We use an unorthodox strat for twins so they aren't always in range (or so you would think) of cleave, but cleave is still 30+% of my damage on that fight. What I found is that on the primary target, cleave will always strike twice, once with the damage buff from having the proper color, and once at a 30% reduction for the improper color. If the Two happen to be withing range for the cleave, the first target gets hit twice, the second target gets hit once as well. So you get 3 hits, one with the damage buff and two at 30% reduction, HOWEVER it is much more damage than a single heroic strike. IMO there is no reason NOT to use cleave on this fight and there is really no reason to not be glyphed for it either.

Do you have a log for the fight that shows this unusual behavior by cleave? In my experience cleave has never struck a single target more than once. Also unless you are saying that the glyph is causing this bugged effect, there is no need for glyphed cleave on this fight.

I would like to reopen some discussion on armor penetration, and gemming for it as a fury warrior.

My Armory: The World of Warcraft Armory

I am not a God-tier warrior. My guild hasn't seen any progression in 25 HToC, and we've only managed 4/5 in 10 HToC. Ever since I got serious about raiding, I've come to places like Elitist Jerks and trusted the info I've found here. There are better, smarter players out there than me, and I recognize that.

That said, I feel like the consensus in the warrior community surrounding armor pen for fury is wrong.

I've always enjoyed the Arms play style, and for kicks last week, I respecced into it to try out the incite build. Since it'd have been difficult for me to accurately ascertain its effectiveness without completely regemming (I was gemmed pure strength) I sucked it up, and spent the 2k or so gold to replace all my strength gems with armor pen ones.

After determining that Arms DPS wasn't as good as Fury (for me) I went back. Not wanting to spend another 2k regemming again, I opted out of investing in new gems right away. After DPSing in a couple 10 mans, I'm positive that I will not be regemming for strength.

Prior to this grand experiment, I was regularly pulling between 5-6k DPS in normal 25 man fights. Fights like Beasts, General, or Razor, or best yet, Patchwork, where bosses don't take extra damage and there aren't excessive cleave opportunities.

Since regemming, I've been pulling higher numbers, but in 10 mans, with fewer buffs, and with raid comps that have been far from optimal.

My making this big long post may be a bit premature since I don't have much concrete data to substantiate any of these claims, but the simple truth of the number is when I DPS now, I see bigger numbers than when I was gemmed pure strength.

Armor pen food + armor pen elixirs + sunders puts my armor pen in zerker around 85%. I don't have a Grim Toll or a Runestone. Running with this setup, I'm seeing my Heroic Strikes inching closer and closer to 10k... a number that I never got close to with pure strength.

I've given up nearly 1000AP and I've only gained ~30% more armor pen, but I feel strongly that my numbers are significantly better.

I have the utmost respect for Landsoul, and all the guys that work on Rawr, but I am currently of the opinion that their SEP values on armor pen are a little bit off right now.

Discuss!

Well I'm in the same situation right now, respecced from arms to fury (I have ArP hardcap as arms), and using the same gear I have the same dps in arms as in fury ; this being without regemming to strength, so I'm actually wondering if I have to regem or if staying with those gems is better, sitting at around 5k4 ap and 80% arp unbuffed.
I actually doubt dropping ArP would be the best thing to do as it scales really good when you have high AP, which you already have with fury spec and raid buffs.

Problem is , you shouldnt look at Heroic Strike damage as a indication of how good arp is.

Heroic strike scales less with AP then Bloodthirst (which is a considerable amount of your damage). Before all the "common buffs" (shared by BT and HS), BT gets around 0.5 dmg per ap point, HS gets 3.6/14 (most likely) = ~0.25.

So most likely your BT went down while HS went up. Additionally your deep wounds hit for significantly less now (another large chunk of dmg).

All of this can affect real SEP of ArP. I'm not saying ArP is bad for you - definitely dont know what kind of gear you have/can have, but those are things you omitted.

Glad you mentioned this. I was only trying to use Heroic Strike as an indicator of the results I'm seeing. My Bloodthirsts, my Whirlwinds, my Executes, even my regular ole white swings are all hitting significantly harder... I'm regularly seeing Bloodthirst crits over 10k, with the average crit sitting around 8.5k. If I had 4k Gold to blow I'd regem strength, take some parses, and then go back to armor pen and do the same so I could justify what I'm trying to say.

That said, I did not look at Deep Wounds numbers at all, so that's a perfectly viable point.

In response to gears... Armory was linked, but for the lazy folk, I'm in four piece T9 (232) with 232/245 pieces rounding out every other slot except trinkets. Weapons are 245 Justicebringer and 232 Obsidian sword.

I've really only paid attention on Twins Hard mode, but even with my improved cleave build I cannot put up as high numbers with cleave as I can with heroic strike. It just doesn't add up. We also raid with two DPS warriors, and we have almost the exact same gear, we are both orcs and we both have a heroic dual blade butcher. He uses cleave exclusively and I use heroic strike and I am usually always 1k dps (or more) ahead of him so he switched to heroic strike. He is now much closer to my damage. All gear differences taken into account we are usually within 500 dps of each other on any given fight. This is pretty damn close in my opinion. Having another warrior around makes it extremely easy to test out different strats.

Glyph of sunder is a great idea, but we use the strat where melee is all the same color. We only switch on the rare occurences where we know exactly what vortex / shield is gonna pop up next. How much dps do you think you gain from having both targets sundered for massive cleavage? For how we do it, I don't think it would be a massive gain, but it might make us switch back to cleave over heroic strike.

Deleted by Author.

Yea well i checked your gear.

1) Instead of gemming str, in your particular case, you should 1st of all go for almost all the socket bonuses out there :P. They are all 4 str or better and require a yellow (not blue) gem to activate.

2) Dont tell us your BT hits harder, when it doesnt. You can argue Landsoul modelling, I often do that (not with regard to very basic , and never more then just "i think your sheet SLIGHTLY undervalues stat x" etc), but when it comes to more complex parts like evaluating hit impact. BT damage with given raid buffs can be calculated absolutely accurately - you dont even have diff crit rates gemming for str or arp to start with. Its very simple and accurate math and individual "i feel like it hits harder", wont change it. Maybe you got more raid buffs since last time, or maybe your gear improved.

Heroic Strike deals only few more damage than Imp cleave, but accounts for a lot of the bloodsurge procs (1 hs every ~2s vs 1bt every 4.x and 1ww every 8.x) and this is probably the biggest difference.
The point where one gets better than the other is dependant on many factors. You lose ~3% of your overall damage from Slam using cleave and T9 4pc bonus increases this loss. However if you replace hs you would get between 18 and 30% additional damage from 2nd target depending on debuffs. You also get more flurry uptime and a lot more deep wounds damage.

If you can get some debuffs on the 2nd target you can see better results cleaving, and especially in execute phase, (where Slam is not used) if you can sustain cleave it provides a dps increase.

While I appreciate the discussion around ArP, and I'm not suggesting that you are right or wrong, but I would like to see the following before you suggest the established and tested wisdom is incorrect.

a: Your armory & a parse with Strength gems*
b: Your armory & a parse with ArP gems*

*Please ensure the parse contains a listing of all the classes and buff uptimes as this may account for the difference you are seeing.

Just to illustrate this by some numbers, assume one swaps 30% additional ArP for 1200 AP (this is equivalent to 30% ArP when considering raid buffs), and starts the fight with 8000 AP raidbuffed when gemmed for STR. Furthermore assuming berserker enchants on both hands and greatness card ... we can up the in fight effective AP value by 750 on average.
Thus we get:

1) In case of being STR gemmed:
8750 AP, 55% ArP => BT hits for 8750 * 0.5 * c * 3/(3-0.55) = c * 5357 , where c = 1 - ArmorMitigationOfMob

2)I In case of being ArP gemmed:
7550 AP, 85% ArP => BT hits for 7550 * 0.5 * c * 3/(3-0.85) = c * 5267

From this you can calculate though, how high your average raidbuffed AP (procs included!) has to be until 30% additional ArP surpasses 1200 additional AP for BT damage assuming base ArP value of 55% (neglecting Deep Wounds contribution to total damage here!). Let x denote said AP value, then following condition must hold

(x-1200)/(3-0.85) > x/(3-0.55) and thus x > 9800

As a sidenote:

Does anyone know for sure how AP contributes to next swing abilities? I assume the base weapon speed is factored in here, because everything else would be non intuitive (read: haste almost worthless). But I have never tested this explicitely.

So, here's what I'm not following. I loaded up your character in the spreadsheet, and the SEP for ArP was > 1. I used the phoney stats section and dropped 242 ArP and added 242 Str (the amount of ArP you have from gems). It dropped ArP to .99xx. It also dropped your DPS slightly. So, what this says to me is the spreadsheet is telling you to gem ArP. How is it undervaluing it if that's exactly what it's telling you to do?

Simple answer is an error of my judgement. I made my first post after running my gear through Rawr and assuming the spreadsheet would match. (Rawr put ArP at 0.93 for me)

That said, does this not imply that armor pen does actually pull ahead of strength?

I feel like you're assuming too much AP. 242 strength comes out to what? Just under 600 AP with imp zerker stance? I also don't have a greatness card. I feel like your numbers are maybe ~500ish AP off, which, using your same math, would again put ArP ahead.

And I do want to openly iterate that I'm not trying to shout "zomg regem armor pen guys!!1!", but to express that I feel my numbers are getting better, and think it warrants some testing.

It was an academical example to show how AP/ArP affect BT damage ^^
And 242 STR adds up to 242*1.2*1.1*1.1*2=700 AP raidbuffed (improved berserker stance, kings and 10% AP buff from shaman/DK/hunter). And having no AP proc (= greatness card e.g.) values STR even more.

But no one has stated that it never does. In fact it is well known that armor penetration at a certain gearlevel gets better than strength. It pulled ahead before the nerf to arp rating and all that the nerf did was setting the required gearlevel higher and now when people start getting full 245/258 item sets we start to reach that point again.

Hi all!

So many info about str vs arp gems and it's so hard to read and understand when you're russian %)
Seen some questions about what is better - hard str gems or hard arp - but isn't it better to get ~735 arp (other from Thorim's trinket) and then socket str gems? Thanks god 245-258 gear has a lot of arp, so it's pretty easy to get ~735. I have 4 t9(245) - all red sockets capped with arp gems, excepting one - already placing str gem. All yellow sockets - str+crit. All in all i have 1844str and 4969ap (without imp zerk), and 729 arp.

So, just want to ask highly-skilled warriors for some advice - should i continue saving ~735 arp and using Thorim's trinket (sorry, haven't the forggiest what's the name of this trink on english) or it's better to socket hard str/hard arp?

My apologise if someone already posted something similar, but it's realy hard to find required info in this mass of messages.

Armory link: The World of Warcraft Armory

Most people who have GT/MR (the arp proc trinkets) do exactly what you have done (stack ArP to ~730 so they're capped when trinket procs). Best way to confirm it is using Landsoul's sheet methinks.

If you don't trust it entirely think of it this way: the trinket has an uptime of about 20% right? So 20% of the time you have 100% ArP. If you replace one of your STR gem for an ArP one, whenever the trinket procs you'll have 20 ArP wasted since you can't go over the cap. But 80% of the time that ArP will still bring in damage. So what you're basically comparing in terms of DPS is: is 20 extra ArP worth it 80% more than 20 STR 100% of the time - 0.8 * ArP SEP > 1 Strength. I personally think not (sheet will prolly tell you the same thing).



Related to this (sort of) a question about possible incoming items. Would an ArP proc trinket similar to MR but with a lower ArP proc value (say 450) be better than MR?
Edit: nvm, figured that one out :V

Pardon my ignorance but how does using cleave result in greater flurry uptime, unless I'm reading your post incorrectly? I'm trying to figure out which, on paper, would result in more damage. I've been using cleave exclusively on that fight as it simply made sense to me that in an unlimited rage situation, why wouldn't I damage both targets instead of one on nearly every swing when BT and WW are on CD.

It increases it maybe by 5% if you are spamming cleave, since you have more opportunity to land crits in the same timeframe, reducing the chance flurry will fall off, but it's not very significant to the overall discussion of the argument.

The problem with the twins discussion is that twins is a gimmick fight with a stacking buff. You will get more damage out of building up your buff stack and blowing your cooldowns when you get empowered, and not wasting cooldowns on shield absorbs. This is classified under cheesing your raid because technically you are supposed to use your cooldowns on the shield and not when it does more damage for you. There are so many things in that matter that can swing your dps one way or the other based on the decision of the player to play correctly or to just simply cheese the damage as much as possible.

Cleave spam is most certainly going to do more damage, it's a no brainer. You lose maybe 3% damage from slam loss but gain 10-15% damage from hitting/critting the offtarget with your main hand, essentially almost doubling (offtarget has less debuffs but should at least have sunder) your heroic strike contribution in an endless rage situation. Everything the same, if someone is doing more damage with heroic strike, someone is doing something wrong.

You also need to consider the contribution of increased deep wound damage on the off-twin with a cleave spam over HS spam. We set up in the doorway, so none of our melee get the orb buff, but DW is usually second on overall damage for me...right behind cleave.

Thanks guys, you've more or less confirmed my thoughts on the whole cleave vs HS discussion on twins. We too use the door method so orbs aren't a factor in this fight for us. Would it be safe to assume though to maximize my contribution to the raid for that fight, using the door method, to cleave on every swing possible and to HS on the shield that we are trying to burn down (since I believe HS does more damage than cleave for single target)? I say HS on shield because when same color twin is shielded and we blow heroism to burn it down, it would feel cheap gimping my own damage to that target just so I can cleave and continue hitting the other target while my raid suffers. Someone please correct me if I'm wrong here.