Wednesday, September 9, 2009

Pimping a blog: Feral Aggression

Haven't done this in a while, and I'm hoping that with the loss of Toskk, Flyv and Karthis there might be others I'm missing out there.

I'd like to say hi to Vallen, author of Feral Aggression. Vallen's a very good writer who writes a bunch about doing cat DPS in various situations, as well as an awesome set of guides on how to solo old-world content for awesome ground mounts (which all druids love, of course). He wrote a great piece on what 3.2.2's armor pen means to you, and while I'm not on board with his or Jacemora's conclusions it's nicely written and brings up good points. So check him out and say hi!


Jacemora said...

My big thing is I don't buy that the increase of damage from increased ArP is worth a loss of Crits. Crits are double damage where a boost of 100 or so ArP from Gemming is not.

For the 100 agility you are getting an increase of Crit, AP, and it scales with Kings. For 100 ArP you get 100 ArP.

ArP might edge out agility at a 100 for 100 swap but it won't be a huge difference. Take into account running from targets and downtime from attacking and I personally think Rip ticks win out. On bosses where you can stay stationary behind the boss for the duration ArP could be the way to go but again not by a huge margin.

Would this change with one of the 2 ArP trinkets? Sure, but now we are talking a small window of 100% armor reduction.

Would I soft cap if I had one of the trinkets... yes. Will I go out of my way to get the trinkets... No, especially with ArP being removed completely in the future.

I am just not a huge fan of simulations dictating gearing, live practice and testing I feel means more. If anything I would like to read a logical argument for ArP over Agility and not "because the simulator ran 1000 boss fights with a huge RNG deviation and ArP came out on top"

How much harder is a shred crit with 300 ArP versus 400 ArP all other stats staying equal for example. 100 more damage? 500 more? What about a simple melee attack?


I scored cheap 2T9 today in hands and shoulders slots. This bumped all my stats up just a hair and I plan on reporting back how this effected my rotations and DPS coming from 4T8.

Kalon said...

That's fair, Jacemora, and the important thing is to gear such that you, personally, do the most DPS. Regardless of simulations. For instance, on Yogg I often use the idol of rip when we're doing things without Freya because doing extra damage on bleeds while I'm not otherwise DPSing is more worthwhile to me. That's not the 'correct' sim result, but it's a result of what works on that encounter.

On 300 armor pen vs 400 armor pen, we need a baseline shred. We'll go with a nice round 5000, with crits of 11330. If you go up to 400 armor pen you're looking at a 3.5% boost, so shreds now do 5175 and crits do 11726. So it's almost 200 damage per shred, and almost 400 damage per crit. Using the calc from Toskk gives us a higher chance of crit with 100 agi, but nowhere near the base damage increase.

If you're actually relying on bleed damage and doing the most damage while away from the boss, the best thing to stack is strength of all things. It improves rip and rake damage more than agility does, it makes all attacks do more and it is multiplied by kings.

My personal finding is that there are very few fights where my bleeds are overpowering the rest of the damage I do to such a degree that I'd want to maximize their damage. When I can't get face time on a boss most of the time I won't be able to refresh my bleeds either. And more often than not, the big boost of damage I can do is dictated not by bleeds, but by short bursty damage that is most directly affected by armor pen, such as FB.

Here's the logical reason for it: armor pen does not scale linearly, and at a high enough value increases damage significantly from non-bleed hits to compensate for the loss of crits and CPs and not working on everything. Looking at the wowwiki chart, we have 100 armor pen increasing the DPS output (for a given affected attack) by anywhere from 2% to 7% depending on where you fall on the spectrum of damage. Hit 1000 armor pen and you're looking at a 40% boost of damage to all attacks that can use it, roughly. Now, let's be really generous and say that rip & rake do 40% of your damage, and you do the rest with armor pen attacks.

So if you go armor pen, your damage will look like (.6 * 1.4 + .4)*x.

If you go agility, your damage will look like (1.0 *y )*x.

In order for agility to be better, y needs to be 1.24. In other words, 1000 agility needs to increase your damage by 24%. Now, 1000 agility will add 12% crit (which isn't exactly 12% increased DPS, it's slightly more) and 1000 AP, which is about 7-8% more damage. So overall that's less.

But wait, you say! That doesn't take into account CP generation! And that's true. It's hard to put a value on what CP generation is in terms of DPS.

Also wait! you say. 1000 armor pen is insane and way higher than most people have. That's true. I have 800 armor pen, for instance, and at 800 armor pen it's more like 30% boosting of damage - or only an 18% boost overall. Compare that to the 9.6% crit chance and 800 AP, and agility starts getting closer.

The 'benefit' if you can call it that, from armor pen, is that it's consistent. Agility can give some big swings one way or another, and that can be great - but it can also suck. I personally like the stability of my damage being fairly consistent; for this, I sacrifice some stability in the cycle and have to pay attention a bit more.

One last final thing to consider: even if you don't like stacking armor pen, you need to recognize that it is still the best secondary stat ferals can have. That means that armor pen is going to exist on gear that you want regardless of whether or not you stack it, which means that it's not that hard to push it to where it really edges out agility. For example, I've got only 242 armor pen from non-armor right now. While it's certainly leaning towards armor pen, it's not like every piece has to have it (and many don't).

Jacemora said...

I think we are on the same page actually.

I just happen to value increasing my RNG chances and cp generation is all.

Note that when Nightcrowler does his testing he has a huge range of outcomes. I understand the concept of ArP giving a more standard outcome than agility where it concerns crit and CP generation RNG but I think even 2% more crit moves that whole range to higher DPS or at least close enough where I don't have to sweat gemming for ArP.

Great explanation though, I appreciate the math.


Skittlesbucket said...

I'm so glad you're still around. Stuff like this is gold.

Alaron said...

Well, Vallen stole my idea. :P Okay, not really, but I'd been working on a more detailed post off and on discussing ArP. Check it out: link

Kalon said...

Alaron, that's an outstanding analysis. Great job. I'm going to make another post to point it out.

Skittles, thanks :)