Yeah the difference in the grand scheme of things isn't particularly high. Still it's another bug that I've been pestering the devs about on the forums and PTR for a while now and I try to keep the Priest module in SimCraft as close to in-game as possible when it comes to little quirks like that. Not really for the DPS differences but more as a reminder of how the spells function in game.

Hmm I've downloaded the latest version of Rawr to do a sanity check against SimCraft results (There's been a lot of structural changes in SimCraft lately so we thought it a good time to compare results against various well known other simulators/models/spreadsheets for the various classes) but I'm having some issues.

Here's a link to the .xml saved file I used:
rawr_priest.xml

I basically took a sort of BiS T8 Priest set (though Rawr doesn't seem to support any of the Algalon items so I had to change some things) and then I stripped out any proc or other such items (changed Lightweave Embroidery to +23 haste, replaced Flare of the Heavens and IotDS with straight +spellpower trinkets etc). I disabled the 4 piece T8 bonus too by changing to the holy helm.

I then removed basically every single buff and also removed all the points from Twisted Faith and removed all the glyphs.

Basically I wanted a rather barebones profile that still had decent gear.
I also created a SimCraft profile to match (and checked the various stats both reported and they were the same or within 3 or 4 rating points etc).

I set latency in both simcraft and rawr to 0, turned off the shadowfiend etc. and made sure that the rotations pretty much matched.

I noticed a few discrepancies between Rawr and SimCraft and at least 1 possible Rawr bug.

The possible bug is that Rawr isn't updating the value for Mind Flay (or the tick amount etc) when I switch on or off Glyphs (in particular Glyph of Shadow Word: Pain)

The discrepancies were most obvious in the amount the spells were hitting for and in the overall DPS.
Rawr reported a sustained DPS (set both to 2 minute fights) of 3873. SimCraft reported 3366 DPS.
Now, I couldn't find an option to enable or disable Heroism/Bloodlust in Rawr so it's possible that Rawr was using it while SimCraft had it turned off.

As for the tick damage values etc, first here's the dump from SimCraft:
Copying to the clipboard in Rawr of the Stats page:
When I hover over the relevant spells in Rawr:
SimCraft has:
The crit chances are near enough to being identical (SimCraft has some variance of course as it's a simulator rather than a model and thus has more RNG).

What is noticeable however is the difference in Tick and Hit amounts as well as the boost Crits get compared to Hits/Ticks.

Rawr models VT, SW:P and DP crits as doing *2.045 the hit amount with the CSD meta gem (as opposed to 2.09 for MB and MF).
SimCraft has 2.09 for all of them. I jumped on the PTR again to verify how it works with VT and it is indeed 2.09, not 2.045.


As for the hit amounts themselves:

Both Rawr and SimCraft are using 2648 spell power and have the same talent spec and lack of glyphs.

For Mind Flay, the talents taken that have an effect on damage done are: Twin Disciplines, Darkness, Shadow Weaving, Shadowform, Misery.

The base damage of a Mind Flay tick is 196.
The base coefficient for a single Mind Flay tick is 0.9*1/3.5 == 0.25714.

This leads to a tick amount of:
(basetick + (coeff * spellpower * misery) ) * (1 + darkness + twin disciplines) * shadowweaving * shadowform
(196 + (0.25714*2648*1.15) ) * (1+0.1+0.05) * 1.1 * 1.15 = 1424.

Both Rawr and SimCraft are assuming a +3 enemy and SimCraft deducts a straight 6% off to account for partial resists against a +3 mob.

So deducting 6% for partial resists: 1424 * 0.94 == 1339.
Which matches SimCraft.

Even assuming that the coefficent was 1*1/3.5 (and thus didn't have the penalty it does) that would still lead to a tick amount of 1551 before partial resists and 1458 after partials.

The other spells are all hitting for different amounts too.

Could you share with me the formula and constants being used to calculate tick damage in Rawr?

shadowpriest.com hasn't suggested that Mind Flay would be around 25% of your overall damage since TBC. It's more likely to do between 35% and 45% of your overall damage.

The current T8 BiS set only has 361 haste rating, but that is more an artifact of what stats appear on the ilvl239 gear than the optimum balance. You'll see pretty decent returns on increasing your haste a bit to around 500 (depending upon your latency) before it starts to fall off again in it's relative worth.

This by the way is why the latest scale factors in SimCraft rate Haste so much higher than expected. It's because the baseline gearset used has relatively little haste. But once you bump it up a bit, the haste scale factor changes a fair bit.

You'll pretty much always see a DPS boost by increasing your Haste rating (the nonsense about losing DPS if you can't fit in an integer number of GCDs between MB cooldowns is just that, nonsense), but that's not to say it's a purely linear line (assuming fixed spellpower and crit values). It does have some steps in it.

So currently, the BiS T8 gear only has 361 haste, but there's no harm in aiming for around 500 haste (depending upon your latency and other stats).

Can't really give you a detailed answer right now Althor, but if you get the Rawr source code, the things you are asking for is in Spells.cs under Rawr/ShadowPriest. Note that the display of each spell in Rawr does not include debuffs or target based resistances. Also I believe I multiplied in Twin Disciplines instead of adding it onto Darkness. Metagem issue should be easy to fix.

Hmmmm the difference between * (darkness + twin disciplines) and * darkness * twindisciplines isn't much. around 6 or so damage after partial resists are removed.

Puts it at 1430 damage per tick before partial resists, and none of the damage is from debuffs on the mob (which would only increase the damage).

I'm raiding atm, but I'll try to give it a look over later.

I've taken a look through the Rawr.Priest module and here are the main issues I picked up (not all of them have any great impact on the overall result of course. Some are incredibly minor).

1) When calculating damage for a spell Rawr is multiplying separately things like Twin Disciplines, Imp. Shadow Word: Pain, Improved Devouring Plague, Darkness and the 2 piece T8 set bonus. All of these effects need to be added together before the multiplication.

2) In a similar fashion, for Mind Flay, Glyph of Shadow Word: Pain and Twisted Faith also form their own little "group". So with both the correct procedure is to multiply by (1 + 0.1 + 0.1).

3) The effects of Glyph of Shadow, Glyph of Shadow Word: Pain and Twisted Faith are not reflected in the display for each spell's damage etc. (Otherwise the correct values are used for the rest of the calculations as far as I can tell, just more of a display issue).

4) The reason why the tick amount for Mind Flay seems so high is because it's overridden the display function so that rather than showing how much it ticks for it's instead showing Dmg / Cast Time (with Haste factored in). In my example profile above there's 10.31% Haste and this is how much off what the actual Tick damage is. Again though this is just a display issue when you hover over Mind Flay on the stats page.

5) As mentioned earlier, with the CSD meta DP, VT and SW:P should all be seeing crits of *2.09 the hit amount rather than 2.045.

6) Also as mentioned previously, Improved Devouring Plague's initial damage component shouldn't be including spell power from spirt in it's calculations nor the boost from 2 piece T8.

7) Twisted Faith isn't checking that SW:P is up when it's giving the boost to Mind Flay and SW:P (Glyph of Shadow however correctly does check)

8) VE still has the old data. It now costs 0 mana, has no cooldown and lasts 300 seconds.

9) There's no dummy rank of Mind Blast for level 80. Most(all?) direct damage cast-time spells gain a small amount of base damage with each level the caster is above the level of the spell. The max rank of Mind Blast is 79 and at level 79 does 992 to 1048 damage (avg of 1020). However at level 80 it does 997 to 1053 damage (avg of 1025).

10) Mana cost reductions aren't quite handled correctly. The correct way is to first floor(basemana * spell's % of base mana) (the floor first is important). Then all of the mana reduction talents are applied as one additively (or subtractively I guess). And then the result is floor()'d again.

So for example with say, Shadow Focus and Focused Mind:

floor( (floor(BaseManaCost / 100f * BaseMana - stats.SpellsManaReduction)
* (1f - character.PriestTalents.ShadowFocus * 0.02f - character.PriestTalents.FocusedMind * 0.05f
- stats.MindBlastCostReduction)
)

11) There's no support at all for the 4 piece T8 bonus. In general it seems that haste rating change procs aren't handled which, yeah, removes support for the 4 piece T8 bonus. This can be worth a decent chunk of DPS.

Hope this helps.

Another thing I noticed (and SimCraft was slightly wrong here too, now fixed) from a bunch of testing tonight is that the coefficient for SW:P is neither 0.9*18/15 (what Rawr uses) nor 0.916666667*18/15 (What SimCraft was using) but is in fact 0.915*18/15.

Sometimes, just sometimes, I wish I knew *why* Blizzard chooses these seemingly random values.

Thanks for the information, did you agree on the values for shadowfiend? I think some of your base damage values have been victim of partial blocks, and might be lower than they should be.

And Rawr doesn't really have a timeline. I know what you are saying, but Rawr is not a simulation, so there is no actual casting, atleast directly. I do try to create a repeatable pattern (which is probably what you are asking for).

I only used non-blocked,absorbed,whatever,non-crit attacks when determining the coefficient.
I threw the combatlog (after some grepping of course) into OpenOffice and only looked at those full-hit attacks.

It's possible that I may be off still with the base damage range (but I got enough data points at both 0 spell power and with high spell power that I'm reasonably sure of it but yeah, I'm pretty confident with my scaling coefficient for the Shadowfiend.

I found my old results (but sadly looks like I didn't record my spellpower) and I've done some more testing tonight on the Shadowfiend.

The old results, even though I lack the spell power so can't use them to determine the coefficient, did still me a damage range. I only counted attacks that were neither Glancing nor Crits. For each attack that passed that test I totaled the hit damage + block amount + resisted + absorbed (as I was testing on the dummies the absorb amount was always 0 of course).

From 553 valid attacks the minimum attack was 1103 and the max 1137. A range of 35. With 553 data points that's probably a fairly accurate figure for the full range.

From previous testing at 0 spell power I had also seen a range of hits between 178 and 209 damage across 67 valid attacks. That's only a range of 32 of course so I was obviously missing some.

I've been testing some more tonight with 2564 spell power, again only counting non-crit, non-glancing attacks.
The range I've seen so far has been a minimum of 1149 dmg and a max of 1182 dmg. A Range of 34.


Going off what I know about the range at 0 spell power, assuming there is a spread of 35dmg from min to max, then either the minimum at 0 spell power is between 175 and 178 and the maximum is between 209 and 212 dmg.

However I also know that either 1148 or 1149 is the minimum when I have 2564 spellpower and either 1182 or 1183 is the maximum.

If 1148 is the minimum, then the maximum is 1182 and if 1149 is the minimum then 1183 is the maximum.

It also makes one of the following correct:

1) (175 + C * 2564) = 1148 and (209 + C * 2564) = 1182. Where C then equals 0.379485
2) (175 + C * 2564) = 1149 and (209 + C * 2564) = 1183. Where C then equals 0.379875
3) (176 + C * 2564) = 1148 and (210 + C * 2564) = 1182. Where C then equals 0.379095
4) (176 + C * 2564) = 1149 and (210 + C * 2564) = 1183. Where C then equals 0.379485
5) (177 + C * 2564) = 1148 and (211 + C * 2564) = 1182. Where C then equals 0.378705
6) (177 + C * 2564) = 1149 and (211 + C * 2564) = 1183. Where C then equals 0.379095
7) (178 + C * 2564) = 1148 and (212 + C * 2564) = 1182. Where C then equals 0.378315
8) (178 + C * 2564) = 1149 and (212 + C * 2564) = 1183. Where C then equals 0.378705

More testing will help me narrow things down better, but for now averaging the coefficients I get 0.379095 and a possible base range of 175 to 212.

I'll have to update the base damage range in SimCraft as it's obviously out. The coefficient is rather close to my previous figure of 0.3787 but I'll update it anyway.

*edit* Of course, due to rounding when doing the spell power tests it's possible the base damage range is less than 35.

Hi, OP feral druid here looking for some help so I can help my friend the shadowpriest.

Wow Web Stats

is the heart phase on our best try for XT Hard mode a few nights ago. I'd love to see my friend do a little more. How can I help him?

Armory: The World of Warcraft Armory

Well, for starters, he's about 41 hit rating over draenei hit cap and is still using hit gems.

In the 29 seconds of the log you linked, he didn't even start casting until 10 seconds in (which lost his illustration stack), and then cast VT twice in a row, and once he got his dots up and summoned his shadowfiend, the first spell he cast was mind flay instead of mind blast. If this is at all indicative of how he normally plays, I would say he needs to spend less time not casting and remember that mind blast is always higher priority than mind flay.

I have two parses from separate nights.
I feel that as an Spriest I'm not doing perhaps the BEST that I could be, sometimes I am outdmgd by the other Spriests, sometimes I am not. I understand that this could be RNG however if there is anything I'm doing wrong that is noticeable I would like to change it. I'm new to this class, and although I've done research I'm prone to make mistakes. Help would be loved. A message to me, a post on here. Help is help. I appreciate it in advance.

Wow Web Stats
Wow Web Stats

I apolologise. Hadn't seen it was the WWS forum. Moved topic to SP basics

I am a shadow priest in need of help. I have read alot on what spells to keep up and when to use others. My DPS has not gotten any higher. Here is my armory The World of Warcraft Armory.

For DPS:
For my normal rotation i usually start with VT (because it takes a bit to tick), MB, MF, DP and SWD to get illustration as high as i can and have shadow weaving at 5 then i put on SWP. I watch to ensure to reapp SWP when other buffs come for instance elemental oath. I use MB, VT, and DP every CD. I have found out that with or withoug clipping MF my dps stays the same (except on bosses like Vezak). I have an old world of logs to show you, my rotation changed just alittle and my dps increased because my gear got better. World of Logs - Real Time Raid Analysis

What is the main stat that a SP should stack? I know some SPs that stack crit and do insane damage and SPs that stack SP and top the charts. I have tried about everything and will try anything to increase my dps (even thought about gettin spell pen gemss and try). Please help when you can.

Taking your Hodir parse, you have 118 SWP ticks but only around 75 VT and/or DP ticks... there lies the majority of your DPS problems (the other being the low% of total damage MF is as opposed to other priests)

It doesn't matter when you cast SW:P relative to Illustration stacks because spellpower is the one thing that always updates on mind flay.

I was lead to believe MF only renewed SW:P. By this i mean when you cast MF your SW:P continues with the same buffs as when you initially cast it. When you recast it yourself SW:P gets all the new buffs that you currently have.

Here is the log from last nights raid World of Logs - Real Time Raid Analysis. It still seems that i have an issue with the DOTs so would it be smart to interupt the cast on MB to reapply the DOTs?

Heya,

Long time stalker, first time poster(so appologies if my post lacks eliteness).

This is interesting as i was pretty sure about the same. The alternative would open for new rotations. Could someone imba comment on wether or not MF renews SW:P with the old stats(when first cast) or with the new ones once renewed by MF? (Nevermind the illustration trinket as i get the part about it using my current SP at each tick)

On the topic of rotations, im somewhat uncertain after a late night experiment on IF dummies at 5am tonight.. My stalking EJ experiences tells me never start SW;P without having SW5 up - yet, i'm phailing badly at beating the DPS i get from Rayrifes rotation on teh tube; YouTube - Spriest DPS test

He's using VT, SWP, MB, DP, SWD, MF, MB, MF, Pet. Which in theory to me is a bad rotation seeing as both SWP and DP start rolling on bad shadow weaving stats and would need a GC later on to buff up. Still 4.1k after 30sec with his stats seem imba to me.. ?

VT, MB, MF off the bat gives SW5 which is the sweetspot to start rolling out the dots. Shouldnt this be beating his 'weak' dots or do the rolling dots while building SW really make up for the lack of dots for the first 6-7 secs of the opener?

Granted, it was 5am, i had had some wine and wasnt doing too extensive research. But! could somebody still be kind enough to put me out of my curiousity-misery by telling me what rotations i should be opening with on a long fight like patch and a short single-target 65k trash in naxx for instant max burst dps?

And, if you're really feeling mercyful, feel free to comment on what the best stat would be to focus on tonight, when im at 2250sp (IF buffed), 21%crit, 420haste and 270'ish hit. Something tells me i badly need more SP.. ?

I'm really getting sick of people having to repeat this constantly.

Mind Flay currently, and has always, applied your current spellpower value to SW:P when it refreshes the dot.
Examples of this include Illustration of the Dragonsoul stacks, the spellpower portion of Wild Magic pots, the spellpower portion of buffs in the Totem of Wrath category, Lightweave Embroidery, and various spellpower trinket procs and on use effects. It also includes encounter-specific buffs such as Shadow Crash in General Vezax.

Mind Flay does not, however, refresh the crit rate of SW:P or any percent-based damage buffs applied to your character. Your chance to crit a mob appears to be totaled at the time SW:P is cast, and then never recalculated for the spell unless it is specifically recast. This means if you gain crit increasing buffs like the Moonkin Aura category, or if the mob gains crit increasing debuffs such those in the Heart of the Crusader category or Scorch category, then SW:P must be recast in order to gain the increased chance for its ticks to crit from these buffs/debuffs. This also includes things like the increased crit rating from Wild Magic potions. As for percent based damage buffs, the primarily notable one is Shadow Weaving. Other than Shadow Weaving, the only other percent-based damage buffs of this type I can think of are encounter-specific, and include things like Rune of Power on Iron Council, the charge buff on Thaddius, and the sparks on Malygos.

The flip side to these two stipulations is that a new Mind Flay cast will NOT reset the critical strike chance or percent based character buffs of the current SW:P. This allows you to do things like roll short term effects like Rune of Power buffs or Wild Magic crit chance buffs for the rest of the encounter (or at least until the mob dies or something causes SW:P to fall off like a phase change).

Effects that cause the target to take increased damage are automatically applied without needing to recast SW:P or even cast Mind Flay. These include debuffs in the Curse of Elements category and encounter-specific buffs like the Singed debuff on Hodir.

So rule of thumb: Wait for 5 weaving stacks and all raid critical strike increasing buffs to be present before initial SW:P cast. Shadow Greenlight is a useful mod for this (although it tends to bug out when multi-dotting). In general, only recast it when your total critical strike chance against the mob increases for some reason, usually due to drinking a Wild Magic potion or some encounter specific buff. Also recast it when there is an encounter-specific damage increasing buff on your character. DO NOT recast it when a trinket procs, when you get a full Illustration stack, when you use a clicky +spellpower trinket, etc.

The current assumption, although I haven't seen this actually tested, is that buffs that increase critical strike *damage* (the only significant one in current content that I can think of is Hodir's Storm Power buff) are not updated on SW:P unless you recast it. So until proven otherwise, recast SW:P when you gain Storm Power.

I'm really really sick of seeing people misunderstand how this works. So many times in Archavon or Naxx pugs I've seen spriests with anemic SW:P ticks due to not understanding these basic concepts. If I can stop one stupid question about SW:P/Mind Flay refresh interaction with this post I'll be happy.

Edit: In case anyone refers to this post in the future, I confirmed Shadow Crash is a percent-based modifier that requires a recast of SW:P, and not a modifier on your current spell power like I thought it was.

Can this huge, amazing and wonderful quote please be added to the Shadow Priest Basics thread - It is a fantastic round up that SPriests who feel the need to post in but not read this forum should come across before asking stupid questions.

Hey everyone, this is actually my first time posting here, and I have to admit, I'm pretty nervous!

I'm main specced Holy, and I usually offspec Shadow or Discipline depending on the encounter, but lately I've been Shadow for our 10 man Algalon attempts (and kill!) to Dispersion all of the Big Bangs. I play Shadow on occasion, usually on Hodir or when we have a general excess of healers, but not enough to have mastered the spec yet.

My question is...my gear is quite good (I've logged out in it, and will continue to log out in it for the next few days), but yet I still can't keep up with most other hybrids or Shadowpriests in DPS. I'm fairly certain I'm using the standard rotation, though I know my haste rating is slightly lower than I'd like it to be, and I gem properly...I just feel like I'm doing something terribly wrong and I can't figure out what it is (is it really the haste?).

Here's a couple of my DPS World of Logs parses; sorry I couldn't find one with a second Shadowpriest for comparison, it's not often I DPS, let alone with another facemelter.

One night of Algalon 10 attempts: World of Logs - Real Time Raid Analysis

Our Algalon 10 kill: World of Logs - Real Time Raid Analysis

Edit: Also, the Dispersion glyph is only there for Algalon. Normally, the Mind Flay glyph would be in its place.