Has anyone tried modeling a dual wield arms build for 3.2? I remember reading in the arms thread that someone tried it on a dummy on the PTR with less than favorable results, but a target dummy is not a raid boss situation. Is this just completely a non-worthwhile avenue of investigation?

For Massacre to become even with berserker the value would need to drop enough to remove 86 ap from Armored to the teeth bonus. Thats 2952 armor lost. I think that works out to 5% of 59k armor, which works out to be about twice what my tanking set has currenty.

Berserking has no negative effect on AttT since 3.1 afaik.

Looks like they're reverting the ArP Buff they gave a few patches back.

MMO-Champion BlueTracker - Armor pen nerf in 3.2.2

No; reducing it. Say the original Wrath release ArP rating conversion is variable R. The rating conversion buff a few patches throttled that up to R*1.25 . The change is that it is now R*1.10 . Make sense? That's a far cry from "reverting", which would be simply switching it back to the original R factor.

We also have no word from a Blue from a design standpoint about the change, and we don't even have a confirmation through patch notes, CMs, developers, or anybody really, even acknowledging that there was a change.

[edit] As of an hour ago, it is confirmed, and in the very link you posted. My apologies.

I have a question about the actual ArP formula. When finding which value to multiply by %ArP, the formula GC gave said the server picked the lowest value of either the ArPcap, calculated using (Armor_Target+C)/3, and the actual value of the target's Armor. However, I also seem to remember it being said that the whole point of this convoluted formula was to make ArP less good against lower armored targets.
So my problem is thus: given the formula as I understand it, ArP works just like the tool-tip says it will against any target under 7410 armor. I feel that that can't possibly be right, and was wondering if someone could confirm/correct me here?

I have 1050 armor pen rating from gear and use the 40 armor pen food. My SEP for ArP on Landsoul's spreadsheet dropped to 1.229 and my total ArP dropped from 98.5% to 87.88%.

Thank you for that. Would you mind checking the SEP values for lower total amounts (say 300 rating)? I can then interpolate the rate of growth and get an idea of whether to gem for ArP or not.

The purpose of converting Ignore Armor into ArP rating and flat amounts of armor ignored into a relative amount was to make it less deadly with clothies and (slightly) more effective on high armor targets. In BC you could easily bring a Mage's armor to 0 with an Executioner proc. Now this is impossible, even if you have 100% passive ArP, since there's a cap.
Well, it would be possible if you have the time to 5/5 sunder.

This has been explained many times, but since I'm in good mood, I'll explain again how the cap works: it depends on the target armor (unmodified), level and class. If the target armor after modifiers (sunder armor, FF) is lower than the cap, then you can bring it actually to 0. If not, you can reduce his armor up to the cap.

This is updated, with a preview of the nerf: Simple ArP calculator.

Changing all my gems for str, and swapping banner of victory took my ArP to the mid 600's (gear just has so much ArP these days, I guess this is why they are nerfing it) at that rating my SEP was around 1.113. However when I flat out subtracted 300 rating from that to bring it in the 300's (which is very artificial as realistically you would have other stats to balance it out) the SEP for ArP was at 1.0003. However because the SEP for ArP goes up the more STR and Crit you have, I would assume at 300 raiting it would still be very worthwhile to gem for ArP, especially when you consider that the more you gem for it the better it becomes.

You misunderstood me there; I was not suggesting that the conversion to percentages itself was implemented to protect plate weerers but rather the funny cap formula which essentially leaves prot paladins with ~13k armor even against a 100% arpen warrior.
The way ArP works right now, it is most effective against targets with ~8k current armor, namely hunters and enhancement shamans and 2hand plate wearers with sunder/expose up. The effect on cloth and leather classes is smaller and is often countered by class mechanics like priests' Focused Will talent.

edit: thank you very much tentfox, your help is appreciated.

Hysteria, Death Wish and Wrecking Crew do not stack with each other - enrage effects will never stack.

Maybe this isn't a simple question, but a search on "effective health" in this thread turned up nothing so I'll ask and see.

What's the effective health of 1 point of armor?

Specifically, I'm comparing the Monarch Crab (+123 stamina, Dodge rating 300 for 10 sec every 1 min) against Glyph of Indomitability (1792 Armor, Dodge rating 512 for 20 sec every 2 min)

Ignoring for the moment the +200 dodge rating, what's the comparison of 123 stamina versus 1792 armor in terms of survivability and effective health?

In general with my calculations (they are all pretty subjective - since everyone has different block which doesnt apply linearly), 11 armor = 1stamina in EH. So the Glyph is more EH, except remember - not every attack is affected by armor - stamina works for everything.

Could someone either explain, or point me towards a reference that explains, precisely how diminishing returns currently work for CC-like effects in PvP and PvE? The best reference I can find is the WoWWiki page on "diminishing returns", but that contains a mishmash of out-of-date and incorrect information. I can't find another reference-ish site with the information, and EJ searches predominantly turn up posts related to either the diminishing utility of various DPS stats in various contexts, or Blizzard's own system for DR on avoidance stats.

People I ask reliably quote the WoWWiki content (DRs expire fifteen seconds after the most recent relevant debuff has expired from the target), but a) it's ambiguous (15 seconds after the expiration of the first relevant debuff since DR's reset, or the most recent that participated in DR?) and b) it's incorrect (if it were, Shockwave, a 4-second stun on a 20-second (untalented) cooldown, would never experience DRs as long as you didn't use other stuns; in my experience, casting shockwave on cooldown always gets DR on the second stun).

I thought that since armor was non-linear curve on amount of damage reduced, then wouldn't your effective health per point of armor depend heavily on what your current armor value is?

I was hoping someone would come back with a graph or something showing with armor value 25,000 then EH per point of armor is X or even just a formula for how to calculate effective health.

i've been reading this thread and let me know if i got this right:
devastate does not give additional threat when applies the sunder effect... but still no tanks would start putting the normal sunder armor on the target until 5 stacks are on... is that right? i guess its just waste of time / rage
but how is that when i'm watching omen, 5 devastate hits generate at least 40-50% more threat than 5 sunder armor hits?

I have been trying to find some stat caps for warriors, and I have found some of what I have needed (defense for tanks, hit caps (263 PvE, 164 PvP, 361 duel wield) expertise (214) and resiliance (which I believe is 1464)... but no where seems to mention the stat caps for strength, agility, dodge/perry, and attack power.... if there is a cap for them anyway.

It's not that simple. The caps for hit you write are all hard caps. This means that having more rating than that will not contribute at all to your dps/tps. For most of the rating stats the caps are soft, meaning that the rating loses in value as you go above it, but still contribute. Stat for stat:

Hit
For non-dual wielding the hard cap is at 263 vs bosses (lvl 83 mobs), 164 vs equal lvl mobs/players. Modify this value if you have other sources giving hit (talent, dranei aura, food). 1 % hit gained from another source equals 32.79 (rounded from Whitehoofs post in the class mechanics subforum) hit rating.

For dual wielding the soft cap is 263 (however, as a DW warrior you should have the Precision talent bringing it down to 164). At that value none of your yellow attacks will miss any longer. This is the absolute minimum of hit you should have. You can go a bit higher, but hit loses about 40-60% in value when above the softcap so beware.

Expertise
For tanking Expertise has a softcap at 214, where none of your swings will be dodged. However, as a dps/tps stat (and mitigation if the boss gets hasted on parries), Expertise still has some value beyond 214.

For DW fury, Expertise has a hard cap at 214, if you assume that all your dps will be done from behind (where the mobs cannot parry). If you're in fights where this is hard to do, extra Expertise might actually help out a bit (making 214 a soft cap). But for almost all cases you can treat this as a hard cap.

For Arms, the cap is at 214, but can be reduced by talents (with Strength of arms the cap is 182, with Strength of arms+Weapon Mastery the cap is 116). It is a tricky stat for arms though, you should probably not aim towards capping it at the expense of other stats. For more info check the spreadsheet thread.

Defense
Has a soft cap at 689 rating (540 skill). This is where the chance of a boss mob critting on you is zero. It still gives avoidance also after 689 (although less efficiently than dodge/parry).

Resilience
Here I'm not 100% certain, but from what I could find Resilience has a softcap at 33% (i.e. maximum 33% overall damage reduction can be obtained from resilience, but crit reduction can be stacked further). Using the updated rating conversions for 3.2, I arrive at a softcap of 3111. Good luck on reaching that

Dodge
Dodge rating suffers from diminishing returns, which essentially means that dodge rating become less valuable the more dodge you have (for details check Whitehoof's post in the class mechanics subforum).

Parry
Also has diminishing returns, but it starts at a lower value than dodge. This makes parry a little bit worse than dodge point per point (since they now have the same rating conversion) for pure mitigation but does of course add some extra dps/tps through parry haste.

Agility
For agility->dodge, suffers from diminishing returns as above. For agi->crit, no cap (well, 100% crit is a definite cap) exists. Agility also gives some armor. For pure mitigation dodge>agi, but if you care about threat/dps, agi can be better than dodge point per point.

Crit
Crit has no simple cap, but there is a still a cap for it (which until 3.2 was more of a theoretical value since you could in principle not reach it). This cap exists due to how the hit table works. For white hits this cap is:
critcap = 100-24-(27-hit%)-(6.5-exp%),
where 24 is from glancing blows and the hit%/exp% terms cannot go below 0.
Example, with 10% hit and exp capped the critcap for white attacks is: 100-24-17=59%

For yellow attacks the glancing blow term disappears. Also, it is possible that yellow attacks are on a two-roll system (as far as I know no one has really been able to pinpoint that mechanic yet), making the cap 100% for crits.

ArP
The effect of Sunders+FF take the boss mob armor down by 25% before all calculations. Additionally there is a cap to how much armor can be reduced based on the armor remaining after these debuffs (see blue post by Ghostcrawler from a while back). Thus ArP from gear and talents only affects a fraction (depending on debuffs) of the total boss armor. These values are after the 3.2.2 nerf of ArP.
ArP needed from gear/trinkets (food/pots):
Fury
No trinkets: 1400
Softcap with Mjolnir: 735
Softcap with Grim Toll: 788

Arms
in battle stance (BSt): 1260 (90%)
BSt+T92p: 1176 (84%)
BSt+Mace: 1050 (75%)
BSt+Mace+T92p: 966 (69%)
Using Mjolnir gives softcaps which are 665 ArP rating lower,
Grim toll 612 rating lower.

Thanks for the input Gristle! An easy way of checking ArP and what you will get out can be found at Simple ArPen calculator

Strength and AP have no caps at all, you will continue to gain dps/tps however much you stack.

edit: clarified arms expertise cap

For Fury in 3.2.2 (rounded up):

Hard cap from gear = 1400 ArPen
Softcap with Mjolnir = 735 ArPen
Softcap with Grim Toll = 788 ArPen

These are the numbers I am working with anyway, based on the figures reported in this post. Feel free to correct me if you think I'm off the mark.

My 2nd spec is Prot for 10m hm's, so I don't really touch Arms. There are indeed a few permutations based on weapon used/whether you have 4T9 etc.

If I'm understanding the question right, the reason is that devastate does damage, which of course causes threat. The inherent threat of a sunder, even with the minor AP scaling it has, isn't anywhere near the threat devastate does just by raw damage these days. The inherent threat of sunder is something like 350. Devastate hits much, much harder than that now.

To put some numbers out there to help, Sunder does (345 + 5%*AP) threat per application. Devastate does (damage + 315 + 5%*AP) threat. Glyphed, it appears Devastate now does (damage + 2*(315+5%*AP)) threat. So, assuming you're doing at least 31 damage with Devastate, it is always more threat per application than straight Sunder.

Numbers come from research in these 2 threads:
Warrior WoW 3.0 Threat Values - TankSpot
Warrior Glyphed Devastate Threat; napkin math - TankSpot