That's 51/7/13 with 1/3 TTT and 2/3 MP for Mutilate, and 15/51/5 CQC for Combat.

To answer in 5 words or less: Mjolnir Runestone and Grim Toll.

To expand on that point: ArPen was always going to raise in value as you got more of it. It still has positive scaling on it, such that the more you have, the better it gets; thus, once you reach some critical amount, it rapidly takes over as the most important stat.

In this case: The Mjolnir/GT synergy spikes the value of ArPen fairly significantly (I think it goes up by about .2 EP just by having those two trinkets equipped); thus, many items with ArPen start to become good. Once one starts equipping these items, the value of ArPen rises further, and eventally it gets up to about 2.1; at this point (since Agi is only about 2.05) it becomes optimal to socket ArPen instead of Agi or AP (and, for that matter, use ArPen food instead of Agi or AP). Switching out alll your red and prismatic gems for ArPen is thus worthwhile, and, thus, gives you that much *more* ArPen, which makes the stat that much better - which makes some ArPen items that weren't quite worth it before better still... and so on. You get the idea.

Thus, when all is said and done, while Agi has an EP value of around 2.05 - nicely ahead of straight AP - ArPen has shot all the way up to 2.38 in BIS gear.

This is, actually, one of the major reasons I revised the FAQ a day or two ago - because once this synergy became apparent, it was important to mention there.

I am a little confused about how Armor Penetration is so valuable when it is stacked so much. Forgive me if the boss value is wrong, as it's the most up to date I could find.

Assuming a bosses armor is at 10643, and according to GC's post the Constant for a level 83 is 16635, this puts the cap for armor ignore at 9092.66. (Meaning we will always have a minimum of 1550 armor up, unless I am misunderstanding).

Now the BiS gear you listed has ~522 ArPen (42.36%), and Grim Toll procs for 612 (49.58%) and Mjolnir for 665 (53.97%). Isn't having ~146% armor ignore redundant? Especially when you factor in Sunder and Faerie fire? At some point I would imagine that's theoretical gain and not actual gain.

Even if past 100% worked, any more than 117% (not factoring in Sunder/FF), would be wasted.

Of course I could be way off base on how stacking that much ArPen actually works.

So is re gemming for arp at a gear level around that of your own currently worth it?, or does it only start to shine with BiS.

Current speculation and testing seems to indicate (if I read the warrior threads correctly, and someone can correct me if not) that Armor Pen seems to be stacking well past 100% at this point in time.

So, your misunderstanding the way ArPen mechanics are currently understood to work. Which is understandable, as GC's post had a number of mistakes and some rather deceptive language.

First off: what GC describes as a "cap" is not a cap in any way, shape, or form. It's a parameter that indicates the strength of ArPen - it doesn't cap anything.

Second, despite what GC said, testing indicates that the constant used is based on attacker level, not defender level; thus, we're using 15232.5 rather than 16635.

Third, Sunder and Faerie Fire reduce what GC calls the cap.

So, the actual calculation for reducing someone to zero armor looks something like this. The target has 10643 armor; this is reduced to 8088.68 by Sunder and Faerie Fire. Thus, the parameter limiting the value of ArPen is (15232.5+8088.68)/3 = 7773.73. Hence, to reduce the enemy armor value to zero, we need 8068.68/7773.73 = 104.05% Armor Penetration, which works out to 1281.5 armor penetration.

However: if you put on more than this - which is easy enough to do with Mjolnir's and Grim Toll - what one finds is that if you have, say, 1400 armor pen on and gouge the combat dummy, the gouge will do more damage than the base damage of the ability. Which is to say: armor penetration has no cap. You can reduce enemy armor past zero, and just keep going.

As an amusing aside: theoretically, if one kept stacking armor penetration until you reached 300% armor penetration, you hit a vertical asymptote. That is to say: with 299.9% armor penetration, your attacks would do several thousand times their base damage. And in the realm of the truly crazy: theoretically, if you go past that (to 300.1%), your attacks would start doing *negative* damage.

Now, this is all purely theoretical, as it's not possible to more than about 250% total ArPen; and in point of fact, there might be some cap between here and 300% - I've done testing up to about 1500 ArPen myself, but there could theoretically be a cap somewhere short of 300% (which would take about 3600 ArPen to reach). But if there is, we haven't found it yet - so, for the moment, this is our best understanding of how things work.

Wow...that clears a lot up. Is this also the case with PvP? Seems like it would be an overly powerful tool lol. Also seems incredibly contrary to logical design of the idea of armor. But hey, works for me.

Thank you very much for the clarification.

wow thanks Aldriana. Is there a list of BiS from non-hard mode ulduar loots?

No such list exists, nor do I think it's necessarily valuable to create such a thing. There is some very good loot from some very easy hard modes.

For instance, I'm reasonably convinced that Thorim 10 Hard Mode is easier than YS25 easy mode, and it drops the best trinket in the game. So excluding it as a "hard mode loot" is sort of... dumb.

Thus: rather than trying to make a clear break between hard mode and easy mode loot, I suggest that you simply use the spreadsheets and figure out what's best from the content you expect to have access too.

Around what armor pen value or gear level exactly would gemming for ArP be more beneficial? Is the requirement for the EP value to outweigh AGI to have both grim toll and runestone? Also what does this speak of for mace spec?

Use a spreadsheet. Having both trinkets is not necessarily sufficient - for instance, I have both and still find it better to socket agility. It's also unclear whether you can reach the crossover point without both trinkets. Basically, the overall situation is too complex to distill down to "it happens at X ArPen". Use the spreadsheet, and see what works out best for you.

Regarding Mace Spec: it is indeed quite powerful; however, the lack of decent OH severely limits the spec. If there was a 1.4 speed OH and an ilvl 239 MH, they would quite possibly be best in slot. It just so happens that there isn't, which is why the BIS setup uses fist/dagger instead.

I find it surprising given the value of armor penetration for combat that 15/51/5 has pulled ahead of 7/51/13. which competed just fine in Naxx when armor penetration was less good. Has something made Lethality better than it was in that tier of content? Or is it just one of those inexplicable things that is what it is?

Im curious to know if agility would still have a higher EP rating than ArPen with the T8 4pc bonus. Also, for combat would it be worth it to pick up serrated blades with the t8 4pc?

I tooled around with the spreadsheet but it still showed it as an overall dps loss - swapped all slots to arpen, grim toll, serrated blades. Put me about 90 dps down from usual combat spec/gemming agi. Of course, changing to Arpen doesn't really benefit 4pc either, since bleeds are AP/crit dependent I suppose.

I am stuck on whether I should use my first runed orbs to craft [Footpads of Silence] or [Death-warmed Belt]? The normal hardmode Yogg is a ways away so I don't expect to get [Soul-Devouring Cinch] any time soon.

Thoughts?

Let's see. One item you won't replace til next content patch. One item you won't replace "any time soon". Your choice, your system, grab all 12 at once for all I know.

So, first off, it was always the case that 7/51/13 didn't scale as well; thus, even if nothing else changed, it wouldn't be too surprising if, as gear continued to improve, it continued to fall behind.

Second, there were significant changes to the way ArPen works in 3.1; for instance, while ArPen rating is a lot stronger now, that's partly because the conversion from ArPen rating to actual armor reduction has changed - that is, 100 ArPen now gives a different amount of armor reduction than 100 ArPen did before. Thus, just because ArPen *rating* has improved doesn't mean that a fixed amount of armor penetration - such as that from serrated blades - is going to be any better.

Third, the way Serrated Blades applies to the new armor penetration formula, it doesn't really give full benefit - it does reduce what GC calls the "cap". Thus, while it does lower enemy armor, it also lowers the efficacy of ArPen, and while you do still come out ahead by taking it, you don't come out as far ahead as you may once have done.

Fourth, I suppose I should probably add that at the moment I'm not modeling the cycle that would likely be optimal for Serrated Blades; hence, it might turn out that it's a bit better than it currently looks. But I wouldn't expect those changes to make 100 DPS difference, so I suspect it still comes out somewhat behind.

I have a question does it worthed geting [Belt of the Twilight Assassin] even that shadowpanther tells it's better than lw [Death-warmed Belt] and [Relic Hunter's Cord] i believe they are both better

Relic Hunter is pretty decidedly better than Twilight Assassin if you're Combat.

If you're Mutilate, however, the decision is a bit more difficult. So, Death-warmed Belt is certainly more flexible in that it gives you those two sockets. Add in the prismatic one, and you can stack whatever stat you want. They are both pretty close to each other, and depending on your other gear one may be slightly ahead of the other. Ultimately, I went with Twilight Assassin because I liked that it gave me a similar amount of Expertise as Soul-Devouring Cinch, which means once I eventually get that, I can swap it with minimal impact on my other gem slots. Furthermore, I figured it would be better to only use guild runed orbs on something that would be a bigger upgrade (the boots) as opposed to taking them for a fairly marginal upgrade in the belt slot.

YMMV, check a spreadsheet etc, but that was my basic thinking on it.

This seems incredibly counterintuitive and quite possibly bugged. Has it been explicitly tested on anything other than a combat dummy?

Is this testable at all? What rating value is needed to hit the asymptote against (say) a naked toon, or even a low level naked toon?

To the first - there's relevant testing here - it's done using warrior talents in addition to pure rating, but all evidence indicates that they work the same way. But I myself have only tested on a target dummy (though it was with all ArPen, so it's not just some weird thing with warrior talents).

To the second: not really, no. In fact, the lower the opponent's armor, the *more* ArPen you need to make the asymptote. It's a fixed 300% for a target who's armor is greater than or equal to 15232.5/2; once armor gets below this level, you actually need *more* ArPen to make it to the asymptote. Thus, short of actually coming up with 300% ArPen (and the best I've been able to piece together is still a good 400 ArPen rating short or so), I don't see that exact testing will be possible.

That said: a test even at some lesser ArPen quantity - say, 200% - would be illustrative. If such testing resulted in the expected behavior - i.e., attacks doing roughly double the base damage - that would be a strong indication that mechanics work the way we think they do. And with mace spec, executioner, Grim Toll, Mjolnir Runestone, and The Dusk Blade, that one might even be attainable without too much pain and suffering.

Exactly my thinking on the Twilight Assassin belt.

Until you see Ruined Orbs selling cheap (like Sunmotes eventually did) it makes more rdps sense to get the HUGE improvements made for other classes and/or items with Ruined Orbs.

As far as ArPen goes, is that how you can maximize your dps, as if to say, you can only reach X DPS mark by using ArPen where if you were to completely ignore it and to gem/gear as you were pre-arpen buff, you couldn't reach that mark.

I realize that the 2 ArPen trinkets just make it ridiculous with them both equipped when stacking it, but can you NOT stack ArPen and do similar dps to people that are stacking it?