spongejordan123

Yes.

 

Lamprey

So, tonight we been doing bunch of General goes and I'm consistently doing 1k dps less than I have done every other week. Cannot seem to figure it out. We don't have our enhance shaman here this week but I dont think that is 1k dps. Only other change i have made is lowering armor pen by changing it to strength because of the 100% cap so I'm sitting at 50% ArP with battle stance. Anyone noticed something new on this fight?

 

◊ Birdemani

I struck up a conversation with Hellorco earlier today regarding the possibility of canceling Sudden Death during execute range. Currently this macro is getting passed around to help with that final bit of burst to kill someone in PvP:

This macro will check if your target is in the 20% execute range, cancel your Sudden Death Proc and use your top rank Execute spell; otherwise it will execute according to the SD rules. I wondered if an Arms PvE warrior would get any benefit from using it during an encounter but I couldn't come up with the math to prove it and I needed some help. Hellorco went to work and came up with this (reposted with his permission):

Well it is indeed quite good for pvp, but for pve is very difficult.
In pvp you want to finish the target and you don't worry about your next hit. You may hit bloodrage just before it.

A 100 rage execute glyphed and talented is probably our best DPGCD move. So to catch the utility of that macro in pve you need a swingtimer and/or a <~1.8s swing.

Our best possible dps would be a gcd execute/op spam where you only use OP in 2 situations:
a) when your rage won't grant an equally damageful execute (see below)
b) when you have SD up and less than 41 rage after the execute.

The case b becomes better if you are in execute phase without BL.
With a swingtimer you do know if you can burn all your remaining rage on execute eventually changing your rotation.

Execute is 1456+0.2AP and 38 dpr. Talented and Glyphed is usable at 10 rage and will hit for +380 damage. With SD up you can grant another exe on the next gcd regardless of your swingtimer. This increases the chances you will generate rage before next "empty" gcd.
Up to 30 rage SD = non SD execute dmg, but SD = 20 cost.
The damage done @30 is 1456+1140=2596+0.2AP.
So I would add to the script an if UnitPower("player",1)>30 where you cancel the buff only after it will slightly cripple your dpgcd.

Let's take an ulduar avg weapon damage of 800.
OP does (800+AP/14*3.3)*1.2 and is 33% faster so OP=OP*1.33 = 1277+0.3762AP. 0.1762AP difference between the 2, so OP scales faster.

@10 rage: EXE = 1836+0.2AP. OP DPS is better than EXE when you pass 3173 AP
@30 rage: EXE = 2596+0.2AP. OP DPS>EXE DPS at 7485AP
@50 rage: EXE = 3356+0.2AP. OP DPS>EXE DPS at 11799 AP
@100 rage: EXE = 5256+0.2AP. OP DPS>EXE DPS at 22582 AP

What you do use is more tied to the swing than rps/dps.
With a 1.8s swing you can avoid OP unless you have less than 25sh rage since in any case you'll hit in between next free gcd.
Consider lag and reaction and you are more or less perfectly hitting every gcd during the phase.

Anyway consider this scenario:
Swing 2.0s. You just swinged and your gcd is up.
You have 2 gcds and your 3rd will happen after another swing.
If you don't have SD and have OP up you use OP regardless if it will starve you (that means you had <15 rage) for 0.5s
If you have SD you will use both execute and OP and you can execute (if over 30 rage you can remove SD) on next gcd since you will have another swing before further gcd.

So as you can see there are too many implications cause of the propagation of the effects of the different cds. Theoricrafting it would mean at best having an average over thousands of samples, due to the small duration of an execute phase.

I think that macro could be useful with a treshold of 40 since SD at 40 rage will still leave you with 10 rage and during exe phase you generally can consider any rage point banked as a static +38 increase unless you overcap rage.

I made some tests with an addon just after 3.0 that based on swing could suggest you when to use SD (since at the time it was using all rage). After 3.08 and SD cap I stopped bothering for it since your rage wasn't an issue and at low levels of AP Execute is undoubtly much better than other specials.

You now pushed the right button on my head and I may get back that work to do a small execute phase "suggestion" tool

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.

 

• landsoul

Bladestorm is definitely the ultimate in arena. A glyphed bladestorm will turn terrible Rogue Priest and Rogue Mage games into less terrible ones. On the PvE side, I generally use bladestorm if my rend won't fall off and directly after a quickly used TFB. Generally it happens like this: MS, TBF procs, OP, Bladestorm, TFB procs, OP, etc. Bladestorm is good single target damage as long as you aren't crowding too many skills.

Visit my Youtube Channel for an increasing selection of warrior videos, including Undermanning, PvP, and LK Raids:
http://www.youtube.com/user/LandsoulWoW

 

pizz

Yes, I noticed the same when switching back to STR gems and lowering ArP to 50% this week. Before the change I was sitting at 75% ArP in BS with hearty Rhino food buff. I am not happy with the results from 2 days of field testing in Ulduar. Even though STR gems and GT brought my AP to 4k. Going back to all ArP gems today.

 

Hamerung

We need to pester Blizz to make haste affect Slam casting time. Is there ANY good reason why it should not?

 

hellord

Haste doesn't reduce the effectiveness of Slam, it's true the opposite.
When you need to theorize Slam value you generally subtract from it the swing% lost. But what counts is the n° of slams in a given fight or if you prefer slam usage per second.

You are indeed getting less benefits from haste DUE to slam usage, up to a point where using slam is worse DPS than letting normal swing happen. This is only reached at very low speeds, just think that you can replace Slam with HS at about 1.75s that is during a Bloodlust+haste pot phase generally.

Having slam affected by haste (more or less like steady shot is) would make sense, and would let it scale much better once raidbuffed.

ArP Whore

 

Lamprey

Yea, I've decided this is what the difference is. I don't care what the spreadsheet says, but this whole "get to 50% ArP w/ battle stance so your grim toll or Mjolnir maxes you to 100%" just isn't right from my experience. Going back to stacking full ArP.

 

RolanNath

1k seems high but it was about what I saw in the same cituation., No enhance shaman really hurts warriors. Mostly the rage generation loss from the loss of the haste buff. Plus less crit, and less Str, and no bloodlust. Its a pretty big drop. We had a hard time killing him in a heavy melee group with no buffs, had to wait to the next day when our enhance was back, and all our dps went up around 1k and it was a 1 shot. I cringe with no enhance shaman or DK with buffs.

 

Shha

Have you considered the simple possibility though, that arms simply got nerfed by ArP change? You are comparing CURRENT str gemming vs OLD ArP. Sure ArP gemming was superior before the cap, and it was a huge nerf when they capped it (bigger then the 4% haste buff definitely). So dont compare current STR gemming to OLD ArP - you might end up with even lower dps if you do ArP again.

 

◊ Eej

The ArPen fix was put in before "last week" so I don't think that is relevant.

 

• Kalroth

It took me a few minutes to wrap my head around it, but the spreadsheet doesn't lie. Any ArPen past 50% drops about 20% in value, when equipping Grim Toll (and even lower for Mjolnir).

I made this page (mainly for myself, since illustration works better than numbers on me):
Why ArPen is bad after 50.32% with Grim Toll

 

hellord

I went a bit deeper into testing this and I had to use my swingbar mod's variables to make it efficient.
What I do is rather simple, if I have more than executeCost+30 rage and I'm before 1.5s from next swing, I cancel the buff.
With just a macro I'm not sure it's easy to get the swing timer, however you can make the rage check.
In fact wheter you have SD or not, at 40 rage it provides the same execute damage (since I suppose you bank the 10 rage from SD for the next execute), but gives you 10 rage you can convert in better damage like OP or another Execute.

Obviously it would be better to use a swingtimer to determine if it's worth or not canceling the buff, otherwise you can execute early and find yourself starved for a few before the swing. If you are over 1.5s from a swing is more DPS using SD and then executing again.

ArP Whore

 

Furrymaker

While your graphic illustrates how passive arp over 50% is wasted a small portion of the time it does not demonstrate why it's SEP value should be that much lower. What it comes down to is that about 20% of the time you have some item budget wasted on armor pen and during this period of time strength is infinitely better. The other 80% of the time armor pen provides better returns than strength. Looking at these facts by themselves and doing the math armor pen has a SEP < 1 as Landsoul's spreadsheet verifies. What I'm not sure is being factored in is that during blood lust for example even with the full grim toll proc coinciding with lust you're still going to spend the majority of lust without the proc. Having higher passive armor pen at this point should yield higher dps overall I would think than more ap/crit based on how it scales. Forgive me for not backing up my intuition with numbers, I guess it's more of a question whether this kind of thing is factored in or if it's even applicable.

Either way I wouldn't expect a drastic difference in the end result dps like some have reported. For the time being I'm going to stay above the soft arp cap but I'll most likely slowly trend towards 50%. Completely regemming would provide minimal dps gains. In my case Landsoul's spreadsheet has me gaining about 20 dps swapping all arp gems for str, dropping me from about 68% arp to 58%.

 

Jayde

Well, I believe the point is that if you lose 20% of the value of ArP, unless ArP was at a SEP of at least 1.20, it will drop below Strength.

If it was, for instance 1.1, it would drop to 0.88, whereby it would be clear that Strength will be more valuable.

That said, I'm not sure dramatically regemming would provide a huge difference if the SEP is pretty close. I mean, if it's in the 0.9s, the increase will be minimal at a large amount of effort.

 

Systema Sephirothicum

Well,there is also "remove grim toll" option.I have used the spreadsheet in a variety of combos and passive armor pen pass about 65% plus another trinket,such as,lets say, wrathstone,has more dps than strenght gemming+ GT.
Sure,some armor pen loses its value but keeping a constand armor pen(I repeat,i am talking on about 70-80% passive) provides much more benefit than just str+proc gemming.At least,this is what I have noticed.
If landsoul could answer to that,I guess would put an end to that kind of debates =) .

I want to have a child and use it as a pulling tool in dungeons.
It's aggro range is huuuuuuuuuuuge!

 

• Kalroth

Put it this way: if you got 75% ArPen (from gear + battle stance + food) then it's a 48% damage increase 80% of the time. The last 20% of the time it's only a 33% damage increase. You need 93.25% passive ArPen just to break even from the loss of 15% damage during the GT proc and that is why the SEP value is much lower.

And these numbers aren't intuition or voodoo magic, I got them from the spreadsheet.

The problem with that is that both Grim Toll and Mjolnir Runestone are very, very good trinkets. If you got 50% ArPen, then it's a 33% damage increase by itself and when GT procs (you reach 100% ArPen) it's a 66% damage increase. So the trinket proc increases your damage by 33% damage for 10 seconds every 55 seconds. Assuming you combine this with other buffs (Wrathstone trinket, Speed Potion, etc.) it becomes a very powerful proc and I can't see any existing trinkets reach their level.

I left out the passive stats of the trinkets (2.53% hit or 2.22% crit) from above numbers, but they simply boost the value of the trinkets even more.

 

Phr00t

I know these are posted all the time, but I'm looking for a little assistance.

I've been lurking these forums for a while, trying to figure out what I'm doing wrong.

With my gear, talents, and the common buffs we use in a 10 man (I always have a draenai in my group, so I'm hit-capped at the moment), the spreadsheet is coming up with 4300 dps.

I have yet to even do 4000dps even in a 25 man, on non-gimmick fights ala XT.

My priority is SD>MS>OP, slam between MS CD's when TFB hasn't procced, and HS on swings when I have over 80 rage.

Couple questions:
Should I be slamming at all during Heroisms?
Why is my DPS not near where it should be according to simulations? Am I missing something?
Should I even bother using bladestorm with my 3.2 weapon?

(Before it's mentioned, I'm replacing my green trinket with Greatness as soon as the faire comes around.)

Any help will be appreciated.

 

Anduryondon

Do you have any WWS from your fights? Personally i slam even on heroism and bladestorm is even with 3.2 speed worth it.

Another question: We're doing Vezax HM right now and i noticed that if i use HS on >80 Rage it is very luck dependend if i have enough rage. Even without using HS sometimes i have rage problems due to unlucky noncrit strikes. Right now i stopped using HS on Vezax, because i think it hurts my DPS more if i have sometimes a GCD that i cant use.
Any opinions?

 

Sabroan

I just tested on training dummies and I'm fairly certain they haven't implemented the ArP cap yet. With 637 passive ArP, I should be capped with a Grim Toll proc and without sunders. However, I got higher numbers on Heroic Throw when the dummy was sundered and even higher when sundered and shattered.

I made sure that the only buff I had up was the Grim Toll proc, so the only explanation is that Blizzard hasn't introduced the ArP cap yet.

 

bcswen

The ArP cap has in fact been introduced. ArP is capped at 100% armor reduction FROM GEAR, and it is important to note that 100% ArP from gear does not reduce the armor of a boss level mob to 0, because of a separate 'cap' on armor reduction.

You would therefore expect to see a damage increase if you stack 100% ArP + sunders + shattering throw.

 

suicuique

Either my math is wrong or yours, so help me on this please.

Assuming the standard scenario of debuffed boss armor > 15232/2; therefore AC ArP Cap is (AC+15232)/3

May x denote the ArP fraction as noted in the Character Screen ([+battle stance + mace spec]), then the DMG increase (realtive to 0 ArP) can be deducted to
DmgIncr = (15232 + AC) / ( 15232 + AC - x * (15232+AC)/3) = ... = 3 / (3 - x)
a very simple formula.

E.g. 100% ArP yields a 50% DMG increase (on armor mitigated attacks obviously).

As such having 50% ArP means you are doing 3/(3-0.5) = 1.2 times dmg, i.e. 20% more damage.

When GT procs (ArP = 100%) you are doing 50% more damage (relative to zero ArP) or 25% more damage relative to unprocced state.

All in all by my calculations you are overestimating GT proc value.
Am I wrong?

 

• Kalroth

50% ArPen is a 20% damage increase if you're hitting a target that isn't debuffed by 5x Sunder Armor and Faerie Fire. In that case the Grim Toll proc is "only" a 29.75% damage increase. However that isn't a very realistic raid scenario, which is why my numbers assume both debuffs are present.

 

suicuique

If a boss (AC = 10643) is debuffed by a full sunder stack and FF the armor is reduced to 10643*0.8*0.95= 8089 which is greater than 15232/2. So my numbers still apply.
To put it more bluntly: ArP Damage increase is dependant only on the amount of ArP (capped by 100%), NOT on the AC as long as AC is greater than 7616.

And how you get the GT DMG increase being 29.75% completely escapes me.
Care to elaborate?

 

hellord

Partly.

On average GT procs gives
@0 ArP = go to 50% arp = (3/(3-0.5))/(3/3) = 20% increase.
@50% ArP = go to 100% arp = (3/(3-1))/(3/(3-0.5)) = 25% increase
@60% ArP = got to 100% arp = (3/(3-1))/(3/(3-0.6)) = 20% increase

Average is 4-5% dmg contribution from proc before softcap and up to +10%. It's still better than any other proc.

If you are in that 10% range and you can swap 8 Arp gems for 8 STR gems so you don't waste anything from the proc, you would have 10% arp less and +128 STR (322 AP after buffs).

@50% ArP going to 60% arp = (3/(3-0.6))/(3/(3-0.5)) = 4.2% increase 80% of the time on 82%sh of total damage.
You give up 2.76% avg damage increase to regain back around 1%.
Is 322AP worth 1.76% of your damage? Most likely yes in this tier.

If you are at 75%+, you can't just replace ArP with STR since you won't have 25% ArP from gems unless you are a blacksmith with practically full BiS. This means you'll have to trade ArP for something else that could be not worth the deal.

There is no wrong or right math on this. It's very linked to how itemization points are spread on gear.
STR scales linearly and has diminishing returns, so there will be a point where 2%sh of your damage is more than the damage provided by 322 AP, but it's still at very high levels of AP rather than at high levels of ArP since you have to deal with itemization limits.

At very high levels of Arp I'm pretty sure I personally would find another trinket rather than regem regardless of what the spreadsheet says. Since I would know it will be closer to what I expect most of the times even if in the long run I could have 1-2% less damage.
After trying many different gear settings what I can say is that the spreadsheet gives you an overview of the average contribution of stats, but you can't just take it as it was the Bible.
GT and MR still procs a buff of 10s every minute or so. It can even not proc at all during a BL and for sure it will never proc twice during those 40s. It's quite different than having x% for sure.

ArP Whore