As I understand it, you're capped at 100% paperdoll armor pen, as that will be 100% of your maximum possible reduction. Check the example given by CG when describing mechanics. However, I haven't seen much testing of how sunder adds to arpen from gear.

ETA: I understood wrong, there's proof over here: http://elitistjerks.com/f31/t29453-c...8/#post1201773

Interestingly, the number for full penetration given in the link is 1282, which seems pretty attainable during trinket procs.

ArPen benefits have already been tested with controlled tests with fixed damage attacks which is what the ArPen calculations were based on and confirmed (was misunderstandings in the existence of an armor modifier). As well as there is no hard cap until the target is at 0 armor at least so you can exceed 100% ArPen on tooltip and it still works (the "cap" from GC's equations seems to just control the effectiveness of ArPen not the literal cap). The only question left with ArPen is its full behavior with various debuffs/buffs but at least with Sunder & FF they seem to apply multiplicative first to the armor debuff then ArPen kicks in.

http://elitistjerks.com/f31/t29453-c...8/#post1202772

As well, should say the benefits of Agi are harder to test than ArPen since Agi adds a random element whereas ArPen is a fixed flat benefit to certain attacks. So if anything you should be questioning the benefits of Agi since a lot of Agi's benefits can go wasted if you are at 4 combo points and you get an extra combo point that is wasted (in fact in spreadsheet models like Rawr they assume no wastage overestimating the value of Agi).

We're supposed to do that intentionally for Rip, always Rip with 5. Keeping Rip/Rake/Mangle and Roar up as much as possible has given me the best result for max damage, and I'll intentionally waste combo points rather than risking a Fero Bite to ensure that those four things are constantly rolling without interruption. This approach is even more appropriate with the possibility that a Fero Bite doesn't crit; with the R&T nerf there is a possibility of that happening.

On our 10 man General Vezax kill last night, I was #1 with 5700 DPS, and 46% of that was Rip/Rake. With that much damage from bleeds alone, I'm hesitant to think that giving up 189 raid buffed agility for 11% armor pen is the way to go (I calculated it out based on my current available sockets). The buff to white damage/shreds would have to be staggering from that amount to counteract the loss of agility, which affects everything.

I'm also not convinced. I just spent the gold to completely regem to ArPen, and I'm not seeing any noticable DPS increase unbuffed against a heroic training dummy. I have 442 ArP rating now, which I assumed would be high enough to notice. But checking again, both Rawr and Toskk's still tell me Agility is better (I had assumed before I regemed that they just didn't have all the bugs worked out). Even with Sunder in the model, both put Agility better with Blessing of Kings and +10% AP.

Putting together a "BiS" set with the limited gear in the latest version of Rawr, I actually saw a slight DPS decrease gemming for ArPen instead of Agil. Does the gear even exist to get to that threshold where ArPen becomes better than Agil? Not to mention, for any fight with movement or burst required, it seems Agil will be better for CP generation.

@ Tauran - i just watch for that upcoming roar and clip it if i see it rather than take a chance.

Vezax is a fight where Shred should be the majority of your DPS because of the black stuff reducing energy cost. Worked out to shred at 38.1% for example on one Vezax attempt with a shred crit up to 15,590. It is really not a good fight to look at due to very bursty damage.

WoW Meter Online - Combatlog Replay in Engrish

I'm guessing your buffs aren't setup or something?

Fully raid buff for me it shows ArPen is worth 1.54 DPS and Agi 1.46 DPS while completely unbuffed ArPen 1.02 DPS and Agi 0.88 DPS add in only Kings and its 1.05 DPS for ArPen and 1.03 DPS for Agi. This is with 546.4 ArPen rating according to Rawr.

FBN seems to clip savage roar a lot, and frankly SR is more important than Rip, which makes SR clipping a Good Thing. That means you're probably not putting up a Rip without SR.

That said, you need to make sure you're watching an ability-timer monitor as well. FBN has one, but, no offense to Nightcrowler, I think BadKitty's timer display is better (though still suboptimal). If you're keeping track of ability timers, you should know whether SR is up or not.

Of course, that might not mean that SR is the best move--or even that Rip is the best move at that moment. FBN is not a substitute for thinking on your feet, and has never been advertised as such. Nightcrowler has done an awesome job with it, but you still have to make the snap decisions on your own--and trust your instincts.

This was in response to Hezkezl mentioning XT-200 dps values so the dps skewing is irrelevant.

He is stating that he has regemmed to ArP while I have stayed with Agi. We are closely geared, I have gone for a combination of Rawr and Nightcrowler for my gear while he seems to have stuck true to Nightcrowler.

If we post our best attempts and others can see what gear and gem choices we have made, it will only be a good thing. I always like to see what other druids are getting on bosses my guild has downed so I know where I am. My guild has downed FL, Razorscale, XT-200, Assembly Of Iron, Kologarn and Auriya so far and I have been first or 2nd out of all the melee in most fights so I would say that our damage increase has been quite nice. Kologarn was the only fight that was difficult to get very high in due to not being able to shred.

Yesterday we downed XT-002, I'm not in an hardcore guild, I'm in good but friendly guild, where people plays with friends, all workers and so on, so our progress is not at hyper-speed. We downed him after 6 attempts. My dps was between 6.9k and 7.6k with an average of 7.45k (I did 7.2k on our kill badly because I became the bomb 4 times in a row and we killed him 2 seconds before the enrage). I have topped the damage meter on every attempt with a rogue very near me, and every other far behind (the third on dps-meter was at 6.7k average).
If you use my addon you should remember to put on the right cycle (for istance we had an arms warrior so I've checked the manglebot check and I've also unchecked berserk suggestion because you want to use it only when the heart is active), on Kologarn I was about 6th on damage meter with about 5.6k average but we didn't do a lot of attempt (we only raid 2.5-3 hours). For Kologarn I've unchecked the shred options in the cycle panel and checked the idol of wastes (obviosly better because you can't shred).

Regarding Arpen (this argument starts to be redundant), as I said before you are not likely to see a "big" dps improvment regeming from agi to Arpen, I've not changed all my gems, simply because knowing the 3.1 before it went live I've already gemmed my gear for Arpen in 3.0. The difference is about 3 dps for each gem, so you are likely to see a 30-40dps increase, not more. Personally I prefer Arpen to agility (at least for cat only gear) because the dps feels more stable, also with too much crit I feel like I have too much waiting time.
Arpen DOESN'T have a real cap, for a druid the Arpen REAL cap is at 104% with sunders and FFF, you need 104% of Arpen to reduce boss armor to 0.

@Kirian: no offence at all, I'll probably made a bad-kitty style display timer in the future, but I'm interested in different concern at the moment expecially because I don't like badkitty style (this is one of the reason I've implemented that display timer in my addon) and people who prefere badkitty style can still hide my cd-timer and use both addons, so it's not a big concern atm.

@Hipnotyk: 46% of rake/RIP makes obviously Agi better than Arpen, but you shouldn't see that numbers. A right rake/RIP value is around 30%, there is no sense in not using FB in your cycle, expecially if you use my addon to keep track of abilities.

About scaling: many feral druids see themselves topping damage meters, talking with a friend mage-therycrafter this morning we were considering the possible nerf of feral druids, I don't think we will see it, our dps will not increase that much with BiS Ulduar gear (about 8-10%, I'll do the exact math after all gear will be discovered) considering that classes like mages will se a dps increase of about 30%.

your algorithm has some awful bugs
YouTube - bugs
please fix if

Thank for your video is very cool (wow what a strange interface setting!). I'll answer your questions one by one:

6 sec RIP, 4 sec to full energy, 1 sec rake and FB?
Yes. It's right, as you said after you had 6 seconds of RIP downtime, just do some math, and you'll see that 6 seconds of RIP downtime (2-3 ticks depend if they are 5.9 seconds or 6.1 seconds actually) is about the same damage as an average FB. Obviosly you can be unlucky, your FB doesn't crit and the next 5 abilities doesn't crit either losing some dps, but you can also crit FB and crit 2 next cp-generators RIPing at less than 6 seconds of downtime. 6 seconds of RIP downtime is exactly the average acceptable downtime for RIP. As for rake, rake and shred does about the same DPE, so you should always rake but not at the expense of other abilities.

Roar when RIP is down?
Obviously. Do you think it's better to do a RIP with Roar down? No it's not. Also it's still better in the average situation (for istance heart exposed on XT-002 is not the case) to lose some RIP time and clip SR, instead of Ripping and then wait a few seconds to put SR up again.

OOC was keeping me away from RIP and SR at the ending?
OOC keeping you away from RIP is ok, Shreding with ooc up is better than ripping also if you waste some cps, as for SR, SR has always the highest priority. That behavior actually could be examinated more (thanks for reporting), I'll look at the algoritm, probably I could check for RIP/SR timers when ooc fires, It's possible that in particuar situation like the one you reported using RIP instead of shred it's better, I'll analize it soon. The main problem is that, that kind of situations are unlikely to happen, so when I run the simulator the dps difference will be very small, but I'll try to figure out the true.

No prediction?
Yes, it's normal, the prediction is 1 second ahead in the future, If you are not doing to use abilities for 1 second, the prediction frame must be empty.

I hope I've solved your concerns.

@BioNit:
4 sec left on roar (they could be 3.1 or 3.9 actually) is enough to choce refreshing instead of rip. If you have ripped, you would a) lose some seconds of SR uptime, b) use SR at 1-2 cps having a short uptime and then you'll not have time for a complete sr/5rip/5fb cycle. If you fill that sr clip is too much you can set "sr refresh cp" in the cycle options to 1 cp instead of the default 2 cps. This will probably solve your problem.

[quote=nightcrowler;1204810]
Regarding Arpen (this argument starts to be redundant), as I said before you are not likely to see a "big" dps improvment regeming from agi to Arpen, I've not changed all my gems, simply because knowing the 3.1 before it went live I've already gemmed my gear for Arpen in 3.0. The difference is about 3 dps for each gem, so you are likely to see a 30-40dps increase, not more. Personally I prefer Arpen to agility (at least for cat only gear) because the dps feels more stable, also with too much crit I feel like I have too much waiting time.
Arpen DOESN'T have a real cap, for a druid the Arpen REAL cap is at 104% with sunders and FFF, you need 104% of Arpen to reduce boss armor to 0.
QUOTE]

How do you feel like you have to much waiting time? I find that having that 50% crit also for smoother rotations. Not having enough crit, I feel will personally get you stuck in a situation where you have SR at 1 sec, rip at 1 sec, and rake at 1 sec.

30 - 40 dps increase though is huge increase in my opinion. Id like to see some numbers from other druids to compare with. Me currently I gem for agi. Id like to see some number from druids who gem with all ArPen for some serious comparrison. But is it worth regeming from all agi to all ArPen? Im not convinced. I guess my jury is still out on the debate.

This is not result of low critchance. Sticky situation you describing is result of our complex rotation. If you are using addon or if you are experienced feral DPS druid, you can reduce this holes in DPS, but you cant completely avoid it. Stacking ArPen move your damage from bleeds to damage from autoattack, shred, mangle and FB. So then, if sometimes your bleeds lose uptime during fight, it has less impact and as was said manytimes before, your DPS become smoother.

Regeming from AGI to ArPen doesnt skyrocket your DPS. As I wrote somewhere before, the models with 100% bleed uptime and no FB, like Toskk, shows AGI endgame gear better then ArPen endgame gear. But on live, where you cant provide 100% bleed uptime, will be both gears on the line with DPS. Just from personal experience with ArPen gear after few fights, ArPen is much sexier. Bigger numbers, you can use FB almost whenever you want and dont be scared, that fallen bleed gimp your DPS.

Thanks Nightcrowler, this ArPen+FB style bring much more fun to my role in raid and its more a pleasure to do DPS as cat, then hard work, like before.

*I apologize for any troll language. Didnt payed attention during english lesons.

I dont know about the troll thing there, but arguement you make about ArPen making you less reliable on Bleed uptime and more so on your own damage output is something very interesting to think about. I never thought about it in that light. But I would be interesting in seeing some of you dps numbers, or maybe a dps video to see if its really worth switching from Agi to ArPen.

As a reminder, based on the post by GC, it's impossible to reduce anyone to 0 Armor.

A = Armor
C = Armor able to be Penetrated
P = Total Armor Penetration %
E = Effective Armor after Penetration
(A+(400+85*targetlevel+4.5*85*(targetlevel-59)))/3 = C. This is the maximum amount of armor you can penetrate (level>60 obviously) You'll notice this value is Lower than boss armor.
A - ((Lower of A or C) * P) = E

For bosses that use the standard 10645 Armor, E = 1551.67 at 100% Armor Pen

Does this imply that once we reach a certain amount of armor pen from gear, sunders, FF and averaged out procs that we should start going for more agility?

Edit: The reason I ask is because Rawr averages out grim toll and actually ranks agility ahead of armor penetration for me now. I have 267 armor penetration from gear, ignoring grim toll. I was wondering if Rawr contains a miscalculation, or whether I have actually reached a 'plateau' on armor pen?

The "cap" mentioned by GC is only the armor point used to calculate the effectiveness of ArP. According to research done in http://elitistjerks.com/f31/t29453-c...8/#post1203636, you can actually get higher than 100% ArP. This will continue to reduce the target's armor to 0. That cap value also changes depending on debuffs on the target (FF/Sunder).

Really nice work, nightcrowler. I absolutely loathe the font and may go and change it, but the rest of the program does everything I want in virtually exactly the way I want. It's the best combination of TMW and Rogue Power Bars I've so far found, and having the CP/Energy in a big box is exceptionally nice.

Please, please change the font or allow it to be changed more easily. Please.

I tried FBN the first time today and really love the Cat version of it. But when i use it while tanking it suggests FFF all the time, though i disabled it in the options.

While using FBN last night on Razorscale, i noticed that a couple of times it asked me to clip rip, I had 4 sec remaining both times it did it. Is it possible that it is not noticing the shred increases correctly? I see it extend the duration on the timer and it only happened a couple of times, just curious if its a little buggy. Also I had an issue several times last night when i could not reapply Rip, did not have another feral in raid and Rip debuff was off target completely so the "more powerful spell" issue that was described in Cat Dps Rotation thread doesnt fly.

Mijae, I don't know why I was assuming we couldn't cross 100% with ArP since so many other abilities can. If Sunder/FF work on target's base armor (likely), and C=15232.5 (uses Player level of 80 not target level of 83) that gives us:
(10645*.8*.95) = 8090.2 + 15232.5 / 3 = 7774.23 making the ArP cap at 104.06% or 1282 ArP Rating. Nightcrowler was correct, and I shouldn't base my calculations off of GC when testing shows he's wrong.