(PvP) How do you determine whether weapon damage or penetration gives better numbers.

Nyladreas

Oftentimes I find myself at a loss while theory crafting when there are no "real" ways to understand (or test) which one of these stats will generate more damage in the long run and offer more overall versatility.

For example, if I'm wanting to create a fun build that's attempting to stack as much crit damage as possible just for pure sheets and giggles (aka IDC about its usefulness just its damage potential), how do I know whether having extra 1000 weapon damage will do more than having 5000 penetration? Not to mention that weapon damage also increases healing...

Is there any calculator or spreadsheet of some sort?

Is there perhaps a common point when penetration becomes less optimal (a soft cap)? Or vice versa?

Is there perhaps a certain critical chance rate at which weapon damage would be more useful than penetration?

Any other factors?

If anyone's knowledgeable about this then pretty pretty please share your power!

BohnT2

The answer is penetration because your procs don't scale with your stamina or weapon damage and building for max stats is useless with how procs overshadow them in all situations.

Nyladreas

BohnT2 wrote: »

The answer is penetration because your procs don't scale with your stamina or weapon damage and building for max stats is useless with how procs overshadow them in all situations.

That's exactly why I said that I don't care about its usefulness but rather just the numbers it would generate (let's say in a theoretical environment completely void of procs) @BohnT2

Olupajmibanan

Google reverse engineering results from Asayre (lost the link, sorry). He did a very deep analysis and revealed all the formulas used by the game.

Red_Feather

It is actually very simple and shocked me. If you only care about damage, and not stuff like healing, and you don't have a group buffing your penetration then going with penetration over weapon damage is always better.

Nyladreas

Red_Feather wrote: »

It is actually very simple and shocked me. If you only care about damage, and not stuff like healing, and you don't have a group buffing your penetration then going with penetration over weapon damage is always better.

Let's say healing would be a factor too, since i'm attempting to stack as much crit damage(and crit healing) as possible.

So.... I've said it kind of wrong. If for example 1000 weapon damage would give me better healing and almost as much damage as penetration (or rather I wouldn't notice that much of a difference) then I'd of course pick weapon damage over penetration any day...

@Red_Feather

Sergykid

https://en.uesp.net/wiki/Special:EsoBuildEditor

click on the three dot icon on the right side of the page, in your case at the "physical penetration" row. There you can see the formula, so you have to do the math yourself if you can't find someone who already did it.

PhysicalPenetration = Item.PhysicalPenetration + Set.PhysicalPenetration + Skill.PhysicalPenetration + CP.PhysicalPenetration + Buff.PhysicalPenetration + Mundus.PhysicalPenetration

Jayman1000

You could try to create two builds, one with desired wep dmg the other with desired penetration, and see which one offers the best dps.

HankTwo

What people need to realize is that there is no general answer to questions like this, since the outcome depends on what offensive stats you already have and what defensive stats your target has.

Since in your example you are only interested in weapon damage and penetration we can look at a simplified damage calculation that ignores other factors like damage done buffs, crits & crit resistance, maim & vulnerability debuffs, as well as target mitigation buffs. It would look like this:

true damage = [base_skill_tooltip] * [all the stuff we ignore for now] * [target_armor_mitigation]

The formula for the base skill tooltip usually looks roughly like this:
base_skill_tooltip = X1 * stamina + X2 * weapon_damage - 1
with X1 and X2 being skill specific coefficients. However, most of the time they share a ratio of X2 / X1 = 10.5, so we can simplify to (I will ignore the minus 1 as well):
base_skill_tooltip = (X/10.5) * stamina + X * weapon damage

For the target armor mitigation we have this formula:
target_armor_mitigation = 1 - (((target_armor + sum_of_target_armor_buffs - sum_of_target_armor_debuffs) * (1 - sum_of_armor_bypass_buffs) - sum_of_penetration) / 66000)
Since you are only interested in penetration we can again simplify this to:
target_armor_mitigation = 1 - ((target_true_armor - sum_of_penetration) / 66000)

Now, since we got the formulae we can do some calculations for your original question of 1000 weapon damage vs 5000 penetration. Lets say in both cases you would have a stam pool of 30000 and already 3000 penetration. For the skill coefficient we will use an arbitrary value of 1. To show you how the result can differ we will look at two different values for the base weapon damage (2000 vs 4000 before you add 1000) and two different values for the true target armor (15000 vs 35000 before penetration), resulting in 4 different scenarios:

1) Low base weapon damage and low target armor:
Adding 1000 weapon damage: true_damage = [(1 / 10.5) * 30000 + 1 * (2000 + 1000)] * [1 - ((15000 - 3000) / 66000)]
= [5857] * [0.8182] = 4792
Adding 5000 penetration: true_damage = [(1 / 10.5) * 30000 + 1 * (2000)] * [1 - ((15000 - (3000 + 5000)) / 66000)]
= [4857] * [0.8939] = 4342
--> extra weapon damage wins and gives 4792 / 4342 = 1.104 ~ 10.4 % more damage

2) Low base weapon damage and high target armor:
Adding 1000 weapon damage: true_damage = [(1 / 10.5) * 30000 + 1 * (2000 + 1000)] * [1 - ((35000 - 3000) / 66000)]
= [5857] * [0.5152] = 3018
Adding 5000 penetration: true_damage = [(1 / 10.5) * 30000 + 1 * (2000)] * [1 - ((35000 - (3000 + 5000)) / 66000)]
= [4857] * [0.5909] = 2870
--> extra weapon damage wins and gives 3018 / 2870 = 1.052 ~ 5.2 % more damage

3) High base weapon damage and low target armor:
Adding 1000 weapon damage: true_damage = [(1 / 10.5) * 30000 + 1 * (4000 + 1000)] * [1 - ((15000 - 3000) / 66000)]
= [7857] * [0.8182] = 6429
Adding 5000 penetration: true_damage = [(1 / 10.5) * 30000 + 1 * (4000)] * [1 - ((15000 - (3000 + 5000)) / 66000)]
= [6857] * [0.8939] = 6129
--> extra weapon damage wins and gives 6429 / 6129 = 1.049 ~ 4.9 % more damage

4) High base weapon damage and high target armor:
Adding 1000 weapon damage: true_damage = [(1 / 10.5) * 30000 + 1 * (4000 + 1000)] * [1 - ((35000 - 3000) / 66000)]
= [7857] * [0.5152] = 4048
Adding 5000 penetration: true_damage = [(1 / 10.5) * 30000 + 1 * (4000)] * [1 - ((35000 - (3000 + 5000)) / 66000)]
= [6857] * [0.5909] = 4052
--> extra penetration wins and gives 4052 / 4048 = 1.001 ~ 0.1 % more damage

As you can see, the only case in our example where the extra penetration was better for damage was the case where you would already have 4k weapon damage and would target a tank with 35k resistances (after buffs and debuffs). This is a case that favors penetration much more than adding more weapon damage, but since even then the difference in damage would be negligible, I would take the 1k extra weapon damage basically every single time, even more so since it would also buff my healing.

Generally speaking, if you already have a lot of weapon damage + stamina, it is better to start investing into other ways of amplifying your damage, like damage done buffs or penetration, and vice versa. However, in your specific example the power budget of 1k weapon damage is just so much more than 5k penetration. Imo, a much fairer comparison would have been something like 600 weapon damage vs 5k penetration. Also, keep in mind that extra pen would boost your glyph damage and potential proc sets.

I hope this in-depth answer will help you in your future theorycrafting

~ Hank

relentless_turnip

A rough way of working this out is to look at your penetration as a percentage. I say rough because this isn't exact, but will give you a vague idea.

Each 660 armor is equal to 1% resistance. So the same can be said for penetration. So major breach for example gives you 5948 penetration. 5948/660= 9.01. So if your tootip reads 10,000 you can roughly workout the additional damage with 10,000*1.0901= 10,901.

It is a good idea to use uesp to see exactly how things a worked out. This is as I said just a vague way of working things out.

Generally pen is much better for damage and WD is better overall. Though it depends on the build and what you are doing defensively to stay alive. A nightblade for instance could stack loads of penetration with very little drawback. If a lot of your heals are based on damage done then penetration could still be the right move.

Kartalin

Jayman1000 wrote: »

You could try to create two builds, one with desired wep dmg the other with desired penetration, and see which one offers the best dps.

Indeed, you could test both on a target dummy. I don't know what the base armor on the dummies are though so if you overpenetrate it might be off.