Is it appropriate to use the same methodology for PvP? It seems that it would be a little different and we don't have a spreadsheet for that.
We need to figure out what each class's cap formula is. He indicated they're different. That's the pvp biggie and will let us figure out more optimal ArP numbers.
It's probably safe to use the one he posted for warriors.
We also need to determine the interaction with sunder-type debuffs... whether they come before or after the cap is determined.
We need to figure out what each class's cap formula is. He indicated they're different. That's the pvp biggie and will let us figure out more optimal ArP numbers.
It's probably safe to use the one he posted for warriors.
We also need to determine the interaction with sunder-type debuffs... whether they come before or after the cap is determined.
I used Landsoul's spreadsheet and equipped all of my PvP gear.
Full Armor Penetration gemming with Melancholy Sabatons: 4,369 dps
Full Strength gemming with Melancholy Sabatons: 4,383 dps.
I guess there might be some class differences for sure. Would resilience have less effect on ArP versus Strength?
We need to figure out what each class's cap formula is. He indicated they're different.
Although he indicated that at first, I think that was just a semi-understanding of the issue in his attempt to describe how complicated it is. I think though, after his second post, that the class of an NPC only determines the amount of armor it has (much like how BT/Hyjal had 2 armor values for all mobs, a lower "caster" one and a higher "warrior" one).
So I'm confused here.... I understand how the cap works in that your ArP rating only affects a portion of a target's armor based on the cap formula. What I don't understand is how the cap changes versus target class, target level, or player level.
I understand that faerie fire and sunder reduces armor level and the armor cap, but how does arms stance come into play? Is arms stance considered as armor penetration rating or is it akin to faerie fire and sunder armor in that it also reduces the cap?
So I'm confused here.... I understand how the cap works in that your ArP rating only affects a portion of a target's armor based on the cap formula. What I don't understand is how the cap changes versus target class, target level, or player level.
I understand that faerie fire and sunder reduces armor level and the armor cap, but how does arms stance come into play? Is arms stance considered as armor penetration rating or is it akin to faerie fire and sunder armor in that it also reduces the cap?
Is armor still 10643 for a boss?
Armor is still 10,643(assuming boss armor matches target dummy armor). Battle stance/mace spec stance still stack additively with arpen or testing would've shown the same strange behavior when combining them as it showed when combining arpen with sunder/FF.
That's what I was just asking myself in the shower.
Stealing Rallik's variable names, A = 1/2 armor constant, B = mob armor...
There exists a fixed value portion of a mob's armor, F, relative to the mob's level and B (and class?) but not exceeding B, for which armor penetration is limited to.
Problem: Is Battle Stance limited by F such that ArP% + BS% <= 100%?
Hypothesis: Yes. From previous testing, I would guess that these together cannot exceed 100% of F.
Test: With ArP% >= 100%, attack a target where 100% > F/B >= 91% while in Battle Stance. If Bloodthirst does full damage, ArP% and BS% can reach 110% of F.
Problem: If F/B < 100%, are Sunder and Faerie Fire capable of removing B-F?
Hypothesis: Yes. Because they stack multiplicatively and affect a % of B. It would be odd, but not unfeasible if they were also limited to F instead of B.
Test: With ArP% >= 100%, attack a target where 100% > F/B > 80% while in Berserker Stance with a full Sunder stack. If Bloodthirst does full damage, ArP% + Sunder + Faerie Fire would be capable of removing 100% of B (or up to 131.57% of F) where F/B >= 76%.
If both of those are true, then there exists a cap on ArP%, E for every F where E = 1/(F/B/76%).
Examples:
F/B = 1 (i.e. F=B): E = 1/(1/0.76), E = 76%, meaning tooltip ArP + Battle Stance has no effect beyond 76%
F/B = 0.81 (the current value for level 83): E = 1/(0.81/0.76), E = 93.927%
F/B = 0.76 (the perfect case where SA and FF remove the other 24%): E = 1/(0.76/0.76), E = 100%
What does this mean (if all this proves to be true)? Grim Toll + Battle Stance is 59.69%. So the soft cap while using Grim Toll vs level 83 bosses is 93.93% - 59.69% = 34.24%, or 422 ArP rating (Yay, we can stack more ArP!).
And if it isn't true? Then Grim Toll kinda... sucks. Previous to now, it was assumed that 84.324% tooltip ArP% was required to remove all armor with SA and FF. If SA and FF are limited by F but proportional to B, then together they remove 25.58% of F. Only 74.42% tooltip ArP% is then required to remove all possible mob armor. Take your 59.69% Battle Stance + Grim Tool from that and you're left with a paltry 14.74% ArP% or 182 ArP Rating from gear to reach the Grim Toll soft cap. I hope you kept your Mirrors of Truth.
-- edit --
Or I might've malfed that entire thing up because apparently F is based off of the current boss armor according to Aldriana's post here (post-SA and FF). Thus, E is always 100%. That means the value of ArP is higher when the mob isn't debuffed, as long as B is less than 2*A.
Last edited by Grayson Carlyle : 04/18/09 at 7:05 AM.
This mechanic doesn't make Grim Toll terrible, it just means that you will have the option to wear a different trinket other than Grim Toll sooner, and switch from ArP gems (arms) to STR gems sooner. Arp is better than STR for arms still so Grim Toll is still a good fury trinket.
Okay, here is a fairly technical explanation we put together for how armor pen works.
We didn’t want Armor Penetration Rating to be too powerful against low armor targets, like it had been in BC. We also didn’t want Armor Penetration Rating to be too powerful against high armor targets.
So, we decided on a system where there is a cap on how much armor the Armor Penetration Rating can be applied to. So, the first X armor on the target is reduced by the percentage listed in the Armor Penetration Rating tooltip, and all armor past that X is unaffected. Another way of understanding that is we multiply the percentage in the tooltip times the minimum of the two values: the cap, and the amount of armor on the target after all other modifiers.
Computing the cap is a little tricky unless you are already familiar with how World of Warcraft armor works. There is an armor constant we’ll call C. C is derived as follows (in some pseudocode):
If (level<60)
C=400+85*targetlevel
Else
C=400+85*targetlevel+4.5*85*(targetlevel-59);
For a level 80 target, C=15232.5. For a level 83, C=16635.
The cap for Armor Penetration then is: (armor + C)/3.
A level 80 warrior creature has 9729 armor. C=15232.5. So, the cap is (9729+15232.5)/3=8320.5. Let’s say a player has 30% armor penetration from armor penetration rating and no other modifiers that complicate the calculation (talents, Battle Stance, Sunder Armor, etc.). The game chooses the minimum of 8320.5 and 9729, so 8320.5. That is multiplied by 30% = 2496.15, and so that much armor is ignored. The effective armor on the target is 7232.85 (9729-2496.15). From a player point of view, the armor penetration rating didn’t ignore the full 30%, but instead ignored 25.66%. (85.5% as effective as expected).
These equations should help you be able to test and verify that Armor Penetration Rating is working correctly and as we designed. The tooltip is not actually inaccurate, as it states: “Enemy armor reduced by up to 30.00%.” That "up to" is key.
Please be sure to test without any other effects which modify the armor calculation (Battle Stance, Sunder Armor, Mace Specialization, etc.) as they may involve other systems that add additional complexity to the calculation.
which completely sucks, because we cant really 0 out an enemy any more, which is where armor pen really shined back at lvl 70.
GC might be wrong about being capped at some level of arpen as well, which we can see from Dysent's OP that GC was replying to. He was fighting a target with 21544 armor, and saw 66.24% armor reduction with 116.5% arpen. If we calculate the arpen "cap" for a target with 21544 armor, we get (21544+15232.5)/3 = 12258.83. That means if ArPen capped at 100% and couldn't reduce armor any further, you'd see 12258.83/21544 = only 56.9% reduction in armor. But, if arpen can go past 100%, 116.5% reduction of 12258.83 would reduce 14281.54 armor, which is a reduction of 14281.54/21544 = 66.29%, close to the 66.24% reduction he stated.
This is easy enough to test on anything. You just need to get gear+GT proc to reduce armor by ~100%, and then get a little more arpen to reduce it by 110% or something. If damage goes up, you clearly can have more than 100% arpen and reduce it past the "cap". If this is the case, a boss mob with 10643 armor and Sunder/FF applied would require 104.05% arpen(sum of rating + bstance + mace spec %) to bring its armor to 0. 10643*.95*.8 = 8088.68. (8088.68+15232.5)/3 = 7773.73. 8088.68/7773.73 = 1.0405
I remember Dysent saying that he was hitting a paladin (was via AIM so not sure) and GC indicated that ArP cap changed with the class.
I remember Dysent saying that he was hitting a paladin (was via AIM so not sure) and GC indicated that ArP cap changed with the class.
I read that as the effective ArP cap changing with mob class, which effects their armor a la the caster vs. melee boss armor split back in the TBC days.
According to the blue post, it seems like the magic number for bosses (lvl83 warrior mobs) is 8317:
That really doesn't really match with the formula he gave, though.
If boss armor is 10643 and L83 C is 16635, the cap should be 8992. Assuming his value of C is accurate, for the cap to be 8317 using that formula boss armor would have to end up being 8316.
Or am I just completely missing something?
EDIT: Nevermind, just checked out the combat ratings thread. GC was wrong and the formula uses attacker level? That makes sense.
not sure about the combat ratings page, but GC's post states that blizz uses the *lower* of the two numbers ( (A+16653/3) vs A ) ). In the case where A = 8317 the two quantites are the same. In any case where A > 8317, blizz continues to use the 8317 value.
Er... not so much, no.
As A increases, the cap increases as well, just at a much slower rate. A = 8317 is the breakpoint if you're using the 16653 value, although testing seems to indicate that ArP calculations instead rely on attacker level, which puts the breakpoint at 7616.25 from C = 15232.5. Either way, though, as A increases past the breakpoint the ArP cap will also increase at A / 3. If a boss had 20,000 armor using the formula, the ArP cap would be 11,744.
Plugging in the boss armor value of 10643 and C of 15232.5, you end up with a cap of 8625.16, or almost exactly 81% armor reduction, which is what had been determined previously.
If I'm not getting it wrong with sunder and ff you'd get:
(8088.68+15232.5)/3 = 7773.56 (104.05% to cap)
You'd need around 19.34% less arp to cap (should be 123.99% without buffs).
Supposing it is possible to overcap and it seems you can from Dysent numbers, then we should reconsider the softcap for Grim Toll and try to test if at 104.05% (would need 44.36% arp in battle stance, 29.36% as mace spec).
The buffs still seems to work additively with gear.
I'll do some tests tomorrow to check what's (and if there is) a hard cap.
It's not that easy to collect data especially cause you can only reach such high arp just with GT procs and they only last 10s, so some sample tests with 123.99%+ with no debuffs, 104.05%+ w/sunder+ff (or 107.57% with only sunder up) could help much.
Right I have trolled till my eyes blead, spent a small fortune in gold on respecs and gems etc in the warm up and live patch and am totally at a loss.
Am I right in deciding that ArP IS the way to go vs. Str? Or is there a cross over point where you need a base passive ammount of ArP before re-geming from 16 str to 16 ArP will be a benifit?
I have been running the spread sheet with my gear and when I change to ArP gems, I loose DPS, both with full raid buffs and under self buff situations.
I have been gimped to Colossal skull clad cleaver and Jawbone as my options of spec / weapon. In a naxx 25 run as axe spec with str gems I was generally out dpsing my fellow Arms warrior who was Mace specced with jawbone, however, I do doubt his general understanding and of both mechanics and propepr gemming - so i take this with a pinch of salt. And furthermore, the mace comes out marginally on top on the spreadsheet, in both ArP and Str gemming situations
I have already spent 2k on gems and respecs trialing this in game as well as on the spreadsheet. I have reaced the conclusion that Axe > mace , since I can pull equal dps in game as with Jawbone.
If I have over looked any thread that explains any form of base stat requirments for gemming ArP or Strength, kindly point me in the correct direction.
I read the blue post concerning ArP so it is working as intended, so I am lead to believe that ArP may well be the gem of choice since PTR was raving over this.
Regarding whether we use 16635 or 15232.5 vs bosses,
If it's calculated against 16635, the armor value is 11623 for bosses. This gives us (a+b)/3 = .8104
If it's calculated against 15232.5, the armor value is 10634 for bosses. This gives us (a+b)/3 = .8108
A good way of figuring out which-is-which would be to use the armor of a beast boss that is known (hunters can determine). Basically get a hunter to give us the armor of the big dinosaur boss in DTK heroic. Then we hit that boss with a BT before the debuffs go up on it and work backwards. That'll give us some finality on non-80 targets.
Also, I definitely got better numbers as I exceeded 100% ArP buffed, as I look at what I've gotten in the past. At 1127 ArP. It's clear that both mace and battle must allow ArP to exceed the cap. That makes the math VERY interesting for those two things.
Base BT 1445, ArP 1127 + 10% +15%:
1442 dmg to lvl80 dummy at 9729 armor, amoutns to 99.67% armor ignored
Formula expects x = 1.165 * .855 = .9964
Formula capping x = 1.00 * .855 = .855 ignored, where I should have hit for 1322.69 if x was capped at 100% including stance/mace.
Also works for hitting the pally.
Base BT still 1445, ArP still 1127, his armor 21544 (suppression factor is .569014)
All the results agreed that I wasn't being capped at 100% including mace/battle.
Once I get one more ArP piece, I'll be able to test 100% in zerk w/o mace or battle (most I can hit atm is 98% tooltip... so close!). It'll be interesting to see if > 100% zerk gives same number as 100% zerk to see if it caps at all.
[edit: went back to the old results and I can't make them match up for any of the "including sunder" tests when using 11623 / 16635... I really want to test sunder more with high ArP values (between 100-125%) and make sure the existing formula holds, using Rallik's original formula but with Z = (A+B)/3 for the suppression factor applied to x instead of .81]
Dysent, thank you so much for your testing and analysis on this... So do you know what the "magic number" is for a warrior in battle stance, but not mace spec? How low in terms of armor ignored can you theoretically get? Also, do we know how sunder and faerie work with arms yet in terms of these caps? I'm still having trouble wrapping my head around this new math, heh.
What I mean to ask is, what does x equal with only battle stance but not mace spec?
It's starting to look like Blade Ward is useless in comparison to Mongoose. However I would like to get some solid numbers together to confirm some of my assumptions. I could use a few pointers on the best way to test it as I've never done this sort of testing before; here is what I was thinking:
*Test One: Standing behind training dummy, auto attack to discover how often it will proc on attack and if there is an internal cooldown. Question: What would be a good amount of time/attacks?
*Test Two: Standing behind dummy, auto attacks combined with yellow attacks to see if instants have any impact on proc rate. Question: Should I just wait until 100 rage and then Devastate down to nothing?
*Test Three: Attacking a high level mob with no expertise to confirm that a parry consumes the proc and puts it on an internal cooldown. Question: I gather auto attack should work here, is there a good mob I could test this on? Assume that I can dual-box a healer.
04/17/09, 10:15 PM
Similar Threads
