This thread, and all of the others like it, seem to have conflicting ideas of whether haste rating or armor penetration is better for a DPS warrior. All three of the weapons have similar top-end values but the sword obviously stands out for its higher DPS.

It should be NO question that a MS warrior PvE build will be 33/28 with blood frenzy (unless you have multiple raiding MS warriors?...) but there is debate whether an improved slam build is better than a MS/WW/spamstring rotation.

PRIOR to Hyjal/BT gear, it is apparent that an improved slam build is moderately better for DPS than a spamstring cycle (but is still limited by the skill of the player). Once haste rating and armor penetration come into play, though, it's not entirely clear. Haste rating is very easy to model, but armor values vary from mob to mob.

I've been searching for some time now to try to find some reliable armor values for raid bosses, but haven't come up with anything. The MobMitigation addon (available at Curse) is a good start, but can be wildly inaccurate since it doesn't account for debuffs and buffs when determining armor of your target. It would take an entire raid full of melee parsing armor values, averaging the results and adjusting for debuffs/buffs to find some good answers. It would be a lot of work, but at this point it's the ONLY way to be able to mathematically analyze which playstyle is better when given Hyjal/BT gear.

The field of possible optimal solutions is:
1) Improved Slam focusing on armor penetration
2) Improved Slam acquiring 2.5 weapon speed throuigh haste, then focusing on armor penetration
3) Spamstring build focusing on haste rating
4) Spamstring build focusing on armor penetration

These options might even be poor choices for many bosses with low armor. It might take only 400-500 penetration to reduce a boss to 0 armor after sunders and other debuffs, but you can equip a maximum of something like 1357 armor penetration with the optimal setup. Walking into a fight like this in the wrong gear would be a terrible waste of itemization points (made even worse with all of the agility on the T6 set). Until we can find reliable data on raid boss armor values, warrior DPS is just a guessing game.

A decent start would be: does beast lore give armor values on bosses like The Lurker Below?

This is my dps at Kaz'rogal using [Torch of the Damned] with my full +haste gear I had at the time which was the [Swiftsteel Shoulders] and the [Plans: Swiftsteel Bracers]

Wow Web Stats

This is the same fight again but now using [Cataclysm's Edge] with full +haste gear, [Swiftsteel Shoulders], [Plans: Swiftsteel Bracers], [Pillager's Gauntlets], [Band of Devastation]

Wow Web Stats

I've now decided though you lose so much stats using the Haste gear primarily that i'm going to go for armour pen over it, so I'm collecting 3 piece T6, the akama gloves [Grips of Silent Justice], the bloodboil legs [Leggings of Divine Retribution], [Choker of Serrated Blades] to go with the [Cataclysm's Edge] to give maximum armour pen

/edit forgot to say was using Slam rotation each time

As much as it freakin pains me to say this(And I stared at this post for a good 5 minutes before clicking post), Blood Frenzy could really use a nerf. Shadow Weaving was far less effective at increasing overall raid damage and it was nerfed from 3% to 2%, mainly from Shadow Priest/Warlocks benefiting.

Blood Frenzy affects All Rogues/All Hunters/All Warriors/Feral Druids/Enhancement Shaman/Ret Paladin DPS, it's 4% of all of their physical white damage and yellow damage, not just 1 arch type like Shadow school of damage, and it can be applied and steadily kept up from just 1 Warrior in the raid, on top of that it's not even in the Warrior DPS talent tree(Fury).

I find it silly that the justification for bringing a 2-hand Warrior to raid isn't because of his personal DPS, but because he brings a 4% physical damage debuff, and nothing more. A large chunk of this thread isn't even about maximizing personal 2-hand DPS, but maximizing it while you keep Blood Frenzy up for your guildmates.

Shadow Weaving is 10%, it's 2% stacked five times. That doesn't really seem to compare that unfavourably to Blood Frenzy.

Don't underestimate Mace Spec for PvE.
Over 9 Illidan attempts (9th being the kill) -

Tota Tick Avg Max
Mace Stun Effect (Rage) 1,632 272 6 6

That's a lot of rage in a short time span, and while it isn't alway needed, I don't think I ever had to interrupt my half slam half spamstring depending on haste procs rotation.

Comparing weapons by their top end damage for DPS purposes is maybe worthy of the official forums, but not of this one in my opinion. What really matters for normalized attacks like Whirlwind or Mortal Strike is the average damage of the weapon.
In this particular scenaro both weapons also have almost identical average damage, but this is not generally the case.

You have to keep in mind the group mana return via Vampiric Touch.

@ Edgewalker
I don't see the extra rage from mace spec as being worth considering. I am always in a group with windfury, so rage is never really a problem. I can rarely spend it fast enough. The only times I run low are when I swap stances and wait for an autoattack (OP/SS) or when I am cleave spamming. Usually on hyjal where I spend tons of rage on cleave, I'm taking damage that feeds the machine.

@ Emeraude
MS warriors are the debuff warrior. We provide MS, imp demoshout (since UW is pretty meh for 2h), imp thunderclap, and blood frenzy. The trade off is that we do a little less single target damage than DW fury warriors, who are usually streamlined for pure dps.

The reason I raised it was because you mentioned having 5/5 Flurry with a Sword as being a significant limiter on DPS due to the haste soft-cap, but with Blood Frenzy you should only be able to cap out at 3/5 Flurry. With only 3/5 Flurry, the haste soft cap for the sword should be considerably higher.

I get to use 2H Fury builds in raids (and it works _very_ well) because we have two DPS warriors, both 2H, and the other warrior is spec'd for Blood Frenzy. If you don't have more than one DPS Warrior... Blood Frenzy is as near to irreplaceable as you can get in terms of talents, and 33/28/0 puts out very high DPS to boot.

Shadow Weaving used to be a 15% increase to Shadow Damage, stacked on top of Misery for a grand total of a 20% DPS boost to all Shadow casters in the raid from a single shadow priest. They dropped it to 10%, with the additional 5% from Misery.

Blood Frenzy is very good, as it's a tad under a 4% DPS boost to Rogues, Enhancement Shaman, Hunters and Feral Druids (a tad under due to non-physical damage - Arcane Shot, poisons, shocks) and a tad over 4% for Warriors due to increased rage... but in its prime, Shadow Weaving comprised a 20% DPS boost to two classes, which is why Affliction Locks and Shadow Priests themselves were nearly unstoppable.

The only issue remaining with the sword is the speed. 3.5 really is very fast for non-normalized attacks; compared to the 3.7 speed Axe, even with only about 2500 AP you're losing about 35 damage on every Slam (before modifiers like 2H spec/Impale) and ~43 on Windfury procs (more AP). Still, Sword Spec is pretty beastly and that is a lot of raw DPS... eh.

I think the confusion came when someone else did their math according to 5/5 flurry, but it was not my intention to pass up blood frenzy. I only wish someone else would bite that talent bullet for me!

I think the sword will be be the best choice. The stuns from mace will proc on trash, but I don't believe its proc will ever help us push through something we've been having trouble with (i.e. extra dps takes down a burn boss). To paraphrase, while mace makes trash easier, trash is already easy.

I hope to have both, however, as the mace looks tasty for pvp. Damn blizzard and their fast swords.

It depends on the speed of the sword and how much haste is on it. I haven't been able to get a readable screencap of it's tooltip.

Where did you get that? Thank you.

I calculate 302 passive haste rating with 3/5 flurry, or 194 passive haste rating with 5/5 flurry to bring 3.7 down to 2.5. Immediately subtract 45 for the weapon itself. The huge difference between 3/5 and 5/5 flurry is why I didn't provide 3/5 numbers to begin with.

Obviously the amount of haste is a bit smaller if you want to be realistic and assume a little latency, 2.7 is good to shoot for, for example. The further you want to drag the weapon, the worse it gets.

If you're shooting for 2.7 as opposed to 2.5, we're looking at 201 haste rating for 3/5, and 102 haste rating with 5/5.

Math:

MMO-Champion - World of Warcraft Guides and Raid Strategies

They have a link to the revealed ZA loots a little way down the page.

MMO-Champion - World of Warcraft Guides and Raid Strategies
or
World of Raids :: Index

I'm just not sure how I feel about haste with MS. It seems like -ac will scale better, due to it applying to yellow attacks as well.

If you look at Canning's WWS parse, for example, only 37% of his damage is white damage. This includes windfury procs, a number of which trigger from instant attacks, and will not speed up with haste. 37 hits and 18 windfury procs. That means almost 50% of the white damage is coming from windfury. It would follow that only about 18-19% of his damage would be modified by haste. This is in contrast to-AC which effects 100% of his damage.

If you're going from Pendant of the Perilous to Serrated Blades, then it's an upgrade. However, I wouldn't recommend passing on Choker of Endless Nightmares for Serrated Blades. That being said, if you want to be nice to the rogues/hunters, then Serrated Blades will be a noticable upgrade.

The T6 helm is only an upgrade (even with all the other armor pen.) over Illidari Shatterer if you don't need the hit rating on HotIS. [Madness of the Betrayer] would also be a good one to get.

You really only have my word and my math scriblings though. On low AC mobs, you'll find you won't pull ahead nearly as much.

I was under the impression that the percentage gain to dps was greater on low AC mobs when using -AC gear?

Yes, however, there are some mobs already in the game where their AC is so low you can bring them down to 0 with just sunder and faerie fire. Unless armor penetration can push armor values below 0 (I highly doubt it), then on those mobs armor penetration would essentially be useless.

Hmm, some of my guildies are saying that, yes, sunder/ff will not reduce below 0, but -ac items will. Requires some testing that will hopefully get done by a willing participant (pincusion) from my guild.

I actually did test this and ran some numbers. While the dps between the two were both extremely close, slam showed itself to be the better use of rage.

here's a really good comparison of hamstring spam vs slam spam, I think: Wow Web Stats

Skru's gear is a bit worse than mine, but not enough to create that much of a gulf. His weapon is stormherald, which amounts to a Torch of the Damned -4dps. We were both in the same group and Anetheron wasn't faerie fired because all we had was a tree but that doesn't really matter. Personally my rotation is hamstring spam below 50% rage, slam spam above. at more than 50% rage I spam slam until whirlwind comes up, then I WW/MS. If I'm below 50% rage I just use MS/WW as they come up. I spend most of my time above 50%, though

In terms of which weapon is best, I think the sword easily wins out for PvE dps, but being a 3.6 weapon it's tough to say whether slam would be better or worse than hamstring spam as a rage dump. Serida does well enough without using slam, but I wouldn't be surprised if slam spam could squeeze out a bit more DPS than even he does

Weapon speed makes a biig difference in slam efficiency, since slam isn't normalized. A 3.8 slam receives quite a bit more of a bonus from your AP pool compared to a faster weapon

After punching a few hundred rats with equivalent listed dps in -ac and non-ac sets, it looks like you are right. Now we need to figure out at what point -AC stacking is wasted on bosses that are FF/Sundered and which bosses to switch out for haste gear.

What about AP vs Crit for fury?