Going to try to have some time to read that. :P thanks guys for all your help. :D
Printable View
Going to try to have some time to read that. :P thanks guys for all your help. :D
Hhahahahha so much false info in one post. Again you play an expert on forums when you have no idea what you're talking about.
If you were able to figure out anything from YOUR own experience... but no. How couldn't you notice that diminishing returns start around 3850-3900 resilience, not 4200? At 4,2k you start getting less than 1% dmg reduction per 100 resilience, which leaded you to conclude that thats the point where dr starts. But its starts giving less than before around 3,9k, so thats where diminishing returns start.
As for 4.3 resilience. This is just false accusation, and you prolly know it. 4.3 NERFED resilience just the way they did on molten. Thats how it should be, unfortunately. And thats one of main reasons this patch is so ******ed for pvp. Resilience was slightly increasing in value in 4.0.6, now its roughly the same until that ~3900 (maybe 3860, something around that), and then starts to rapidly decrease.
You could at least google if u dont have any idea but urgently wanna write a post...
So I posted the wrong value. But still.
Just appears that the difference between 3.9 and 4.2 was that much is wasn't even noticeable until after 3.9.
You know some people do get things wrong ever now and then, you don't need to be a complete dick about it.
Quote:
Resilience scaling has been modified for linear returns, as opposed to increasing returns. Under the new formula, going from 30 resilience to 40 resilience gives players the same increase to survivability as going from 0 to 10. Resilience now scales in the same way armor and magic resistances do. A player with 32.5% damage reduction from resilience in 4.0.6 should see their damage reduction unchanged in 4.1. Those with less than 32.5% will gain slightly. Those with more will lose some damage reduction, increasingly so as their resilience climbs.
Input your values and you can compare it yourself.Quote:
1 - 0.99^(Resilience/79.12785)
Multiply with 100 and add % at the end and you get by how much % does damage get reduced from resilience.
Ah o.o so an exponential function. Math time :o
With
y = damage reduction
x = Resil
the function we have is
Y= 1 - 0.99^(x/79.12785) = 1 - 0.99987^x [.99^(1/79.12785)=79.12785th root of .99]
Y' = - ln(0.99987) X (0.99987^x)
Y" = - {ln(0.99987)}^2 X (0.99987^x)
Now since the second derivative of this function is megative from 0 to 5500. Our function Y has a downward concavity throughout [0;5500]
=> the function has a downward concavity throughout (0 =< x =< 5500).
"DR" starts as soon as you receive resilience. It's just not noticeable till the further end of our range.
Also note: to get a damage reduction of 50%, you need somewhere between 5400 to 5500 resil.
5400 gives 49.63% and 5500 gives 50.27%.
just before anyone start to flame otherwise, blue's function is basically a re-writen function of the one given in Original's link. cept his link gives an easier function to operate with o.o (using e and ln)
The graph of the function (taken from the guide) matches the one I have on my calculator.
http://img26.imageshack.us/img26/5240/resilience01.jpg
It's less noticeable because of the amount you're already mitigating. Honestly, I'm not even sure why people are busting out their math. It's not a mathematical issue. It's a logical issue. The amount of mitigation you actually get per point of resilience goes down (diminishing returns) as you gain more of it, but the value of every additional point of resilience increases exponentially. Obviously you're not going to notice much between 3.9k and 4.2k resilience since you're adding more mitigation to an already large amount of mitigation and you are well over your break point. But this doesn't change that for each extra 1% of mitigation you gain, your effective health increases exponentially.
Here, let me quote DamnOriginal's link.
"What's the effect of 25% extra mitigation when you already have 0% or 25% mitigation? If you have 0 resilience your mitigation is obviously 0%. If you add 25%, for each 100 damage you're dealt before mitigation is applied, you take 75. You've logically reduced your damage by 25%. What happens if you go from 25% to 50%? Instead of taking 75 you take 50 damage, so by increasing your mitigation by 25% you've actually reduced incoming damage by 33%."
The reality of it is, when you increase your resilience from 25% to 50%, you're not decreasing your damage intake by another 25%, but by 33% against the damage you were taking before. Thus, the more resilience you have, the more it is worth.
As I said, it's an issue of logic. The involvement of math is minute and makes no real difference.
Actually, the math proves that you gain less mitigation as you increase resil. As soon as i get back to my desk ill calculate mitigation % rate of increase in increments of 20s for 5 intervals on that chart.
Im not saying youre wrong grace, im just saying youre arguing a totally different argument than what I was trying to prove.
This is actually very hard to grasp, logic involved. Im still not sure about that but one may say it makes no sense. How could it apply new amount of damage reduction to previous amount, not to your basic (aka 0) value? If im not wrong, no other stat works like that. Your haste rating increases your haste by X % of default, not of the value you got with current haste, and so.
Lol its so tangled but i hope someone gets the point.
Btw speaking about 25 and 50% resilience is misleading, you and damn surely meant damage reduction. But going from 25 to 50% dr requires a lot more resilience increase than from 0-25% dr, and thats the whole point.
People will always claim resilience is bugged so long as they feel they die too quickly. This being a private server, that sort of thing tends to become a scapegoat. That being said, nobody doubts that resilience does indeed DR like it is supposed to. But when it comes to the actual worth of a certain amount of resilience, sure, you can say "it mitigates X amount of damage", but what does making that statement do? All it does is give you a predetermined number that won't even cross your mind while you're eating a Chaos Bolt or having a Ret Paladin with full cd's popped sticking to you like white on rice. The actual numerical amount that resilience is worth is irrelevant (in my opinion, I guess?) because it is not these values people think of. Instead, they think of how much they feel their resilience is worth and whether or not they feel are taking too much damage.
So then when they feel they are taking too much damage, or they feel that there is no point in stacking resilience past X amount, they use a statement such as "because resilience has diminishing returns and has no effect at X amount" to back up their argument. I don't doubt that Toshika has found a "sweet spot" that is right for her, where she takes little enough damage due to her amount of resilience so that she can stack damage stats instead. That's perfectly fine. But to state that resilience is worthless because of dimishing returns is a silly fallacy based in nothing but a misunderstanding of how the mitigation truly works.
Armor works exactly the same way against physical attacks with one exception: mitigation through armor has a cap at a total of 75%. I know that it isn't exactly the easiest thing to grasp. It's not because it has weird mechanics, though. The mechanics of the stat are fairly straight-forward. It's just measuring the actual value of the stat as you gain more of it is more complex.
Normally I would simply say that "Resilience's worth increases at an exponential rate the more of it you gain", and hope that people "trust" me and take my word for it, because that is the easiest way to explain it without trying to use numbers for illustration.
But here, I'll try numbers, my own illustration of what I quoted from DamnOriginal's link.
Mitigate 25% of 1,000.
Result is 750.
This is your current resilience.
Add another 25% to your resilience.
Result is mitigating another 250 damage from 1,000.
But comparing that 250 to the 750 you have already been taking, 250 is one-third of 750, which is 33%.
(33% less damage taken with 25% gain in resilience from 25% base resilience.)
Example #2 with more base resilience.
Mitigate 50% of 1,000.
Result is 500.
This is your current resilience.
Add another 25% to your resilience.
Result is mitigating another 250 damage from 1,000.
But comparing that 250 to the 500 you have already been taking, 250 is half of 500, which is 50%.
(50% less damage taken with 25% gain in resilience from 50% base resilience.)
Yes, diminishing returns certainly requires that you need much more resilience to reach a higher percentage of damage resuction as you gain more of the stat. This mechanic exists so that you cannot get "too much" damage reduction. However, the value of the resilience stat is damage reduction. As such, you have to the damage reduction. The value of that far exceeds any amount of diminishing returns in the stat's current formula.
The logic in that post is in regards to the "feel" of damage reduction.
Like the example lets say you take 100 damage with 0 resil, with 25% reduction you take 75 damage, meaning 25% reduced from original. On 50% reduction you take 50 damage.
But when comparing between the reduced damage when you have 50% to that of 25%, youve actually increased mitigation by a third.
This is logically right but needs much deeper analysis since we're comparing 2 very big intervals. Along with the rate of increase from above, ill provide this later tonight.
@grace: also yes increasing mitigation is always beneficial but what youre looking at with that example is a damage taken as a function of % mitigated. While I was looking at mitigation as a function of resilience. What you over saw in that example is the amount of resil required to go from 0-25% is significantly smaller than that of 25-50.
Such "sweet spot" here is when gaining x amount of resil isnt as beneficial as gaining damage from a pve gear, so on and so forth.
I will not be calculating the above since it will require a multiple regression analysis of codes performance against % mitigation and multitudes of other stats.
Honestly, I would say that we should go back to Toshika's statements with this. The point at which it becomes more viable to increase your output stats instead of your resilience is the "sweet spot" I mentioned above in my posts. When you get to a point where you feel you can reduce enough damage to keep you alive and proceed to do what you need to do (heal, dps, or whatever), that is where you should go for your output stats instead of resilience. This varies depending on what role you play, how much you get focused, how bad you lag, how good or bad you are at the game, and whatever feels right to you.
And this is the part of the thread that I don't understand. What deeper analysis can you provide that will have any worth outside of arguing over bits of math?
Sine we're splinting math over here already (you could have just referred to my add on)
For you that are not so good with math.Quote:
Physical reduction : (1-((1-(Armor/(Armor+26070)))*(1-(1-0.99^(Resilience/79.12785)))))*100
Spell Reduction : (1-((1-(SpellRes)/100)*(1-(1-0.99^(Resilience/79.12785)))))*100
Lets take an example of 30k armor and 4k vs 4.4k ress
4k resil = 72.03% Physical reduction
4.4k resil = 73.41% Physical reduction
So extra 400 resilience is worth ~1.4% damage reduction
Spell ress of 97 and 4k vs 4.4k resil (97 resistance = 14.17% spell reduction)
4k resil = 48.36% spell damage reduction
4.4k resil = 50.92% spell damage reduction
So extra 400 resilience is worth ~ 2.5% damage reduction
Seems like you've forgot that resilience is secondary level reduction, so besides DR on itself it gains exponential DR from the primary reduction.
The less armor and resistance you have the more resilience is worth, the more you have of both the less its worth.
Heres a simple graph illustration Resilience as secodnary effect, vs resilience as only effect.
(forgot to put the markers. Y axis is reduction %, X axis is resilience value. Red line is resilience reduction alone, blue is armor + resil reduction [total physical reduction])
Cloth/Leather armor
http://i.imgur.com/upNUZwb.jpg
Plate armor
http://i.imgur.com/VsvQ1ZN.jpg
You can see that resilience has lesser value the higher the primary reduction is and higher it gets.
Which addon is that? :O does it provide all these beautiful equations??
Well basically to answer the OP: How much is 400 resil worth? I will show the increase of % mitigated from an increase of 400 resil on these main points: 0, 500, 1000, 1500, 2000, 2500 3000, 3500, 4000, 4500, and 5000.
Mainly using the first derivative of our damage reduction equation from my original post:
- At 0 resil: the rate of increase is 1.2701 x 10^-4 which means an increase by 5.08% reduction for the first 400 resil.Quote:
Y' = - ln(0.99987) X (0.99987^x)
- at 500 resil: 400 resil gives you an increase of 4.77%.
- at 1000 resil: 4.47%
- at 1500: 4.19%
- at 2000: 3.94%
- at 2500: 3.7%
- at 3000: 3.47%
- at 3500: 3.26%
- at 4000: 3.06%
- at 4500: 2.87%
- at 5000: 2.69%
so as you can see coming from 0 resil to 5k resil, the amount of mitigation increase for every 400 resil went down by almost 50%.
the numbers are even more significant as you compare the increase of damage reduction as resil increases by 1.
Now thanks to Blue, we can further understand damage reduction as a function of 3 unknowns (armor, spell resistance and resil).
Like I said, to calculate the actual sweet spot you'd need to factor in:
Not to mention what stat is beneficial for your class, how each stats affect your output, so on and so forth.