheets than what it should be according to my calculations. I know WWS isn't exactly gospel, but perhaps someone could explain the reasoning for the following:

The World of Warcraft Armory
(My Armory link, which indicates a spell critical strike rating of 26.03% [18.03 from Intellect and bonus crit rating, 8.00 from talents])

Wow Web Stats
(A WWS parse of me and 200ish Shadow Bolts vs. Dr. Boom, which indicates a critical strike rating of 20.45%)

I am wearing the same gear (except for maybe a slightly different damage trinket--no change in spell critical strike rating) on my character that I did during the Dr. Boom test. Two hundred seems like a reasonable sample size to see a relatively close correlation (+/- 2%) in critical strike rating, but this analysis seems to indicate a statistically significant discontinuity. If anyone has any insight on this, I'd really appreciate it.

Thanks!

 

galzohar

In short, expect big deviation in your crit chance if you're not casting over 1875 bolts or something like that.

As for the "always lower", make sure you don't "forget" the WWS parses where it was higher. People tend to quickly forget those.

 

Aliah_KT

Thanks for the statistics--very informative. While it does make me feel better about my numbers' variance, I'm still not 100% convinced we're seeing the whole picture. I'm sure it's only in my head, but I can't seem to remember a single time where I looked at my critical strike statistics and found the rate significantly higher (certainly nowhere near 5-10%) than what I expected. Perhaps I am just unlucky. It would be interesting to see if anyone has ever done a test on the order of a thousand or more spell strikes to more accurately gauge Blizzard's numbers. Seeing as how major a factor spell critical strikes are for our highest-DPS class raiding build (0/21/40), as well as other classes', I'd be surprised if there wasn't some more hard experimental data available.

Thanks again, and I'd welcome anyone else's input into this.

 

galzohar

Keep in mind that even if your crit was 5% less, assuming it's just 5% less but still you gain 1% per 22.1 and per point in devastation etc, your stat choices would be nearly the same. And it's probably not the case anyway - feel free to sit on DR. boom and shoot 2000+ shadowbolts and provide wws. And if lower crit on bosses is the case, then it must be small enough for people to not notice, and again a fixed reduction to crit will make very little change on how you look at stats.

 

Kalle

The tooltip already includes backlash. The reason why it doesn't include devastation is simply because devastation only increases your critchance with destruction spells whereas backlash increases your total critchance.

For a sample of casts the standard deviation for the critchance is
,
where is the critchance. So for and a sample size of 220 the standard deviation is as large as 2.8%.

 

• Emolate

Bloodlust/Heroism is Haste, not Reduction, so the usual rules about Haste apply.

 

Aliah_KT

Thanks for letting me know this. That makes a lot more sense and gives me a better feeling about being slightly lower with my critical strike rate. Being off by 2-3% does not bother me nearly as much as 5-6% for that many attempts.

 

Roywyn

You gem for hit, that means you're doing something wrong with your gear setup.
Get [Belt of Blasting] with spinels, [Mana Attuned Band], replace orange gems with [Potent Pyrestone] and get [Shroud of the Highborne] (or [Cloak of the Illidari Council]) for a ~15 DPS upgrade in the sandbox sheet.

 

Pentamorfi

I've seen a lot of people talk about using [Vestments of the Sea-Witch] over [Robe of the Malefic]. This would mean people would drop [Leggings of Channeled Elements] for [Leggings of the Malefic], in order to maintain 4pc T6.

I don't see the reasoning behind it. Assuming you gem every available gem slot with crimson spinels, you're giving up 30 spell damage for 34 crit rating. Even in a destro spec, 1 crit is worth about 0.7 damage, resulting in what looks like a DPS loss. What am I missing?

 

Necrostar

You fail to calculate the fact that I have haste items (5 items with haste) & there is no such thing as haste gems ... That's why I have hit gems.

Sure, I could put BoB on this, but if you want to get a decent haste rating your gonna have to sacrifice some items. The important thing is to stay hit capped , improve haste rating w/o saccing too much spell/crit.

 

Krazen

It's 27 damage, not 30. The Channeled Elements combo is slightly better for personal dps, but only by 1-2 dps for my gear. I took the Vestments so I don't have to go back to Hyjal 100 times for a high demand item.

 

Eph

http://elitistjerks.com/573011-post567.html

This post from the past also lists a bit more info about 4/5 t6 setups. I remember coming to the conclusion that [Vestments of the Sea-Witch] were the way to go, though looking back now I can't remember why. [Leggings of Channeled Elements] also have 5.6dmg worth of spirit, pushing it even further ahead.

It seems that there are many ideas about the best gear setup. This mostly comes from the cap on hit and the fact that there is so much hit available you have to choose which item slots should have any. I think I'm going to load up some numbers comparing total sets instead of just between items and see if I can find what the overall optimum setup would be.

 

galzohar

Improved Divine Spirit Rank 2

Your Divine Spirit and Prayer of Spirit spells also increase the target's spell damage and healing by an amount equal to 10% of their total Spirit.


Also don't forget somewhere around 80 int gives 1% crit, so int is even less neglicible than spirit for your dps, as with kings 1 int gives something like 0.3 crit rating, and when items have 20-40 int on them it's far from neglicible.

 

Ammanas

At some point you can have too much haste, because you're SBs will be incredibly hasted which will cause more mana usage but your life taps will still be at the 1.5 sec GCD no matter how much haste you have. The extra SBs you gain from the haste will be canceled out by the extra GCDs you are using life tapping. I'm not bashing haste - its a great stat and all 21/40 locks should have a healthy amount - but you don't want to have too much, especially at the cost of other stats.

I would personally never consider using the Supremus belt, the Anetheron one with its sockets is a much better belt (just so much more damage) and the Belt of Blasting is the only other alternative (if you need hit). You could even it out a little by picking up the Anetheron belt, the Najentus ring (or the ZA chest one if you prefer it still has haste), and re-gemming you're oranges as crit/damage gems.

As far as you're other question, yes Blood Lust works like all other haste and would stack with the trinket/passive haste.