Combat Ratings at level 85 (Cataclysm) - Page 30

Mavanas · 04/19/09, 1:33 PM

The cap is not really a cap in the sense that it will prevent you from reducing the armor to zero. It is a way to dampen the effect of armor penetration for high armor targets.

S	F	SB	BS	MS	Cap	APEN
0	0	0	0	0	8625,17	1520
1	0	0	0	0	7915,63	1325
1	1	0	0	0	7773,73	1282
1	1	0	0	0	7773,73	1282
1	1	0	1	0	7773,73	1159
1	1	0	1	1	7773,73	974
1	1	1	1	1	7448,68	924
1	1	0	1	0	7773,73	1159
1	1	0	1	1	7773,73	974
1	1	1	1	1	7448,68	924
1	1	1	0	1	7448,68	1047
1	1	1	0	0	7448,68	1232
0	0	0	1	0	8625,17	1397
0	0	0	0	1	8625,17	1336
0	0	0	1	1	8625,17	1212

Legend:
S = five sunders; F = faerie fire; SB = serrated blades (rogues); BS = Battle Stance; MS = Mace Spec; Cap is the minimum of Boss armor after sunder, FF and SB reductions, and (Reduced Boss Armor + A)/3 as described in GC post; APEN is armor penetration rating from gear required to reduce boss armor to 0.

I included most common combinations of armor debuffs, but not all possible.

P.S. Normally as Aldriana indicated on the previous page the cap is equal to (Reduced Boss Armor + A)/3. However, under extreme cases of armor reduction, this cap can actually change to simply Reduced Boss Armor. So for instance if Reduced Boss Armor includes the 640 armor reduction from Serrated Blades (this is assumed in the table above and is based on Aldriana's observations), then with full sunders and Faerie Fire, the minimum cap changes to Reduced Boss Armor.

Also if you add Shattering Blow and it's also included in the Reduced Boss Armor, together with sunder the minimum will also be Reduced Boss Armor and not (Reduced Boss Armor + A)/3.

Thus assuming Shattering Blow is a debuff similar to sunders and faerie fire, the final formula should be:

$\frac{A}{A+[B(1-y)-SB]-x*CAP}$ ,

where

$1-y = (1-s)*(1-ff)*(1-ST)$ , a multiplicative combination of sunders, faeire fire and shattering throw;
$CAP = min([B(1-y)-SB],\frac{[B(1-y)-SB]+A}{3})$

landsoul · 04/20/09, 2:44 AM

So... where's your proof that you went over the cap with 100% armor penetration, got more than 100% armor penetration, and still removed armor passed 100% penetration? You say that it's a measure of effectiveness but you have no data or test to show? You basically summarized a bunch of information which was already talked about which is fine, but what's the answer, and where's the test?

hellord · 04/20/09, 7:13 AM

I didn't do all the tests yet, but Dysent's numbers seems to suggest so.

Target armor = 21544
expected cap = 12258.83
expected cap % = 56.9%
reduction with 116.5% arp = 66.29%

The value is very close to what Dysent shows in his post, suggesting you can "overcap" the value from the formula.
With a 21k armor target there should be a hard cap of 175.74%, so those 16.5% arp over 100% still seems to grant around 9% damage reduction decrease.

Dontmindme · 04/20/09, 12:22 PM

Originally Posted by Rallik

For the sake of completeness, I did some short testing today to lock down the boss armor value precisely at 10643.

Test method:
1) Decide if armor is less than or equal to 10643 or greater than or equal to 10644. This was done by finding a borderline level of AP such that if armor was 10643, the possible damage values for BT shifts down by 1 compared to 10644. The AP value I used for this test was 3214, giving an armorless BT damage of 1607.

1607*15232.5/(15232.5+10643) = 946.016, crit = 1892.031
1607*15232.5/(15232.5+10644) = 945.979, crit = 1891.958

So, if there is either a 947 hit or a 1893 crit, the armor value must be less than or equal to 10643. If there is a 945 hit or a 1891 crit, the armor value must be greater than or equal to 10644.

After a few dozen BT's landing for 946 and 1892, I saw a 947 hit, meaning the armor must be less than or equal to 10643.

2) Decide if armor is less than or equal to 10642 or greater than or equal to 10643 by the same method as above. The AP value I used for this test was 3044, giving an armorless BT damage of 1522.

1522*15232.5/(15232.5+10642) = 896.012, crit = 1792.024
1522*15232.5/(15232.5+10643) = 895.977, crit = 1791.955

After a few dozen BTs, I saw a crit for 1791, which means the armor value must be greater than or equal to 10643.

Since the armor value must be less than or equal to 10643 and must be greater than or equal to 10643, it can only possibly be 10643.

Actually, I see a flaw in the logic here. It is my understanding the damage works as follows:
You have the damage of the attack +/- 0.5 then following natural rounding as your base damage. As evidence, I recall a bow from vanilla wow which did a fixed amount of damage, but when you check the tooltip, you gain a damage range one apart. (I had done other testing that made me believe this was true as well)

So, in the above example, conclusion #1 is still correct.
Solving for the maximum value gives:
1607.5*15232.5/(15232.5+x)=946.5 {the minimum value that would round up and still show 947 damage}
becomes
1607.5*15232.5/946.5-15232.5=x=10637.81 {maximum possible armor to show 947 damage}
doing the same for the minimum
1606.5*15232.5/(15232.5+x)=946.5 {the maximum value that would round down and still show 946 damage}
becomes
1606.5*15232.5/947-15232.5=x=10621.71 (minimum possible armor to show 946 damage)

So far we are between 10621.71 and 10637.81
-----------------
Calculating for the 1892 crit
(1607.5*15232.5/(15232.5+x))*2=1891.5 {the minimum value that would round up and still show 1892 damage}
becomes
1607.5*15232.5/945.75-15232.5=x=10658.32 {maximum possible armor to show 1892 damage}
doing the same for the minimum
(1606.5*15232.5/(15232.5+x))*2=1892.5 {the maximum value that would round down and still show 1891 damage}
becomes
1606.5*15232.5/946.25-15232.5=x=10628.54 (minimum possible armor to show 1892 damage)

So we are now between 10628.54 and 10643.43

-----------------
Test 2 is where this goes wrong
Calculating for the 1791 crit
(1522.5*15232.5/(15232.5+x))*2=1790.5 {the minimum value that would round up and still show 1791 damage}
becomes
1522.5*15232.5/895.25-15232.5=x=10672.53 {maximum possible armor to show 1791 damage}
doing the same for the minimum
(1521.5*15232.5/(15232.5+x))*2=1791.5 {the maximum value that would round down and still show 1791 damage}
becomes
1521.5*15232.5/895.75-15232.5=x=10641.07 (minimum possible armor to show 1791 damage)

Thus this test is only proving the armor value between 10641.07 and 10643.43.
We are not pinned down further than that...

dysent · 04/20/09, 2:45 PM

Originally Posted by hellord

I didn't do all the tests yet, but Dysent's numbers seems to suggest so.

Target armor = 21544
expected cap = 12258.83
expected cap % = 56.9%
reduction with 116.5% arp = 66.29%

The value is very close to what Dysent shows in his post, suggesting you can "overcap" the value from the formula.
With a 21k armor target there should be a hard cap of 175.74%, so those 16.5% arp over 100% still seems to grant around 9% damage reduction decrease.

Yeah, I did a bunch of data at 1127 ArP (self) + mace + battle. I went back to those numbers with GC's formula and the only way I can come up with them is if I'm getting the entire 116.5%. If anyone wants to run them themselves, here's what I got at the top end:

1445 BT base,

_____________	_________	_________	_________
Target Armor	9729	13954	21544
dmg@926 ArP	1315	1124	893
dmg@1087ArP	1417	1212	962
dmg@1127ArP	1442	1233	978

Numbers were gathered from: Lvl80 dummy (9729), paladin w/ 2h (13954), paladin w/ shield (21544)
Paladin did not use his 6% dmg reduction, nor did he use devo aura, since I wasn't sure how it would interact.

Those are the base tests that the percentages Hellord quotes were built on.

The only way those calc out properly is if I don't cap x at 1.00 after I add the 123 (battle) and 185 (mace). As I was saying in warr thread, I'm getting close to being able to test 100% tooltip ArP in zerk w/o a mace. Once me (or anyone else) can do that, we can determine if you can exceed 1.00 all the time, or only with battle/mace/etc.

Mavanas · 04/20/09, 5:36 PM

Dysent,

are you sure there is no typo in the armor penetration rating in the first line. Using the formula, I can match the reported BT damage for 1087 and 1127 armor penetration with 0.05% accuracy, while the reported damage for 926 armor penetration is below the theoretical one by 6-9 damage (0.7% discrepancy).

If you had 909 armor penetration in that first line, the formula would work with 0.05% accuracy for all 3 levels of armor.

Other explanation of course is that the formula is incorrect, and mace/battle stance and armor penetration interact in some other way, not additively.

Mavanas · 04/21/09, 10:56 AM

I know Rallik has done extensive testing of boss armor on heroic dummy using BT, but here is just another independent confirmation. I asked a hunter to use Beast Lore on Sanctum Sentry, the boss level cat add of Auriaya in Ulduar. The armor is exactly 10643, which matches Rallik's testing.

Dontmindme · 04/21/09, 1:33 PM

Which merely narrows it down to a number that will round to 10643. Once upon a time, armor values tended to be very even numbers (ending with a 5 or 0). Now they are not, but this leads to the question of what has changed.

Looking at an armor value given by Ghostcrawler for a level 80 we have 9729. This is a very interesting number in that it happens to be divisible by 9. Now, at some point we heard about a 10% armor reduction for mobs. Divide 9729 by .9 and you get 10810, a very even number. Now do the same with 10643 and you get 11825.555555 but lets look at 11825*.9=10642.5 which would naturally round to 10643 for display purposes.

Thus, I'm still not convinced the correct number isn't 10642.5.

Rallik · 04/21/09, 7:41 PM

Originally Posted by Dontmindme

Actually, I see a flaw in the logic here. It is my understanding the damage works as follows:
You have the damage of the attack +/- 0.5 then following natural rounding as your base damage. As evidence, I recall a bow from vanilla wow which did a fixed amount of damage, but when you check the tooltip, you gain a damage range one apart. (I had done other testing that made me believe this was true as well)

Your assumption that normal rounding goes on is incorrect. Damage is rounded probabilistically. If an attack should have dealt 100.1 damage, it will deal 100 damage 90% of the time and 101 damage 10% of the time. If this ability crit, its damage would become 200.2, and it would deal 200 80% of the time and 201 20% of the time. That's why I used AP values where a +/- 1 armor shift would cause BT damage to fall into a different range.

I did assume that armor only had integer values. If that is true, armor is exactly 10643. If it is not true, then it would be easy enough to begin testing in the same fashion I used to nail it down to an arbitrary degree of accuracy. Finding AP levels that give you those nice breakpoints is the hard part though.

Edit: Fixed flip/flopped percentages

drumbum · 04/22/09, 8:07 AM

Originally Posted by Rallik

Your assumption that normal rounding goes on is incorrect. Damage is rounded probabilistically. If an attack should have dealt 100.1 damage, it will deal 100 damage 10% of the time and 101 damage 90% of the time. If this ability crit, its damage would become 200.2, and it would deal 200 20% of the time and 201 80% of the time. That's why I used AP values where a +/- 1 armor shift would cause BT damage to fall into a different range.

It would actually be 101 damage 10% of the time, and 100 damage 90% of the time -- I'm sure this was what you intended to say, but I wanted to clarify for anyone else trying to understand how it works. (Similarly, your second example would be 201 damage 20% of the time, and 200 damage 80% of the time.)

Rallik · 04/22/09, 4:27 PM

Originally Posted by drumbum

It would actually be 101 damage 10% of the time, and 100 damage 90% of the time -- I'm sure this was what you intended to say, but I wanted to clarify for anyone else trying to understand how it works. (Similarly, your second example would be 201 damage 20% of the time, and 200 damage 80% of the time.)

Err yeah, haha said it backwards, but that's what I meant. I'll edit the post, thanks.

Slafsinator · 04/23/09, 5:00 PM

Originally Posted by landsoul

So... where's your proof that you went over the cap with 100% armor penetration, got more than 100% armor penetration, and still removed armor passed 100% penetration? You say that it's a measure of effectiveness but you have no data or test to show? You basically summarized a bunch of information which was already talked about which is fine, but what's the answer, and where's the test?

We did some tests on our warriors and posted here about them, I'll quote the post as we found some interesting data applicable to any class -- you can reduce armor below 0 with armor penetration above 100% and damage will keep increasing. Findings in the quote below:

Data
Attack power: 2745
Gear + food + elixir: 50.91% (627 rating)
Character sheet ArPen with Grim Toll proc: 100.60% (1239 rating)
Battlestance: Yes (makes it a total of 110.60%)
Sunder Armor: 5x
Thunder Clap, base value: 629.4
Heroic Throw, base value: 1385

No damage modifying talents such as Blood Frenzy, Wrecking Crew or 2H weapon spec was in effect.
You can see what items and talent spec I used on my armory profile (only for today though, I'm not going to be allowed in raids with my current spec!

We used this spreadsheet for numbers.

Results
Against a level 80 mob with 9729 armor (Hulking Abomination in Icecrown, just outside Dalaran)

Damage values with Grim Toll proc and 5x Sunder Armor
Thunderclap hits for: 660 damage
Heroic Throw hits for: 1452 damage

This matches our data of -702 armor on the target.

Kalroth · 04/23/09, 6:28 PM

Originally Posted by Slafsinator

We did some tests on our warriors and posted here about them, I'll quote the post as we found some interesting data applicable to any class -- you can reduce armor below 0 with armor penetration above 100% and damage will keep increasing. Findings in the quote below:

Just to follow up on our findings. I got some more ArPen gear and managed to reach 59.60% ArPen from gear + elixir and food, which brings me to a total of 119.29% ArPen including battle stance with Grim Toll proc up. My armory is updated with the gear and talents I used for below test.

If there is a cap, then it's very high. :)

Talents: 0/0/0
Attack power: 2815
ArPen from gear + gems + elixir + food: 734 rating (59.60% ArPen)
ArPen from gear + gems + elixir + food + Grim Toll: 1346 rating (109.29% ArPen)
Battlestance: Yes (119.29% ArPen)
Sunder Armor: 5x (20% debuff)
Heroic Throw base: 1420
Thunder Clap base: 637.8

* Level 80 mob with 9729 armor (Hulking Abomination in Icecrown, right outside of Dalaran)
All available ArPen buffs and 5x Sunder Armor:
Heroic Throw hits for: 1559-1560 damage (3212 crits, with 3% meta gem)
Thunder Clap hits for: 701 damage
-1368.6 armor

* Level ?? boss with 10643 armor (Heroic Training Dummy in Orgrimmar)
All available ArPen buffs and 5x Sunder Armor:
Heroic Throw hits for: 1512 damage (3114 crit, with 3% meta gem)
Thunder Clap hits for: 679 damage
-928.2 armor

Slafsinator · 04/25/09, 6:44 PM

A peculiar effect of stacking Armor Penetration to the point of negative armor is that the closer the base/debuffed armor of the target is to the so-called Armor Penetration cap (see Ghostcrawler's post), the more effective Armor Penetration is. Using Shattered Throw to reduce the target's debuffed armor further than with Sunder/Faerie Fire can decrease the DPS of those having above 100% Armor Penetration.

To simplify it -- in one case you debuff someone to 3000 armor and you get 200% Armor Penetration. You end up at -3000 armor on the target -- great. In the other case you debuff the target to 8000 armor and you still have that juicy, tasty 200% Armor Penetration. Result: -8000 armor on the mob.

Omnicia · 04/27/09, 4:58 PM

I can't imagine blizz will allow ArP to continue to operate as it currently does. the nature of the mechanic as it is now just seems un-blizzlike