Some time after Armor Up: where do we stand?

[Axeyu]

Consider two builds, one with 1000 hit points and no damage mitigation and the other with 500 hit points and 50% damage mitigation. Using the formula EHP=HP/M we get identical EHP of 1000.

The argument is that these two builds are then equivalent.

Now place them into an encounter where the monster does 750 damage. Make maximum healing 250 for each character and assume that the rate of damage is the same for both so that two heals take place before each hit.

Here is what happens:

A with 1000 hp

1000-750=250+500=750-750=0, character is unconcious.

B with 500 hp

500-375=125+500=500-375=125+500=500, etc. (Note: HP cannot exceed 500).

These two builds with exactly the same EHP are not equivalent builds because one dies on the second hit while the other continues fighting indefinitely.

We only see this if we take EHP and derive HTK from it. Stopping short does not reveal the whole story.

And, HTK alone does not reveal the whole story either. It only allows us to get to the external information concerning healing.

This is why EHP cannot be used to compare builds for equivalency. It is two steps removed from the actual information needed. HTK may be imperfect but it is closer to the actual information we must have to make an equivalency determination.

No, HTK is not in any way closer, because it requires you to rectally derive a damage number that is not representative of anything.
The goal is to evaluate the builds without damage number, because the reality is that the incoming damage varies greatly so any evaluations based on a single damage number is flawed.
I have been saying this since page 9, please let it sink in now. You were wrong about EHP, consider that you might be wrong about this aswell.


If we start evaluating healing we leave the realm of the OP, which is why I have not done it.
But for now, lets try to evaluate the survivability of those two builds with healing, and then see if it matches the situation.

Build 1: 1000 HP, 1000 EHP. HP/EHP= 1000/1000 =1
Build 2: 500 HP, 1000 EHP. HP/EHP = 500/1000 = 0,5

That means that build 2 requires half as much healing as build 1 to stay alive. Or conversely that build 1 requires twice as much healing as build 2.

Lets see under the conditons you set up what happens if this is true:
Now place them into an encounter where the monster does 750 damage. Make maximum healing 500 for build 1 and 250 for build 2 and assume that the rate of damage is the same for both so that two heals take place before each hit.

Here is what happens:

A with 1000 hp

1000-750=250+1000=1000-750=250+1000=1000, etc. (Note: HP cannot exceed 1000).

B with 500 hp

500-375=125+500=500-375=125+500=500, etc. (Note: HP cannot exceed 500).

These two builds with exactly the same EHP continues fighting indefinitely.

Now that we know that it works we can use HP/EHP (or EHP/HP) to evaluate the relative need for healing (or the efficiency of healing) between builds.

[Baktiotha]

Now place them into an encounter where the monster does 750 damage. Make maximum healing 500 for build 1 and 250 for build 2 and assume that the rate of damage is the same for both so that two heals take place before each hit.

Here is what happens:

A with 1000 hp

1000-750=250+1000=1000-750=250+1000=1000, etc. (Note: HP cannot exceed 1000).

B with 500 hp

500-375=125+500=500-375=125+500=500, etc. (Note: HP cannot exceed 500).

These two builds with exactly the same EHP continues fighting indefinitely.

Why do you allow twice the amount of healing for build 1? Is there any reason to do this other than it makes the math work?

We are comparing like things. Like things have the same amount of whatever. In this case the two builds have the same EHP, take the same damage, and receive the same healing.

But you set up the problem to compare unlike things. They have the same EHP, take the same damage, but one gets double the healing of the other.

That is inherently not an equal comparison.

But, homogenization and build equivalency *requires* that things be the same or be trending to the same. You attempt to show sameness by selecting elements that are most definitely NOT the same.

Edit:

Put differently, you recognize that the two builds are not the same as soon as healing is introduced. Although the EHP values say that the two are identical for defense purposes it is immediately clear that the two are NOT functionally equivalent at all.

Declaring the two as functionally equivalent would mislead readers. The caveat that they are functionally equivalent in terms of armor only will be lost.

Thus we cannot begin to say that there is homogenization because the two builds are not in fact trending towards or becoming the same. Yes, they have common parts and yes, some of those parts are similar. But the builds are different -- thus showing the claims of homogenization to be a lie.

This runs all the way back to the early pages when I noted that what we have instead is balance. What you have done is introduce balance by doubling the healing of the first build. Now the builds perform the same NOT because they are homogenized but because they are balanced.

[Baktiotha]

wow. So that's what I get for giving you the benefit of the doubt and assuming that the following are just typos:

So "math genius", how about you show how HTK*HP=EHP?

Actually, the first is a typo and the second is just repeating the first error. The correct formula should be EHP=D*HTK.

But, you knew that.

And, I'm not entirely certain what your complaint is. You have fought against using HTK at all and yet there is a direct relationship between HTK and EHP.

That has not stopped you from rejecting HTK=HP/(D*M) and trying to introduce a different formula in its place.

The disagreement between us comes down to your rejection of using HTK to get to the fine level of detail that will produce meaningful information for players. I took a while to come to how EHP is used and we still don't have clarity on how much variation is permitted in order to assert functional equivalence. Still, I'm willing to see how it gets us to describing viability. The thread seems to assert that the path lies through HtD but nobody seems able to describe how that works.

You, on the other hand, seem unable to permit any alternative exploration. You seem to be unable to accept that in the EHP pathway if D is ever introduced you have effectively allowed HTK. If that happens then EHP is an extraneous step that can be eliminated and we can move directly to HTK.

At the moment EHP only serves to define the upper and lower limits of arbitrary damage amounts that describe functionally equivalent builds. The allowance for a band based on a percentage difference describes arbitrary damages because the setting of the percent value is arbitrary. So, logically, you are producing arbitrary D values just as I am.

And, if the bands are arbitrary then there's no need to base them off of EHP -- we can just select D in increments.

What it looks like to me is that, unless your methodology uses HtD and avoids D entirely, we are headed towards the same conclusions. It really does not matter that I cannot see how to get there using HtD and avoiding D. What matters is you describing it so that the other readers can see how to get there.

[Baktiotha]

Alright, real life interferes again. Last time I needed to be gone the thread went into hibernation. I will be away most of Saturday and all day Sunday thru Tuesday so will not get a chance to post. Let's see if while I am gone those pushing EHP can explain the relationship of that, HtD and viability without injecting any discussion of damage.

[sirgog]

That's not really an accurate description of the problem. Firstly because there was no "simply" about it except in terms of understanding what you needed to achieve. Actually achieving those number was near impossible once you hit mid teens without major DPS and other sacrifices. The people who could had worked for it, and it was an asset to the game because it was a goal that was hard to reach.

The issue was not the 5% where you could not be hit, it was that after L10 you either sacrificed everything and reached that magic 5% or... well, for the vast majority of the time you may as well not have been wearing any armour whatsoever. You wouldn't even get close to the numbers required to make mobs even miss on a roll of a 2.

So you either built for max armour and no DPS, or you built for evasion and dodge and more or less kept your DPS except for extreme cases.

In terms of dealing with those problems, the first AC revamp certainly nailed it. As I recall it overnight more or less destroyed tank builds and led to a ton of people exiting the game because that incredible sacrifice was now (in their eyes) totally worthless. i think they were overreacting personally - they'd gone from 95% miss to 80% miss for the most part which was still out of reach for a lot of folk even with the changes. Nevertheless, that's how they felt.

All the other armour tweaks since then are basically I think supposed to have made tanking viable again, and I think that's probably worked, for those who understand all the tweaks. I certainly see a lot of tanky builds these days at least in terms of taking the punishment in EE.

Confirming this - I was one of the few people that bothered building an AC-oriented character at the 20 cap.

Yes, they could eventually facetank epic quest bosses (not quite epic raid bosses) when you stacked curses on the boss and landed other debuffs. But they got destroyed by trash that hadn't been debuffed.

Levelling, from 1 to 10 you always focused on AC or DR and were unhittable or unhurtable. 11-17 you could be unhittable with effort but it was faster to be all-offence. 18 to epic content required extreme focus to be unhittable.

[Axeyu]

Why do you allow twice the amount of healing for build 1? Is there any reason to do this other than it makes the math work?

I am demonstrating that build 1 needs twice as much healing as build 2 to be kept alive, which is far more useful information than the outcome of the incredibly specific case of the 750 damage and 2x250 healing.

We are comparing like things. Like things have the same amount of whatever. In this case the two builds have the same EHP, take the same damage, and receive the same healing.

But you set up the problem to compare unlike things. They have the same EHP, take the same damage, but one gets double the healing of the other.

That is inherently not an equal comparison.

But, homogenization and build equivalency *requires* that things be the same or be trending to the same. You attempt to show sameness by selecting elements that are most definitely NOT the same.

Edit:

Put differently, you recognize that the two builds are not the same as soon as healing is introduced. Although the EHP values say that the two are identical for defense purposes it is immediately clear that the two are NOT functionally equivalent at all.

Declaring the two as functionally equivalent would mislead readers. The caveat that they are functionally equivalent in terms of armor only will be lost.

I also recognize that they are not the same when shoe lace colour and favorite tea is introduced, but that does not change how much damage then can take, which is what the OP is about.
But you also have to understand that even if we are to consider this and expand what we are talking functional equivalency for, EHP is better than HTK in every single way, as your analysis fail when you bring in an assumed damage number.
You just need to use EHP in the right way, and if you had paid attention you would have seen that you can use it to compare how much healing builds need relative to eachother.

[Axeyu]

Actually, the first is a typo and the second is just repeating the first error.

lol. I guess you score the points where you can.

The correct formula should be EHP=D*HTK.

But, you knew that.

And, I'm not entirely certain what your complaint is. You have fought against using HTK at all and yet there is a direct relationship between HTK and EHP.

That has not stopped you from rejecting HTK=HP/(D*M) and trying to introduce a different formula in its place.

How can you not know what my complaint is? The relationship depends on a random number that does not represent anthing. The HTK number is meaningless unless used as EHP and in that case there was no need to add in D in the first place.

Let me introduce you to a new number BAD = HP/(S*M), where S is the number of sandwhiches you have eaten this month. There is a direct relationship between BAD and EHP; BAD*S=EHP. It must be important!

I took a while to come to how EHP is used and we still don't have clarity on how much variation is permitted in order to assert functional equivalence.

How long is a rope? When does green stop being green? When is the rent to **** high?
My point is that the problem lies in the question.
There are no rules for how much variation is permited in order to assert functional equivalence. There is no universal truth to reach here.

Still, I'm willing to see how it gets us to describing viability. The thread seems to assert that the path lies through HtD but nobody seems able to describe how that works.

Viability is not about comparing builds to eachother. Don't move the goalposts.

You, on the other hand, seem unable to permit any alternative exploration. You seem to be unable to accept that in the EHP pathway if D is ever introduced you have effectively allowed HTK. If that happens then EHP is an extraneous step that can be eliminated and we can move directly to HTK.

We have explored HTK for almost 30 pages now and the conclusion is that EHP is clearly superior. We saw in your grand post that was supposed to show the superiority of HTK and uselessness of EHP that even with a known D you get more information, and faster, from just using EHP.

At the moment EHP only serves to define the upper and lower limits of arbitrary damage amounts that describe functionally equivalent builds. The allowance for a band based on a percentage difference describes arbitrary damages because the setting of the percent value is arbitrary. So, logically, you are producing arbitrary D values just as I am.

And, if the bands are arbitrary then there's no need to base them off of EHP -- we can just select D in increments.

No.
"They can be regarded as functionally equivalent if their EHP is within 5% of eachother" is not the same thing as "they can be regarded as functionally equivalent if they take 1000 damage". If you select D in increments you will at best approximate the information you get from EHP.

You clearly have no idea of what you are talking about still. You should do more examples and continue to learn instead of assuming, again, that you understand things you don't know anything about.

What it looks like to me is that, unless your methodology uses HtD and avoids D entirely, we are headed towards the same conclusions. It really does not matter that I cannot see how to get there using HtD and avoiding D. What matters is you describing it so that the other readers can see how to get there.

If you know how to use EHP you also know how to use HtD. D should be avoided when comparing builds for reasons I have repeated since page 9.

[nokowi]

It is incorrect to subtract 4 in the place of DR. That leads to error. My method and my math is correct.

I should post QED, but why gloat.

EHP=1700/(100/(100+209)=1700/(100/308)=1700/.3246=5236.


Here is where you added 209 and 100 and got 308, and used it in your calculations.

Yes, keep gloating about your wonderful math skills.

[nokowi]

Two comments. I think the prr of that rogue is too low, and you picked the tankiest build to compare.

When using twf or thf I hope I have convinced you that you can get to very similar defensiveness with a rogue and a paladin (or a barbarian, or a ranger...).

I understand it is a bit tricky to pick someone else a numbers, but you will agree with me that you picked the tankless build I posted with somehow maxxed prr (all pls, prr feats that are kind of unnecessary for a shielder) and then compared to a non maxed out rogue. When you bring the rogue in line with the other builds the results in expected damage become very similar.

Now for everyone, what would be a better statistic that hits to die. Expected damage you can soak before dying? Combining EPH with avoidance based damage mitigation. Other ideas?

Please read your own comments above and below. You clearly showed us your concern that dodge was too high, or at a minimum that you are concerned about the pecking order of heavy vs other armor types.

To respond to your original posts, the math calculations with AC miss chance clearly show that the high dodge toons are much less defensive than heavy armor toons. Please read your own comments, and how they referred to ANY class (below).

Now that the data shows your original comments were incorrect, moving the goal posts to "TWF builds with similar PRR and low dodge have the same defense" should be an entirely different thread.

Your original statements have been proven wrong, that all armor types are not equivalent.

Someone in this very thread said they had 324 PRR. Your sample builds are not maximum defensive builds, so neither should the rogue build for comparison. If you want to argue the rogue only takes something like 60-80% more effective damage (instead of 100% more effective damage) with a maximum PRR rogue build choices that would actually gimp the toon, then I would say your original statements are still wrong and this distinction is meaningless to the thread.

But importantly we have also seen unprecedented power creep in dodge via itemization, with a single item providing 17% dodge as we speak (legendary animated rope).

So the question that begs is: where do we stand? Has armor been nerfed to pre buff situations in the pecking order or does it hold its ground well?

My conclusion so far is that heavy armor per se doesn't pull ahead any class defensively.

Rogue Including 28.75% AC miss Chance (let's bump this rogue up to 150 PRR):

(1-0.2875)*.5*.68*.75=0.1817 chance to be hit
0.1817*(100/(100+150)) = 7.27% damage taken
Damage taken: 1000*.727=72.7
Hits to kill: 1000/72.7 = 13.8

__________________________________________________ _________________

S&B Paladin Including 59.3% AC miss Chance:

(1-0.593)*.5*.94*.9=0.172 chance to be hit
0.172*(100/(100+289)) = 4.42% damage taken
Damage taken: 1000*.0442=44.2.
Hits to kill: 1000/44.2 = 22.6




After including AC and bumping PRR up to 150 (a value I have never seen another rogue post in-game or on the forums), we can see that those players who said all builds were equally survivable were way off base. Note that pallys actually get 30 more AC from shields at level 30 (not 15 as used in the miss chance calculations), so this calculation is under-representing the difference at end game.

If we include differences in HP (pally will win), the difference gets bigger.
If we include differences in self healing (pally wins big time), the difference gets much bigger.

We can conclude that low HP, low PRR toons get the shaft, just like they did when armor up came out.

We can conclude that devs can continue to increase dodge chance items without fear of changing this fact.

[levy1964]

12/6/2 only for me own chellenge :P i have hope this life give me some ability to kill LE sorjeck

Sorjeck : You're that Sanjo.

Sanjo : I've been waiting for you, Sorjeck.

Sorjeck : Who are you?

Sanjo : Neither enemy nor friend. I am back from a world where such
words are meaningless. I've removed all obstacles. Now you and I
will battle to the death.

Sorjeck : What do you want?

Sanjo : I've waited a long time for this day. Now I wanto to enjoy the
moment.

Eryhn : What... what's with these guys? It's like one of my Japanese
animes...

Sanjo : I've come from another world to do battle with you.

Sorjeck : What is it? Revenge?

Sanjo : It is nothing so trivial as revenge. A fight to the death with
you. Only in that can my soul find respite. I will kill you or
you will kill me... it makes no difference.

lovly MGS

[BigErkyKid]

Please read your own comments above and below. You clearly showed us your concern that dodge was too high, or at a minimum that you are concerned about the pecking order of heavy vs other armor types.

To respond to your original posts, the math calculations with AC miss chance clearly show that the high dodge toons are much less defensive than heavy armor toons. Please read your own comments, and how they referred to ANY class (below).

Now that the data shows your original comments were incorrect, moving the goal posts to "TWF builds with similar PRR and low dodge have the same defense" should be an entirely different thread.

Your original statements have been proven wrong, that all armor types are not equivalent.

Someone in this very thread said they had 324 PRR. Your sample builds are not maximum defensive builds, so neither should the rogue build for comparison. If you want to argue the rogue only takes something like 60-80% more effective damage (instead of 100% more effective damage) with a maximum PRR rogue build choices that would actually gimp the toon, then I would say your original statements are still wrong and this distinction is meaningless to the thread.







Rogue Including 28.75% AC miss Chance (let's bump this rogue up to 150 PRR):

(1-0.2875)*.5*.68*.75=0.1817 chance to be hit
0.1817*(100/(100+150)) = 7.27% damage taken
Damage taken: 1000*.727=72.7
Hits to kill: 1000/72.7 = 13.8

__________________________________________________ _________________

S&B Paladin Including 59.3% AC miss Chance:

(1-0.593)*.5*.94*.9=0.172 chance to be hit
0.172*(100/(100+289)) = 4.42% damage taken
Damage taken: 1000*.0442=44.2.
Hits to kill: 1000/44.2 = 22.6




After including AC and bumping PRR up to 150 (a value I have never seen another rogue post in-game or on the forums), we can see that those players who said all builds were equally survivable were way off base. Note that pallys actually get 30 more AC from shields at level 30 (not 15 as used in the miss chance calculations), so this calculation is under-representing the difference at end game.

If we include differences in HP (pally will win), the difference gets bigger.
If we include differences in self healing (pally wins big time), the difference gets much bigger.

We can conclude that low HP, low PRR toons get the shaft, just like they did when armor up came out.

We can conclude that devs can continue to increase dodge chance items without fear of changing this fact.

I already replied to this kind of comment.

You pick the shielded paladin with full PRR enhancements and feats and compare it to a rogue that has completely different PRR choices (does not blitz nor take PRR feats) and claim this proves me wrong.

The whole premise was to pick builds with similar choices and different armor types and classes. If you break that premise everything flies. I already have given you a rogue with around 200 PRR in a configuration that is not gimped and is comparable to that used by the other builds, a barbarian, a thf paladin, a ranger.

When compared to similar builds the rogue is somehow equivalent. You have the numbers spelled out to the last enhancement and gear choice. Can I fabricate scenarios where this does not hold? Yes of course.

So you picking up a completely different build with different build choices, arbitrary PRR values, a different destiny, and then comparing it to my standardized paladin shield build is a nonsense scenario. I can instantly bring down the differences by moving the paladin to divine crusader, taking out prr feats (already high prr, it does not need so much) and then get a completely different result. Then a guy will appear who plays his max prr rogue in clothes (sneaky Boulder build) and who knows what else. And all that will be pointless.


The point was whether they have washed out the inherent differences between heavy armor and other styles. And the answer is yes, because via reasonable itemization and gear choices you can obtain roughly equivalent defensiveness across a wide variety of classes that use different armor types. If you instead choose a different choice with implications for defensiveness you can create differences. But it is you creating them, not the fact that the armor type is different.

[nokowi]

I already replied to this kind of comment.

You pick the shielded paladin with full PRR enhancements and feats and compare it to a rogue that has completely different PRR choices (does not blitz nor take PRR feats) and claim this proves me wrong.

The whole premise was to pick builds with similar choices and different armor types and classes. If you break that premise everything flies. I already have given you a rogue with around 200 PRR in a configuration that is not gimped and is comparable to that used by the other builds, a barbarian, a thf paladin, a ranger.

When compared to similar builds the rogue is somehow equivalent. You have the numbers spelled out to the last enhancement and gear choice. Can I fabricate scenarios where this does not hold? Yes of course.

So you picking up a completely different build with different build choices, arbitrary PRR values, a different destiny, and then comparing it to my standardized paladin shield build is a nonsense scenario. I can instantly bring down the differences by moving the paladin to divine crusader, taking out prr feats (already high prr, it does not need so much) and then get a completely different result. Then a guy will appear who plays his max prr rogue in clothes (sneaky Boulder build) and who knows what else. And all that will be pointless.


The point was whether they have washed out the inherent differences between heavy armor and other styles. And the answer is yes, because via reasonable itemization and gear choices you can obtain roughly equivalent defensiveness across a wide variety of classes that use different armor types. If you instead choose a different choice with implications for defensiveness you can create differences. But it is you creating them, not the fact that the armor type is different.

No, I bumped the PRR up to 150 for the rogue, which is very near the values you requested. I would say after 40+ rogue lives I probably have a better understanding of what rogue PRR should be than you. Bumping it to 160 or 170 wont change the final result.

If you want to ignore all reasonable class and feat choices, just post the armor values.

You started a thread by picking class choices, you can't now argue that I cant do the same.

Show me your own rogue build that is equivalent before you argue they are equivalent. Stop avoiding the issue and actually defend your comments.

Until you do this, you cant argue they are the same. Do the math and prove your point.

I suspect you are unable to actually do this (make your point, not do the math - i know the math would be easy for you to do), because you are simply wrong in your OP. It should be simple for you to prove me wrong by getting off your rear and actually proving your own calculations with regard to rogue.

I'm willing to be wrong. Prove it.

Is your argument now that the worst possible paladin build is equivalent to the best possible rogue build, and that the entire point of your thread was pointless from the getgo?

[nokowi]

The whole premise was to pick builds with similar choices and different armor types and classes.

Not all builds can choose the same choices. A rogue can't get +1 HP per AP spent, +10% to armor/shield PRR, etc.

I will repeat, why did you spell out any builds if the builds must be exactly the same for your argument to work?

I already showed the difference between barb and pally was significant, even without self healing.

You have really wasted our time...

[nokowi]

And the answer is yes, because via reasonable itemization and gear choices you can obtain roughly equivalent defensiveness across a wide variety of classes that use different armor types.

You did not prove this. You backed off this and said that similar low dodge TWF builds had roughly the same defense.

Your final conclusion is that similar builds with the same armor have similar defense. (DUH!)


A wide variety of classes and armor types should include low PRR dodge toons in light armor. (You know, the ones you complain about being equivalent...)

[nokowi]

Dodgy characters were supposed to be master of dodge, less capable of sustainable damage due to spike (which was the real of heavy armor toons). And all that class differentiation stuff.

I think that heavy armor, light armor, medium armor, it all looks pretty much the same right now given the right class.

When are you going to prove light armor dodge toons are equivalent using the same type of calculations you used for all other builds?

Why are you afraid to do this?

[DrRomz]

S&B Paladin build: 289 PRR (25% taken)/ 6 Dodge To be hit: 40%Edamage: 7.7%

Umm, when I went to school a quarter of 40 was 10 not 7.7!

[BigErkyKid]

No, I bumped the PRR up to 150 for the rogue, which is very near the values you requested. I would say after 40+ rogue lives I probably have a better understanding of what rogue PRR should be than you. Bumping it to 160 or 170 wont change the final result.

If you want to ignore all reasonable class and feat choices, just post the armor values.

You started a thread by picking class choices, you can't now argue that I cant do the same.

Show me your own rogue build that is equivalent before you argue they are equivalent. Stop avoiding the issue and actually defend your comments.

Until you do this, you cant argue they are the same. Do the math and prove your point.

I suspect you are unable to actually do this (make your point, not do the math - i know the math would be easy for you to do), because you are simply wrong in your OP. It should be simple for you to prove me wrong by getting off your rear and actually proving your own calculations with regard to rogue.

I'm willing to be wrong. Prove it.

Is your argument now that the worst possible paladin build is equivalent to the best possible rogue build, and that the entire point of your thread was pointless from the getgo?

I already showed you the rogue equivalent build. Must have been buried in the piles of posts.

Let's bring back first the example. MOB hits 1000 raw damage, how many hits to kill the rogue?

Chance to be hit:
.95*.5*.68*.75=0.24225

Mitigation:
100 PRR (50%)

Expected damage taken (Edamage):
0.24225*.5=12%

Hits to kill the rogue
Damage taken: 1000*.12=120.
Hits to kill: 1000/120= 8.3


Now let's have some quick numbers on PRR achievable by rogue, using the same configuration as for the other builds.

TWF rogue 165 PRR (38% taken) / (34% blitz)
30 Larmor
6 Enhancements
38 Celestial armor
25 Insight+quality item
10 Epic DR
20 Scion of earth
36 PLs
(30) Blitz
30+6+38+25+10+20+36=165 (195)
Damage taken: (100/(100+165))=0.3773585

So in LD you get 34% physical mitigation and in SD 38%, given this set up.

Going back to the example above:

Expected damage taken (Edamage) Shadowdancer:
0.24225*.38=.9%
Expected damage taken (Edamage) LD:
0.24225*.34=.8%

Hits to kill the rogue (shadow dancer)
Damage taken: 1000*.09=90.
Hits to kill: 1000/90= 11.1

Hits to kill the rogue (blitzing)
Damage taken: 1000*.08=00.
Hits to kill: 1000/80= 12.5

So where does the rogue stand?

It takes 13,6 hits for a paladin to die, 14,6 for a ranger, and 14,1 for a barbarian, 11.1 for the rogue in Shadowdancer, 12.5 for the rogue in Legendary D.

The difference between the THF paladin and the rogue in LD is less than one hit to die.

[BigErkyKid]

Not all builds can choose the same choices. A rogue can't get +1 HP per AP spent, +10% to armor/shield PRR, etc.

I will repeat, why did you spell out any builds if the builds must be exactly the same for your argument to work?

I already showed the difference between barb and pally was significant, even without self healing.

You have really wasted our time...

Hold it man! I have been open and transparent about build choices. Non are impossible, or even whacky. I refer you to my previous post regarding build choices. The experiment is to pick similar okish build choices and different classes and check the result. As I already said, picking completely different choices breaks the equivalence. But if you chose to, you can build fairly equivalent builds across a wide range of archetypes. So it really isn't the armor, which provides a very minimal part of the end prr.

[nokowi]

I already showed you the rogue equivalent build. Must have been buried in the piles of posts.

This is a build without AC miss chance, something you promised you would include much earlier in the thread.

I just showed results with 150 PRR (a build close to yours) with all miss chances included.

Please take the time to make a rogue build that includes AC miss chance so that you can participate in the discussion in a meaningful way.

You can simply copy my calculations with your own build choices. It shouldn't take long. Then you can explain why you rejected my results.

[nokowi]

Hold it man! I have been open and transparent about build choices. Non are impossible, or even whacky. I refer you to my previous post regarding build choices. The experiment is to pick similar okish build choices and different classes and check the result. As I already said, picking completely different choices breaks the equivalence. But if you chose to, you can build fairly equivalent builds across a wide range of archetypes. So it really isn't the armor, which provides a very minimal part of the end prr.

I'm fine with you using your choices, just include AC miss chance and be honest about the results.

I just posted something very close to your build that you just rejected, so I have my doubts about you.