Get the Desktop App for Battle.net Now
- All your games in 1 place
- Log in once
- Automatic game updates
Will +50 armor when you have 1000 armor give you the same damage reduction as +50 armor when you have 9000 armor? Or do later armor amounts have diminishing returns so that it's better to get a little bit of resist and armor and dodge? Or does it not diminish and it doesn't matter if you try to stack just pure armor?
I know that the %dmg reduction it gives you goes down, but +1% damage reduction at 50% damage reduction is twice as good as +1% damage reduction at 0% damage reduction.
If you don't understand the 2nd paragraph please don't reply. :)
The displayed damage reduction amount will go down, so there are diminishing returns in that sense, but effectively there are no diminishing returns. Here's why:
(I'm going to be pulling some numbers out of my !@# but bear with me)
Say @ 1000 Armor you have 30% Damage Reduction. An additional 50 Armor might get you to 31% Damage Reduction, or a 1% increase.
@ 7000 Armor you might have...60% Damage Reduction. 50 more Armor gets you to 60.55% Damage Reduction, or a 0.55% increase.
50 Armor seemingly gave you less DR than before, right? Nope! The numbers are misleading.
@ 1000 Armor, you went from 30% DR (70% damage taken) to 31% DR (69% damage taken).
---->69/70 = 98.6% less damage taken.
@ 7000 Armor, you went from 60% DR (40% damage taken) to 60.55% DR (39.45% damage taken).
----> 39.45/40 = 98.6% less damage taken.
Identical results, as you can see. Looking at EHP is also perfectly valid and is essentially an extension of what I did here.
EDIT: You answered your own question in the second paragraph. Though others have found that since it's easy to stack at lower levels (ie easier to get more armor when you're only at 3000 to begin with, or easier to get resists when you have none), it's optimum to have similar damage reduction from Armor and from Resists.
10 Armor = 1 Resistance in terms of EHP. So optimally you have something like 6000 Armor and 600 Resist All, or 7000/700, etc etc.
At 6000 Armor and 500 Resist, for example, adding 100 Resist would give you more overall DR than adding 1000 Armor, I believe.
Edited by AlphaMagnum#1559 on 5/30/2012 9:14 AM PDT
That' doesn't make sense though. Your entire post I agree with alpha except the very last sentence. I would assume it wouldn't matter.
For example: If you are at 3000 armor and 0 resist. Getting either +1000 armor or +100 resist should be the same damage reduction. From what I was believing it really doesn't matter what your current armor/resist is. Adding x armor/resist is always going to have y damage reduction.
Ucross, you are wrong.
There arent diminishing returns on stacking DR, but there *are* diminishing returns on the amount of DR provided by changes to your Armor and Resistance stats.
Damage vs level 63 mobs:
1000 armor = 75.90%
2000 armor = 61.17%
3000 armor = 51.22%
4000 armor = 44.06%
The additional DR provided by adding 1000 armor is
1000 armor, adding 1000 = 19.42%
2000 armor, adding 1000 = 16.26%
3000 armor, adding 1000 = 13.99%
The diminishing returns are plainly evident.
The numbers for Resistance are exactly the same, except resistance is worth 10 times as much as armor (instead of 1000, 2000, 3000, 4000, ... its 100, 200, 300, 400, ...)
Now, adding armor always increases DR, and adding Resistance always increases DR. However, if your starting Armor is not precisely 10 times your Resistance then increases one of them has greater DR benefits than increasing the other precisely because of those diminishing returns.
If your ratio is 11:1 then the Armor stat is further along the diminishing returns than Resistance.
if your ratio is 9:1 then the Resistance stat is further long the diminishing returns than Armor.
^ This last post should be a sticky.
Adding armor/resists does provide a linear percentage reduction of damage taken, however, linear percentage is different than linear base value. Currently, the more armor you have means you take less damage but this also means the actual value of that linear % reduction per armor point becomes a smaller number the more you have. Essentially, each point of armor does less work the more you have (becomes inneficient to stack).
Noting Dusty's 10:1 ratio, this is a build based on perfect damage reduction efficiency. In practical terms, I would venture to guess that these exact numbers are unattainable unless you invest massive amounts of time and/or money. Having 11:1 or 9:11 isnt "bad" because really it depends on what those numbers actually are. Having larger numbers on a 9:1 ratio can potentially be better than lower numbers on a 10:1 scale (so don't panic if you dont have 10:1).
I know that Dusty isn't claiming that an unoptimized ratio is bad, I just wanted to let the people reading know that they can still be doing well without it. Dusty's post is da bomb.
**edited for spelling.
Edited by Grenolf#1756 on 5/31/2012 2:33 PM PDT
Er, I wouldn't call that diminishing returns, however.
Effective HP (how big a hit you can take in "raw" damage before keeling over) is bilinear in armor and resistances. That is, if you hold one constant you get linear returns on the other. Sure, you get the best overall damage reduction by keeping the damage reductions from armor and resists equal... that is not diminishing returns ;/
Edit: Diminishing returns would be something like EHP = base_hp * (1+const*sqrt(armor)) * (1 + const* sqrt(resist)) or something explicitly nonlinear. That is not the case.
Edited by Feynman#1683 on 5/31/2012 3:25 PM PDT
Diminishing returns = anything that the more you do, the less you get back (its sort of like overfishing :D)
Diminishing returns on stuns are a common example throughout many games. Each successive stun stuns for a shorter amount of time until the unit being stunned can no longer be stunned. Considering that the more armor you use, the less effective each point of it becomes, I would call this a diminishing return effect. This is what makes the 10:1 ratio so effective (50:50). If armor didn't have diminishing returns, resists would effectively be useless.
Additional resist or armor past the first point is subject to diminishing returns regardless of the ratio of armor:resists.
With all this being said, would I be correct to state that 900 resist with 6000 armor is not as good as 1300 resist with 6000 armor? Common sense would tell me this is true.
Can anyone calc out the difference in damage reduction for these amounts? Also, wouldn't 1200/6000 be better DR than something like 400/4000?
Yes to both your questions. Both would provide a greater damage reduction than the other.
However, 400/4000 is more efficient than 1200/6000 in the sense that it optimises the diminishing returns problem. By sheer numbers 1200/6000 is more effective in damage mitigation than 400/4000 (provides a larger % reduction overall) but is inefficient because you can probably attain a greater damage reduction doing say 660/6600 (rough number) at a theoretically similar cost assuming 1point of resist costs the same as 10points of armor.
Note that effective =! efficient.
Edited by IKICKKITTENS#6127 on 7/26/2012 7:20 PM PDT
the one thing no one mentioned here is that from a gear point of view, armor is ridiculously hard to get. you can get like 400 extra armor on an archon piece with +300 armor over a similar lvl 62 piece. you can get 110 phys resist on one piece pretty easily. you are naturally going to end up with small amounts of armor relative to your resists unless you want to use defensive passives.
Just to add to this slightly (as well as once again mention my calculator), if anyone wants to see the effect of the 'diminishing returns' formulas on armor or resistances (or both) on their time-to-live, as well as the relative value ratio of armor:resistance based on their actual stats and (more importantly) buffs, that calculator is available here:
For example, try adding 1000 to the armor field in the calculator, and looking at the time-to-live bonus for 1 armor.. it's the same as it was before. i.e. each point of armor has a linear effect on your time-to-live, even though your actual percentage mitigation per point of armor is decreasing.
As another example, try switching the calculator to output values as 'x:armor'.. it will show the relative value of each stat and skill as a number of armor-equivalents. For most Monks, with a much higher bonus to armor % than resistance %, the relative value of these two stats is quite a bit different than the 10:1 ratio often talked about.. usually closer to 6:1.
Here's the best I can explain it.
Armor increases your HP effectively by 100% for every 3000 armor.
Resistance increases your HP effectively by 100% for every 300 all resist.
For example 0 Armor = 1x HP, 3000 armor = 2x HP, 6000 armor = 3x and so on.
For All Resist, 0 All Resist = 1x HP, 300 = 2x HP, 600 = 3x HP and so on.
These effects multiply, therefor the most effective way to gain effective hp is to balance the two.
Examples of calculating effective HP:
Example 1: 900 All Resist, 6000 Armor.
4x bonus from All Resist, 3x from Armor. Effective HP is 12x base HP or 91.67% resistance
Example 2: 600 All Resist, 3000 armor
3x bonus from All Resist, 2x from Armor. Effective HP is 6x base HP or 83.33% resistance
If you want to talk about the ratio of effect for armor vs all resist, it is exactly 1:10 as long as your all resist total is exactly 1/10 of your armor.
So are the effects of armor and resistance diminishing? No. I would actually argue that the returns from armor and all resist are augmented the more you get.
300 AR / 3000 Armor = 4x Effective HP
600 AR / 6000 Armor = 9x Effective HP
900 AR / 9000 Armor = 16x Effective HP
1200 AR / 12000 Armor = 25x Effective HP
Edited by Noobasaurus#1377 on 7/31/2012 3:11 PM PDT
Threats of violence. We take these seriously and will alert the proper authorities.
Posts containing personal information about other players. This includes physical addresses, e-mail addresses, phone numbers, and inappropriate photos and/or videos.
Harassing or discriminatory language. This will not be tolerated.