The soft ArPen cap is 566, that is taking into consideration the Grim Toll proc that would put me at the ArPen hard cap. I currently have less than 566 ArPen (without the proc) so it should suggest more ArPen gemming. Either way, it shouldnt suggest I replace *all* my gems with +Agi gems.

The thing is, other people in the Druid forums who imported my character from Armory say Rawr is correctly suggesting they gem ArPen, so I assume its something in the settings I messed up.

About the +85 ap enchant, I'm just waiting for a better weapon to slap massacre on it.

Removed post.

Well, when I load my char from armory the stat value shows ArPen > Agi. However what I'm doing is just clicking on Optimize, and allowing it to regem my gear. It always ends up switching every gem I have to +agi for some reason. I *know* mathematically thats incorrect, because I'm better off using ArPen gems up until the soft cap, which I havent reached even with full ArPen gemming. Furthermore, others in the Druid thread who loaded my character from Armory also report Rawr suggesting ArPen gemming. I'm a loss as to what it is I'm doing wrong.

Do you have Fractured Rubies 'greened' for Optimizer?

What character? Are you buffing yourself? What basis do you have for "knowing mathematically" that ArPen>Agi for you? Again, just post your character file, and I can explain whatever questions you have specifically.

Just a heads up, the proc rate for Extract of Necromantic Power is wrong. The internal cooldown is marked down at 45 seconds, while it's actually 15 seconds (just tested it on a dummy). If it makes any difference, I'm talking about the Boomkin module, although I imagine the proc values are the same across all modules.

I've manually editted the item to make it into a 15 second internal cooldown, but I'm not sure how to test and calculate the proc chance to check if it's really 10% or higher. If you guys got any idea, I'd appreciate the info as I'm deciding between that and another trinket atm

The basis for knowing ArPen>Agi is FeralbyNight (Cat & Bear simulation tool). It has been proven that over 230ish ArPen its better to stack it over Agi. I'm sure whoevers making the Feral model for rawr can confirm this aswell.

Anyways, here's the char file:
http://www.faction-gaming.co.uk/menoferal.xml

thanks for the help.

That's not particollary true, last I heard FBN dps value are the nearest to reality (and still seems too high).
FBN is not a theorycraft tool but a simulation tool, it doesn't make assumption and everyone can play with the code and change the cycle if they find a better one, atm I failed at finding a better cycle than the one I used. Rawr on the other hand make assumption for energy usage and dot uptime that are far from reality. Being a theorycraft tool that's obvious otherwise it will take ages (like FBN) to compute a simple stat value.
I use Rawr my self because is pretty simple to use and has a good interface, but I know its limit (and many times I choce to ignore them).

I think we are mixing up Apples and Oranges. First of all FBN and Rawr uses different cycles.
Second in rawr you put in your char and see the stat allocation, with FBN you can do the same but is far more complex, so I used to push a BiS gear for each tier with the relative stat allocation. The 230 value is refered to a T7 BiS geared druid with Grim Toll if I remember correctly. Different gear set will give you different stat allocation for sure. That's really doesn't matter because the important think is selecting the right gear and giving it the right value to choce your drops. I near BiS gear both FBN and Rawr put Arpen ahead of other stats, they also give very similar values for stat allocation (around 0.03 dps difference from each other, for istance with my bis gear rawr gives me 1.78 for Arpen and FBN gives 1.8).

This point is really the issue here, in my opinion. With 4 piece T7.5 Rawr says ArP gives me more dps, but honestly, how valuable is the small amount (in my case it's only 7 dps) compared to the added value of being able to switch to bear form and have the very valuable stats provided by agility? Even the best guilds out there have things go wrong at times and will need us to go bear unexpectedly and prevent a wipe, and that added armor, crit for savage defense, dodge, etc, can make that little bit of difference.

If you NEVER will be put in that position, I would consider spending the gold for the few extra dps and regem, but stay tuned. 3.2 is coming, and we don't know final adjustments yet.

Those are really 2 different things.Firstly I'm not sure how much those 4-5% extra dodge will save my life in that situation, second is a matter of personal feeling, I'm the first who is specced with only 1/5 FA while dpsing because I prefer to take ImpLoP and improved swipe, but I know that single target dps will be lower. It's a trade off. Max dps theorycraft is different from what is better to do in real game because real game depends on your raid setup, wich boss you are aiming for and many other un-theorycraftable factors (or simply not worth to theorycraft).

The "what if I have to tank suddenly" argument still seems very flawed; does it really matter that much if you have ArP gems or Agi gems when presumably ALL your enchants and your meta will be dps-oriented? I've put agi gems in any piece of gear I have to use for 'double-duty' but that's still far from ideal - just sit back a minute and think of what you're missing out on defensively (even assuming you've gemmed all agi):

meta: 21agi/ 3% crit damage vs 32stam/2% armor from items
helm: 50ap/20crit vs 37stam/20def
shoulders: 40ap/15crit vs 20dodge/15def
bracers: 50ap vs 40stam
legs: 75ap/22crit vs 55stam/22agi
boots: 12crit/12hit vs 15stam/runspeed (or 22 stam I guess, but I'd still take runspeed)

As an enchanter I also have 64ap on my rings where I could have 48stam (similar differnces for applicable tradeskills like LW etc).

For the slots that have enchants, only chest, gloves, cloak, and weapon have enchants that are suitable for both cat and bear--you must make significant sacrafices everywhere else. Can you mitigate your losses by putting agi in every gem slot instead of ArP? Sure, but after all the other losses you've incurred with your enchants is it really going to make an appreciable difference in your ability to tank anything remotely dangerous? You don't have near the HP of a properly equipped tank, and more than likely there are deeper problems with the raid (that caused your tank to get gibbed in the first place) that will wipe you regardless. If you're at the top of threat when the tank eats it, you're likely to A) get splattered before you react or B) get splattered before your healers react; and if you're not top threat, you're likely to lose several DPS before you re-stabilize for the same aforementioned reasons--and probably eventually die to an enrage timer (soft or hard).

I've been on a hiatus for much of wrath, but throughout all of TBC, I can count the number of times on one hand I've turned a wipe into a kill by taking over tanking after an MT death (once on a PUG mag, once on Teron, and a few other times that I can't even recall which boss it was). Those were the days when agi was pretty much king for both jobs, and it was only made possible due to PvP gear shenanigans in PvE and how easy it was to hit a "minimum HP threshold" as a druid (esp. a tauren druid). Bosses hit for so much more these days, I wouldn't expect to achieve similar success even in such limited capacity.

So, TL;DR - you can argue the merits of ArP vs Agi for dps and people are certainly free to gem which ever they please(we seem to be in agreement its very close either way), but the "accidental tank" scenario is inevitably a strawman arguement that, in my opinion, doesn't really hold up well in practice.