Madlax

I have to once again intrude into the shadowy realm of priests for a bunch of questions that I couldn't find an answer myself to.
Given the following guidelines:
SWP(including muru) gear and lower available mana-battery-priest.
Something like chardev.org - A World of Warcraft character planner v.3.a (spell hit wise)
Shaman in group.
Default 14/47 spec.

Question 1
On a flat out DPS encounter, like Brutallus, would a shadow priest ever be able to go OOM?
How about an encounter without being able to use fiend, or no shaman in group.
Does any points in Focused Mind change that - or is that only a minor improvement?

Question 2 - so I get my thought straight.
A shadow priest swapping a random trinket for Skull of Gul'Dan and thus being able to put 1 spare point into Focused Mind does not receive any actual DPS bonus from this talent restructuring - does he?

Question 3
I saw that in the later posts - while I don't think its necessary but asking helps.
At which point is it actually possible to drop points from meditation?

 

• Snowy

1: Impossible to run out of mana on Brutallus with Sunwell gear, even with no shadow fiend. Shaman probably is irrelevant as well -- I've had an elemental so only mana spring.

2: If you were moving one point from Shadow Focus to Focused Mind, you would not receive any DPS bonus. (this is typically NOT the tradeoff you're making though)

3: Fully BT-geared and above you can consider it. Dropping a point from Focused Mind is better though. Consider that Mind Blast costs 450 mana and Mind Flay costs 230 mana. One point in Focused Mind saves you 22.5 mana per MB and 11.5 mana per MF.

Looking at my latest Brutallus: Wow Web Stats

I cast 39 MB and 62 MF in 5m29s. That's 8.43 MB's per minute, and 5.30 MF's per minute. Dropping one point from Focused Mind would cost me (8.43*22.5)+(5.30*11.5) ~ 250 mana per minute or ~21 mp/5.

Unbuffed I have 517 intellect and 305 spirit. Raid buffed would put me at (517+40+18)*1.1=632 intellect and (305+50+18+20)*1.1=432 spirit. My regen according to the calculator is 152 mp/5 while casting with 3/3 Meditation, so it's easy to see that each point in Meditation is worth a little over 50 mp/5. Dropping Focused Mind first is clearly the way to go if you still want to kep a 14/0/47 spec at a Sunwell gear level. Lower amounts of gear draw the a little closer but not by THAT much -- for example, someone with, say, 600 int and 350 spirit raid buffed would have 120 mp/5 while casting, making each point in Meditation worth 40 mp/5.

edit: Also for my own curiosity, if I use an Elixir of Draenic Wisdom, my regen while casting is 168 mp/5, for an increase of 16 mp/5 -- the exact same increase as an Elixir of Major Mageblood, but without the extra benefits.

 

Madlax

I sure enjoy accurate infos, tya
Going to the break even point - where would a priest start to struggle with his mana DPS wise?
1000?
1200?

I´ve only been looking into shadow priest itemization slightly but I'm still a bit confused.
I have, my own, shadow priests putting up the claim that Skull would be a major DPS upgrade for them - ignoring warlocks and mages which gain full benefit of it.
The gear link( chardev.org - A World of Warcraft character planner v.3.a ) is what I came up with after a few reads and made me wonder whether or not that could actually be true.
With a bare minimum of items I could reach close to 6% hit, in my example even bluntly ignoring rings or Illidan headpiece. I checked with shadowpriest.com - Skull is ranked 4th without spell hit and first with spell hit and wondered whether or not that hit would actually could be counted in.

So, should a shadow priest actually consider taking skull - given that he can take any other trinket(muru, hex, timbal) or would that hardly make any difference?

 

rooj

From what I understand the only reason to use the Skull is the very real possibility of being hit starved.

 

Anedris

In full Sunwell gear a shadow priest will be able to make full use of the Skull's hit and it will be either the best or second best trinket in the game for us. In my own opinion it's still better in the hands of a mage or warlock though (assuming both players are equally skilled etc.) - their DPS simply scales better than ours, and their sole job in the raid is DPS (whereas we're a mix of healer, DPS, and mana battery).

To be clear though the Skull is an extremely good shadow priest trinket and would not be replaced before Wrath.

 

cheebamonkey

Ok, this goes out to anyone who has a sufficient amount of SW/Haste gear. Back before SW came out I did a lot of simulations that showed scaling of dps with haste the result of which was that haste was worth about .9 SP. While I was doing it I had a fairly strong hunch that spell haste would cap out for us at some point despite the fact that simulations showed otherwise. In any case I can range my spell haste from 200-300 (i can go lower but i would be making stupid trade outs to do so) and from all the raiding I've done with this gear I can't see any point in going more than 240 on haste. For me and my latency (about 250ms on average) this allows me to get exactly 2 full MF's in between each MB. Whenever I try to go higher on haste I just find time i'm sitting there, not doing anything, or my spells come up at weird times.

So anyways my question is are there any priests running into the same thing or any who are able to make higher haste work? My concern is that once I have all my SW gear I'll have too much haste and will essentially be wasting points that could have been used better by the other casters in my guild.

 

tedv

I'm generally around 290 haste (18%) and I haven't had this problem. No matter how much haste you have, some spells will come up at weird times. I can tell that there would be fewer cycle conflicts if my haste rating were in the 265 to 270 range, but it's still a net benefit. Also this means that I should reach another "sweet spot" once I'm in the 320 to 340 range. The trick to using haste well is proper clipping of mind flays. Download and install the thirds texture for quartz (sorry, I don't have a link, but it was earlier in this thread). I noted a roughly 40 DPS increase between not clipping and installing the texture and clipping.

 

• Snowy

For every time you actually cast 2 flays back to back and have to wait a small portion of a second to cast MB, there's more times where you've cast dots back to back, etc., and the lowered GCD has gained that amount of DPS back and then some. Hopefully I'm being clear with what I mean there.

If there's less than 15-20 secs on VE and I'm at an awkward moment where MB/SWD are <1 sec from being cooled down, I'll usually take advantage to refresh VE then, rather than waiting until it's about to drop off and use a full GCD then that could be used on a MB or SWD.

 

cheebamonkey

That thirds bar texture would really help. Time to sort back through this monster trying to find it.

 

Fail44

I have been browsing through this thread and find it quite informational. I'm interested in getting some advise or just opinions on where I should go with my gear at this point. My guild is on Archimonde and we haven't stepped a foot in BT yet. On average I think I'm usually pulling 900-1000 dps. It depends on the fight. I am sitting at 1354 shadow damage unbuffed. But 0 haste. And around 135 spell hit. I've been pretty much focusing on getting my damage up, and was hesitant on how spell haste actually effects a shadow priests dps. Which is why i have none . My main question is, at this amount of damage should i start replacing some of it with spell haste? From what i have been reading on this thread the answer is probably yes. Maybe a estimated amount of haste i should shoot for to increase my dps in raids?

 

tedv

If you are averaging less than 1000 DPS with 1354 spell damage unbuffed, the problem is not gear. There is something you are doing wrong. You should be doing at least 1200 DPS.

For the record, estimate 1 spell damage = 1 spell haste and pick loot accordingly.

 

Fail44

Alrighty, thanks for the tip. I guess I'll just have to push my dps a bit harder. I keep my dots up, MF whenever MB is on CD and throw SWD in when I can. I don't think I'm doing anything wrong really, but I'm open for suggestions.

 

tedv

I would post a WWS and have us look over it. What exactly did you plan on doing to "push your DPS a bit harder"? In my experience, most people who are low on DPS aren't low because they know how to do better and just aren't doing it.

 

rooj

Without seeing a WWS, I will take a wild stab and guess that you are refreshing your dots too soon. I had no idea how much of an effect this had on my dps, until I started to pay attention. Refreshing SWP early can severely affect your dps. I have a little less spelldmg (about 1290 unbuffed) than you, but around 85 haste. On Hyjal trash I am consistently at about 1150dps and depending on the boss can hit around 1200+.

Actually, I think one of the reasons haste is so good for us, is that unlike other classes it doesnt' actually reduce mana efficiency. Remember, our mana regen is based on dps, and haste increases dps, which means the faster we do dmg, the faster we gain mana. At a certain gear level and with other consumables (oils, flasks), I find myself nearly never having to pot outside of either, 1) dying early, 2) having to cast a ton of dispels or PWSs. I would imagine that destro pots would be fine regardless of the length of the fight.

As for gear level, I am at about 1296 shadow dmg with 85 haste and don't have a need for supermana pots in normal situations as long as I use spelldmg food, weapon oil and flask.

 

• Snowy

When you're first getting to Council, Illidan, I don't think you can afford to use Destruction pots. Your damage and mana efficiency (and more importantly the fights take longer when you first encounter them) pretty much dictate keeping your potion timer free for mana pots. Even going back now, I occasionally have to pot on Council. You aren't going to have a huge abundance of haste at a BT-gear level so it doesn't really play into the equation. That's Sunwell gear, and once you get into the Sunwell fights, mana shouldn't really be an issue at all. I've used very few mana potions in Sunwell -- mostly at M'uru and even then it was only until I learned the rhythm/time of the fight to reliably Shadowfiend each time.

 

tedv

Due to the influx of "help me please!" questions, I made a "help me" thread for shadow priests: http://elitistjerks.com/f31/t26400-s...me_melt_faces/

So hopefully we can keep this main theorycrafting thread on track with, well, theorycrafting.

 

Fail44

I figure I'll work on my rotation and try an figure out what I'm doing wrong there. I do fine on the damage meters but my damage per second is low. I thought 1000 dps was pretty good ha ha, but you guys say it's low. And if I can get a WWS I'll try and toss it up.

Alright I'll take this over to the help me thread. Sorry for cluttering this one up.

 

◊ Asgorath

Threes texture for those that need it, was cleaning up my HD and found the TGA image. Note that I created the archive with command-line zip on my Mac, apparently people have had problems reading my archives on Windows in the past. Just let me know if it doesn't work for you.

Attached Files

threes.zip (1.5 KB, 614 views)

 

Missa

http://home.comcast.net/~mischievous...dowpvecalc.zip

Shadow Priest PvE Calculator
Missa, <Aftermath> Lightning’s Blade – US
Aftermath :: Lightning's Blade
theories@comcast.net

Requires Latest JAVA: Verify Java Version

About this Calculator
This calculator was designed to aid raiding shadow priests in making gear choices as well as more accurately predicting raid DPS than current spreadsheets available. The project was conceived after the haste change in 2.4, and the way that procs were calculated to go off seemed incorrect in the current calculation, so a hasted means to calculate the number of times a shadow priest can hit a mob was developed that also tracked what skill was hitting the mob. This calculator was not intended to be an attack at the current system, but a revision, and I feel like I’ve made a strong case for my approaches and hopefully this will show how I’ve come to my conclusions. This calculator was designed using JAVA for the ease of use of the GUI drag-n-drop interface, and it’s a simple enough program that C++ seemed overkill.

Calculator Functions
- Compute pseudo-damage value for items, trinket and non-trinket, so comparison will be more simplistic.
- Compute raid DPS based off the user’s base stats plus buffs chosen by the user to simulate a raid setting. Support for drums, bloodlust/heroism, and power infusion included to produce a more accurate haste calculation. DPS is calculated assuming the 1% natural resist rate and 20 unmitigated boss resist, using the formula from WoWWiki - Your guide to the World of Warcraft. Curse of Shadows and Malediction are also supported as well as ISB uptime being assumed at 35% (average crit rate of warlocks in Aftermath).

Calculator TO-DO
- Longevity calculation built into the raid DPS calculator to see if the user will expend more mana than returned/regenned/potted for a given fight duration.
- Threat generation built into the raid DPS calculator, assuming buffs like Blessing of Salvation, Subtlety, and the threat talents.
- User input for the ISB uptime calculation through input of the number of locks in the raid, their crit chances, and the number of other shadow priests present.

Formulas Used

Spell Haste
From WoWWiki - Your guide to the World of Warcraft

At level 70, 15.7 (as of 2.2) points of spell haste rating grants 1% spell haste bonus. As of 2.4.0 spell haste also influences spellcasting global cooldown.

New Casting Time = (Base Casting Time * 1570) / (1570 + Spell Haste Rating)
For the purpose of this formula, you should use the improved cast time from your if you have that talent, or the formula will not give the correct result.

New Global Cooldown = 1.5 / (1 + (Spell Haste Rating / 1570))
It would require 785 Spell Haste to bring the GCD down to 1 second.

DPS Testing was done to determine the equilibrium point where 1 point of haste was equal to 1 point of damage, and the approximate point that I found was 1570 damage.

Spell Critical Rating
From WoWWiki - Your guide to the World of Warcraft

At level 70, 22.1 spell critical strike rating increase the chance to critically hit with a spell by 1%.

(Intellect/80) + (Spell Critical Strike Rating/22.08) + Class Specific Constant
The class specific coefficient for a priest is 1.24 and this is used when determining the spell critical rate of the user as calculated from base stats + buffs.

Pseudo-damage Calculation
From shadowpriest.com Index page

5.57 crit rating = 1 dmg (with SW:D in rotation)
1 haste = 1 dmg (presuming ~1400 dmg)
10 spirit = 1.1 dmg (1 dmg without kings with imp DS)
80 int = 1 crit% = ~4.36 dmg (3.97 dmg without kings)
1 hit = 1.364 dmg (when not hit-capped)
hit capped = dmg + crit/5.57 + int*0.055 + spirit*0.11 + haste + sockets [+12 gems] + meta [MSD or ESD]
hit needed = "hit capped" + hit * 1.364
“CSD is rated at 9.8 dmg and requires at least two blue gems which would further lower the overall gain versus ESD. Additionally MSD is rated at roughly 19.1 dmg, but requires more blue gems than yellow, which immediately reduces the difference in most helms to satisfy the meta requirement. But it also rules out the use of Reckless Pyrestones to gain useful socket bonuses, so ESD is used for all meta slots.”

Set Bonus Calculation
From shadowpriest.com Index page

“Mefadin wrote:
Additionally, I did the math for set bonuses. Pieces are first calculated alone. Then if a bonus exists for a 2 pieces of a set, half of the bonus is added to the piece. If a bonus exists for only 4 pieces of a set, a quarter of the bonus is added to the piece. If both a 2 and 4-piece bonus exists then the bonuses are summed together and a fourth of the total is added to the piece.”

From http://wiki.shadowpriest.com/index.p...rinkets/Priest

+damage Trinket / Proc / Effect
55.0 Tier 6 – 4pc Bonus
31.8 Tier 4 – 4pc Bonus
27.2 Tier 5 – 4pc Bonus
20.7 Tier 6 – 2pc Bonus

The 12/0/49 setup was used to calculate these values as that build realistically includes all possible talents that will increase a priest’s DPS.

DPS Calculation
From Shadow Priest By The Numbers

--Rank 7 Mindflay --
580.8 = base damage
.627 = Spell coefficient
General equation: for 1300 spell damage per tick ((580.8+(1300*.627))*1.32825))/3=618
Which can be simplified to
y=.27760425X+257.1492
Y=Tick
X=bonus shadow

--Rank 10 Shadow Word: Pain--
1812.8= base damage
1.61 = Spell coefficient
General equation: for 1300 spell damage per tick (1812.8+(1300*1.61))*1.32825))/8= 649
Which can be simplified to
Y=.26519X+304
X=shadow damage
Y=SWP tick value

-- Rank 3 Vampiric Touch--
715= base damage
1.10 = Spell coefficient
General equation: for 1300 spell damage per tick (715+(1300*1.1))*1.32825))/5= 570
Which can be simplified to
Y=.292215X+199.93975
X=shadow damage
Y=VT tick value

--Rank 2 Shadow Word: Death --
629 - 731 =base damage
680= average base
.429= bonus Spell damage coefficient
General equation: for 1300 spell damage ((680+(1300*.429))*1.32825))=1644
Which can be simplified with a Crit model for 150% bonus to damage
Y=(1+.5(Z/100)( .56982X+903.20965)
Y= AVG Damage per cast
X= spell damage
Z= Crit %

--Rank 7 Mindblast --
782-827 =base damage
804.5= average base
.429= bonus Spell damage coefficient
General equation: for 1300 spell damage ((804.5+(1300*.429))*1.32825))=1809
Which can be simplified with a Crit model for 150% bonus to damage
Y=(1+.5(Z/100)( .569826X+ 1068.56754)
Y= AVG Damage per cast
X= spell damage
Z= Crit %

HASTE-MF
3(.27760425X+257.1492)/ (3-.00135H)
Y=DPS
X=bonus shadow
H=Haste rating

DPS is calculated based on the length of the fight, the user’s damage, and the user’s crit rate. Haste is used to evaluate cast times. Priority has been set to SWP->VT->MB->Death->Flay, allowing a flay to be cut short if MB or SWD comes up.

Features are discussed more in-depth in the included document file inside the program's zip file.

 

rooj

I am getting an error message, it is not letting me launch and the details informs me that I do not have a version of JRE installed locally, and then it makes an attempt to do so, but is unable to.

 

Caligula

I thought ISB uptime was quite a bit higher than 35%. Like 60% or so? I could be making shit up in my mind though.

http://sig.gamerdna.com/quizzes/INFL...s/Caligula.png

 

Missa

I wasn't completely sure, and polled my locks. I mean, they could be idiots, let's face it. I plan on making it so you can calculate out the ISB yourself to make it better for you. And @ the post about not working: make sure that you have the latest java installed (I put a link to it). You need at least 1.6 I believe, and sometimes Java is picky about that. I guess going for easy to make (since the coding is easy) has it's downfalls, but it autoupdates as Java so I dunno.

EDIT: Forgot to reapply VE in my calculation so it should be a lot more accurate now. Also, added a slider (yay!) for letting the user input ISB uptimes. I will now work on adding in latency and DPS time/presence. Please give me feedback on version 1.3.

EDIT 2:

To Do:
-- Reverse engineering calculator to determine what min dmg and haste you need to output a certain DPS (assuming, of course, perfect situations/casting sequence).
-- Pushback factor, calculated out like ISB where user inputs approximate pushback in the fight (think RoS/Gorefiend sort of thing).
-- Latency factor to increase time between casts (forces perfect sequence to be less than perfect)
-- Mana Efficiency and Threat Generation calculation.

 

Crepusculu

For you Sunwell priests, I made a haste table a while back for spell haste optimized around mind blast cycling
My theory is that spell haste plateaus around these values and doesn't follow a linear progression (such as assigning a direct spell damage value to haste). This is designed to match mind blast's cooldown and the number of global casts one can fit inbetween.

I just reached 246 haste, right next to a 5 GCD marker Has anyone reached near the 400 range and noticed any change in their casting? I'm simply wondering whether these are indeed magic numbers or if our varying dot rotation makes it inconsequential.

Addons
http://www.wowinterface.com/list.php?skinnerid=35267

 

tedv

Where did you come up with these values for points in improved Mind Blast? Anyone can spend up to 5 points if they can use them, just by stealing one point from Shadow Weaving and losing a point in 15% shadow crit. The real question is whether the fifth point for a 5.5 second Mind Blast cooldown will result in more blasts than a 6 second Mind Blast cooldown. Once you have even 100 spell haste, the answer is yes, you can cast two mind flays in a row in less than 6 seconds.

I guess I just don't understand why having 196 spell haste means you want to go up to an 8 second Mind Blast cooldown. I had this much spell haste for a while and 5/5 really wasn't a waste at all. Can you explain your logic?

 

Malleus

Crepusculu's rationale is to find the breakpoints where an exact number of GCDs fits into the CD of Mind Blast, so you're constantly casting. At 196 haste, 6 GCDs is 8 seconds.

The problem with this logic is twofold. First, it assumes you can react perfectly in zero time. This is manifestly not true for anyone. Second, it's engineering instead of reverse engineering. Crepusculu is thinking in terms of how best to maximise DPS output for a given value of haste, when what he should be doing is thinking about how much haste will maximise his DPS for a given value of Imp MB.

As for why Imp MB wasn't wasted for you: taking 2/5 Imp MB at 196 haste gives you 7s CD, which fits five GCDs of 1.33s each with .33s spare. Taking 4/5 gives you 6s CD, which fits four GCDs with .68s spare. Both of them give you leeway to react, and both will give more DPS than 0/5 anyway because you're casting MB more often.