Riot

There's no reason not to use Conc Blow/Shockwave for pure TPS unless you need them to deal with specific situations, such as spiderlings on Maexxna, to use an example.

Whenever an S&B proc lights up though, you should take advantage of it. Conc blow is massive threat, and Shockwave has an extremely generous AP modifier that let's it do very good damage, for less rage.

However, I'm not sure if Shockwave is "high" threat. Sorry if I missed that?

Overslept, / So tired. / If late, / Get fired. / Why bother? / Why the pain? / Just go home / Do it again. - The Commuter's Lament
Norman B. Colp

 

Ultramax

WoW 3.0 Threat Values - TankSpot

Unless you've got more recent testing to show otherwise both Shockwave and Conc blow are damage only, no bonus threat.

 

Rustyboy

Yes they are both damage only.. however there damage is based on 75% of AP.. so as your AP increases so does there damage and threat.

A Full Naxx Clear

Shockwave 2.5k avg on bosses
Concussion Blow 2.4k avg on bosses
Devastate 720 avg on bosses + 200 (%5 AP) innate threat from AP.

Which one would you choose for single target threat when SS/REV/SB are not up?

Even Thunder Clap averaged for 755 damage on bosses with its 1.85 threat modifier your looking at 1396.75 for 1:1 dmg/threat which would be better than a Devastate as a direct comparison. Again Heroic throw was averaging for 1.4k damage + 1.5 threat modifier and your looking at an extra ~600 innate threat.

The only issue is whether using Devastate at a particular time is worth doing over other abilities knowing that S&B procs are going to enable you to use SS more often. At the start of SS CD I think it is, at the middle or the end of the SS CD probably not unless S&B had a proc chance higher than 50% which is does not.

If you were in a 5 man dungeon and you happen to be only sitting on 3k AP because of the group composition you may want to Devastate more, however in 5 mans Conc Blow/Shockwave/Thunder Clap/Heroic Throw usefulness is scaled a lot higher than on a boss due to there AOE threat and CC uses.

Devastate is important when your getting your sunders up, to keep it up and to try and proc S&B when its viable. It can also be useful for very low health single target uses when other things are on CD because it will still give your next devastate a higher chance to crit.

 

dest

I went ahead and "modeled" the first 90 seconds of a boss encounter using the SS/Rev/Dev/X and "Use it when it's up" rotations. Obviously these results will be quite varied based upon S&B procs (and the fact that it resets the rotation entirely...), but I believe the point will be received just from the modeling... take it with a grain of salt, obviously.

The bottom line :

Not factoring in additional S&B Threat

SS/Rev/Dev/X Rotation :

Total Threat : 122651
Number of "usable" S&B procs : 9

"Use it when it's up :

Total Threat : 124673
Number of "usable" S&B procs : 3


So there is a 2022 threat differential between the two. However, there are 6 additional "usable" S&B procs (meaning it doesn't proc on the GCD before shield slam would come up). I believe this would more than make up the difference between the two, and edge the SS/Rev/Dev/X rotation into the lead.


The spreadsheet with the values is attached. Feel free to let me know what I've done wrong. I'm sure that there is something :-).

Attached Files

threatmodel.xls (34.0 KB, 112 views)

 

Yomon

I made a program where you can see how much threat the specials do with a certain amount of attackpower and block value - I also included crit, as it is the only stat where all specials scale but Rend does not.

With this program you can easly see how good Rend really is - and its very good, here some examples:

4000 ap, 15 % crit, 1.6 Speed, 171 Weapon DPS (Last Laugh), 1200 block value, improved Rend, Mangle/Trauma, improved Revenge:

Edit:New values in Post 76.

So in most cases it is good to keep Rend up. Also the improved Rend talent instead of improved Charge might be worth it.
Also Concussion Blow and Shockwave outscale Revenge at 3338 ap or improved Revenge at 4256.

EDIT:A short explanation what Devastate /w S&B means. If you use Devastate as the first or second special after Shield Slam it has a higher value than the third special, because Shield Slam will be up after the third special anyways and a S&B procc wound be wasted. A S&B procc is calculated as a Shield Slam instead of a Devastate.

Threat Calculator (Filefront)

 

• Jamor

I really, really wish they would allow devastate to add the sunder threat while stacking again. And increase the inatte threat a bit. It's extremely painful on threat to put up -- and feel like, by far, the worst threat move we have. The numbers show it's low, but it feels even worse when using it.

 

Yomon

As a DPS Fury Warrior you have two global cooldowns between the first and second Bloodthirst; since most Warrior are near to both, hitcap and expertise dodge cap, sunder armor is very likely to land and can help you a lot in the very first seconds of a boss encounter.
If you have a DPS Warrior with you, ask him to do that, it will but increase DPS as well as threat generation if Sunder Armor is as soon as possible at 5 stacks.

If you use Rend, keep Thunderclap up and refresh it at least 7 seconds before it drops (it might get resisted), Concussion Blow and Shockwave you don't have to use Devastate that often.
Devastate is not that good for threat - but hey, its better than Sunder Armor, increases DPS and has a high chance to procc Deep Wounds or Sword and Board.

 

Xerophyte

I spent the evening getting a decent closed form expression down for Devastate threat. It's not quite perfect, I make enough assumptions that a simulator would be preferable, but it should be decently indicative of how things play out. First, note that for abilities with cooldown we cannot simply say that shortening or lengthening it's cooldown by one or more GCDs is a complete loss or gain of that ability's damage. For instance, shortening Shield Slam from 6s and 4 GCDs to 3s and 2 GCDs by a Sword and Board proc means that we've gained a 50% of doing one more full Shield Slam before the fight ends, at the ultimate cost of a 50% chance of doing one less Devastate over the course of the fight. Similarly, delaying Concussion Blow by 2 GCDs is only a 10% chance of losing a full CB, and so on.


With that out of the way, we can deal with Devastate more completely. If we assume that our baseline attack is Devastate and we can usually expect no other abilities to be available then opting to Devastate rather than use some long cooldown ability will:
- Always cause 1 GCD of ability delay for your Cooldown ability in exchange for Devastate 1 GCD earlier
- If immediately following a Shield Slam, have a 30%*50% = 15% chance of giving you (Shield Slam - Devastate) extra threat
- If 2 GCDs after Shield Slam, have a 7.5% chance of (Shield Slam - Devastate) extra threat
- Have a 30% chance of delaying your cooldown ability by 2 GCDs rather than 1.

The expression for total threat gain is thus ( Threat(SS) - Threat(Dev) ) * 0.3 * CD(SS)/CD_tot(SS) - 1.3 ( Threat(Other) - Threat(Dev) ) * GCD/CD_tot(other).

I calculated the result for the 2nd set of threat values posted by Yomon and got the preferred ability choice as:

CD Ability	Dev 1	Dev 2	Rev 1	Rev 2
Rend >75%	Rend	Rend	Revenge	Revenge
Rend	Devastate	Rend	Revenge	Revenge
Heroic Throw	Devastate	Devastate	Revenge	Revenge
Concussion Blow	Devastate	Devastate	Revenge	Revenge
Shockwave	Devastate	Devastate	Revenge	Revenge
Thunderclap	Devastate	Devastate	Revenge	Revenge

"Dev 1" refers to 1 GCD after Shield Slam, "Dev 2" to 2 GCDs after Shield Slam

Notable errors that I'm aware of are not accounting for differences in miss, dodge or parry and not accounting for more than one ability being delayed by choosing to Devastate. If anyone wants to play around with it; use other threat values as these were picked mostly at random from a handy post and such then I'm attaching the rather bare bones spreadsheet used.

If some results seem odd here then please bear in mind that the basic rule for cooldowns is that delaying long ones does very little whereas diminishing a short cooldown can do quite a bit. Yes, Heroic Throw does good threat, but losing 5% of a Heroic Throw is worth gaining 15% of a Shield Slam - even if we assume the Devastate used to proc it does no threat in and of itself then the chance of gaining a SS is worth delaying most long cooldown abilities simply because the less valuable Devastate is the more we gain from making sure as few abilities as possible in our cycle are Devastates.

TL;DR version: Far as I can tell Devastating in the two slots immediately following your Shield Slams to optimise Sword and Board procs is almost always worth it.

Attached Files

DevThreat.zip (17.7 KB, 35 views)

 

Xav

Well I had written up a nice lengthy post about stuff and then got crit in the face by my laptop being a piece of shit and causing me to somehow hit "back" and lose the post, and since it's early I didn't have the common sense active yet to write it up in notepad first. I'll re-write it but won't have nearly as much detail, apologies!

Anyway, as for threat rotation thing: I actually do tend to replace abilities at the end of the rotation before SS (devastates that is) with a long-timer one - this is something I noticed myself doing this past week as I paid closer attention. I got a slightly different impression from that original post though, but what you said does make sense. Due to the spontaneity of tanking in general it's easy to use a less than ideal cycle, but when you're able to sit still and pay closer attention, that's how it works out.

With that done, I wanted to (re)address the threat table Yamon posted. I downloaded that threat tool as well. I put in my typical raid buffed stats which are about 4800 AP, 15% crit, 171.6 DPS, 1.6 speed. TC comes up barely ahead of devastate with no S&B proc.

However, TC has a 1.85x damage/threat multiplier, and Devastate is simply damage+5% of ap. (However, with the Devastate Glyph, Devastate is Damage+10% of ap, Warrior Glyph of Devastate (double check me) - TankSpot testing was done there, and some shitty math by me on glyph comparison)

My average Thunder Clap on a raid encounter was ~850, ~750 for devastate. (Sapphiron as the fight, so I'm wearing max stam/survival stuff)

At 4800 AP and 750 Devastate, that's 750+240 for no glyph, and 750+480 for glyph, or 990 / 1230 threat baseline for Devastate.

At 850 TC damage that's 850 * 1.85 = 1573 average baseline. With no Dev glyph, TC is over 50% ahead baseline. With the glyph, it's about 25% ahead.

Now, TC crits off Incite only, maybe Cruelty (haven't done that testing), and only crits 150%, no Impale bonus. Devastate crits off melee crit and S&B, 200%, and Impale.

Are the crit bonuses of Devastate enough to pull ahead of TC's large threat advantage baseline? Further, is that table calculating it wrong, are Sat's numbers wrong, or am I a moron?

 

• Jamor

Hmm...this news about the Devastate glyph has certainly peaked my interest. I am going to give it a shot this week and see how I like it. One nice bonus for threat, especially against casters with the dev glyph is that you are hitting harder quicker -- which can really help against caster DPS.

There are not too many threat sensative tanking times currently in the game, but I can see this being helpful with our 3 drake sarth strat. We heroism right away on the first drake, and depending on movement due to lava waves, I sometimes find myself in a threat hole against mages in particular <who can put out crazy TPS w/ Heroism, depening on hot streak>. Also note that I put Vig on the Sarth tank, as we don't have a prot pally and don't group 8 them (in general).


As far as prioritizing devastate, I always use it over TC -- for all the reasons you mention. Impale and the higher crit chance adding to deep wounds being two major reasons. Sword and Board procs are nice as well.

 

Xav

Well yeah, actual reasons for using Devastate over TC can vary, and most of the time Devastate will naturally be better just by design. But I was moreso questioning the position (and threat value) of TC on that list, it seemed far too low, as if it was calculated wrong.

 

Yomon

I was not aware that TC crits only for 150% damage, but I think Cruelty increases TC critical strike chance - wowhead data indicates that TC is a physical school spell (but why does it crit for 150% then?). Maybe we have to test Thunder Clap furthermore.

Well, I added the TC 150% crit modifier, a Glyph of Devastate option and a Impale option. In this version Cruelty still works with Thunder Clap.
Since Rend does not scale with Impale as well, Impale was a important addition I simply forgot.

Here is the data for Xavs raidbuffed stats: I included improved Rend, Trauma/Mangle, Improved Revenge, Glyph of Devastate, Impale 1200 BV, 15 % crit, 4800 ap, 1.6 speed, 171.6 dps:

Edit:New values in Post 76.

Threat Calculator:Filefront

Edit: I used this formula for Thunder Clap: (300+(ap*0.12))*1.85*1.3
300 is the base damage, it scales with 12% of ap, has a 1.85 threat modifier and its damage is increased by 130% through talents.
Damage is reduced by (13083)/(15233+(13083)) or 46%, maybe I should add a option for applied Sunder Armor that wound be (9158)/(15233+(9158)) or 37,5% damage reduction. But that wound only change the relative value of Rend, since all other abilities scale with boss armor reduction (or does Thunderclap not?).

 

• Jamor

From my experience -- have run no numbers on this. TC is pretty horrible in general. I would expect it to be at the very bottom.

 

Xav

The TC numbers are close to my own, although using that formula and that armor reduction, I'm not sure why it's coming up lower. Are you not counting minor armor reduction (curse of reck, faerie fire, etc)? Naturally though they should all be affected the same. If you're going off base damage numbers you'd think TC would be even higher in your example, because it has such a large threat modifier, compared to Devastate which is nearly pure damage and a minor % ap coefficient.

I'm doing my scratch pad math based on actual results from a WWS, and it isn't matching up, Yomon. It's certainly likely I'm doing something wrong, but I can't figure out what it is.

Your devastate numbers just seem far too high (if you don't count the S&B part of it). Then again, you're taking crit into consideration, so that may be it, which answers my older question.

Hmm, maybe it isn't the crit thing after all. I downloaded your newer version and put in 0% crit, and then -15% crit (to nullify S&B and Incite), and Thunder Clap was still significantly lower than devastate (20%+). With it being completely impossible to crit, it should go only off of the base damage, which is nearly the same. Then it's the passive modifiers, which is 1.85x vs 5 or 10% AP. What's going on?

 

Thommy

I was just curious if anyone has looked into a pure block value set and it's viability. Of course, making sure to keep high amounts of stamina, defense capped, and not gemming purely for strength. I currently gem Solid, Enduring, Guardians, and Sovreign in that order. I can get around 1600 block value unbuffed right now. With glyph, shield block, and trinket up I was almost 4k. I was curious how that incorporates into using devistate over TC/CB/HT/SW. I'm working on maxxing out the set just to see how it scales with threat/mitigation/survivability. It should be around 2k block value unbuffed when completed. Would a high shield block set change the priority of the cycle much or just in the 3rd GC of a non S&B procced rotation?

 

Yomon

Maybe I made a mistake. The problem of a program instead of a spreadsheet is, that nobody can point out mistakes.
I just read up the abilities on wowhead and did my math. This is my math:

Thunder Clap:
(300+(ap*0.12))*1.85

Base damage plus 12% ap, 185% threat

Devastate
505 + (0.5*dps*speed) + ((ap/14) * 2.4)+(0.05*ap):

101 base damage per Sunder Armor stack plus 0.5*weapon damage + normalized weapon damage (maces, swords, fists = 2.4)+5%ap

The Values are increased by Talents (Thunderclap:130%) and decreased by Armor. The advantage of Devastate is here, that it's extra threat is not decreased by Armor.

What might be confusing is that I multiplied the damage with the crit rate and thus the values are all quite high, but the values overall are the same; that also is with armor. Faerie fire decreases armor, increases the values of all abilities but does not change the relative values since all scale the same way.

Edit: I made up my mind - Thunder Clap does scale better than other abilities with armor reduction, I'll fix that asap.

Edit2:The difference is about 9 % more relative threat for Thunder clap.

Threat Calculator1.1:Filefront

Improved Rend, Trauma/Mangle, Improved Revenge, Glyph of Devastate, Impale, Sunder Armor 5/5, CoR/Faerie Fire 1200 BV, 15 % crit, 4800 ap, 1.6 speed, 171.6 dps:

Edit:New values in Post 76.



You can also see that the 770 threat modifier of Shield Slam does not scale with armor reduction, and now CB, SW and HT are above Shield Slam.

Edit: I'll add hit, dodge and parry tomorrow.
Edit2: There was something wrong about Impale, I fixed it.

 

Fellwraith

Are you guys also factoring parry/dodge/miss into your equations? Since tclap and shockwave can't be parried or dodged that should help the total threat per GCD vs the other attacks. Most of us aren't anywhere near the expertise cap in a standard tank suit and even hit capping isn't that realistic in full T7.5 quality gear (it's more likely in earlier T7/heroic gear).


One other qualitative thing I'll point out is that you can't always shockwave on every boss. Maybe it's just me, but it seems that you need to have shockwave hit the center of a mob's hit box in order for it to register. Anytime you have a mob who cleaves, that cleave can easily chain the melee if you're too close to the center (this is especially true with dragons since they also usually have a tailwhip, so melee are more paranoid than usually about going too far to the back end).

 

◊ Birdemani

Wasn't there a blue comment that Shockwave may be falling short a few yards of it's intended range right now?

Relwin: Besides, the BB is not some ivory tower of WoW knowledge, it's just less stupid here than elsewhere.
DeeNogger: Not less stupid, better stupid. The BB takes stupid very seriously. Now if you will excuse me, I have to go misspell the word fire.

 

Xav

Yomon, if it's possible, can you post the numbers you see if you edit your program (don't have to upload it) not to take crit into consideration at all, that is, 0% crit rate on abilities? Just for my own sake.

 

• Jamor

Yeah, but I am guessing it will still be unusable on large model mobs.

It's supposed to be 10 yards, and estimations put it around 7 right now.

 

Zartion

I don't really have hard data to back this up, but from anecdotal evidence, I've seen the graphic pass through the mob, go another yard or two, and not hit them. It seems really inconsistant, and I also get wierd things, when I move my camera one direction but my body is facing another, it'll still shockwave the direction I have my camera turned.

Kinda sucks when your trying to pick up adds.

Stupidity, like the plague, is best cured with fire, and hysterical violence. =)

 

Dogmeat

All this talk about how good Rend works for threat has me curious to try it out. I'm just wondering if its worth picking up if I DO NOT also spec down to Trauma. If not I would like to see some possible tanking specs that have Trauma included. Just curious since speccing down that far into the Arms tree means you are gonna obviously have to give up quite a few pts elsewhere, and I really can't see how I can give up that many pts in the prot tree to spec for one ability.

But, if you DO NOT pick up Trauma does Imp Rend at least make it a superior choice over Devastate as far as threat is concerned??

 

Yomon

I have updated my program, it now also includes the increased critical strike chances through Incite, Critical Block and Sword and Board; it also includes One-handed Specialization.

Please point out if you see some mistakes or if something is missing (like a talent etc.).

Here are the new stats of a raidbuffed warrior with: 4800ap, 1200 BV, 15% crit, 8.25 expertise, 1.6 speed, 171.6 DPS (Last Laugh), imp Rend, imp Revenge, Mangle/Trauma, Glyph of Devastate, Impale, Sunder Armor 5/5, Faerie Fire/CoR:

Edit:New values in Post 76.

Devastate now looks a lot better, and Rend is worse.

Program: Threat Calculator (Firefront)


You do not have to pick up Trauma - a DPS Warrior will do, or a Feral Druid with Mangle. Since I cannot see your stats in armory you have to download my program and type in your stats; then you can see if Rend will be superior to Devastate in your case.

Edit: Added Glyph of Rending.

 

Branar

Is there any chance you can incorporate [Glyph of Rending] for completeness in the next version?

It's really pretty amazing how good Rend is in your Broken Promise scenario.

 

Xerophyte

Updating the priority chart with those threat values gives the priority list as

CD Ability	Dev 1	Dev 2	Rev 1	Rev 2
Rend >75%	Devastate	Rend	Revenge	Revenge
Rend	Devastate	Devastate	Revenge	Revenge
Heroic Throw	Devastate	Devastate	Revenge	Revenge
Concussion Blow	Devastate	Devastate	Revenge	Revenge
Shockwave	Devastate	Shockwave	Revenge	Revenge
Thunderclap	Devastate	Devastate	Revenge	Revenge

Again, '1' is the first GCD after Shield Slam, 2 is the 2nd. Same assumptions as before: only one ability cooldown is being delayed at a time and avoidance is not accounted for beyond the initial threat values. Basically Shockwave and Rend are the only abilities that are competitive with Devastate threat + the chance of a Shield Slam, as they're the only ones with cooldowns short enough for the delay to be a significant damage loss. You should still almost always Devastate or Revenge apart from the GCD slot just before your Shield Slam or possibly when multiple longer cooldown abilities are up.