61 réponses à ce sujet

[SkillSpray]

Technically, I asked for Armor Penetration VS Attack, not VS Critical Damage... I don't see how this could help me with that in mind

 

I do appreciate the effort though!

 

 

Woops. 

 

Still, you did the calculations for attack, i did a spreadsheet for crit dmg. Should still be useful info.

I understand you were debating whether to use armor penetration or attack on grip. I got a bit sidetracked and decided to do the math on what rings would be better, as I was curious about this myself.

 

 

 

Conclusion
With those results, I conclude 62 armor is where 18% Armor Penetration becomes better than 10.5% Attack.

From that conclusion, 2 questions arise:
1- Did I made a mistake while calculating that?
2- How many enemies have 62+ armor? (is there any spreadsheet with those values like there was in ME3?)

 

There's a thing you missed in your calculations.

The more attack you have (promotions, weapons, etc), the less valuable that 10.5% attack on the dagger becomes.

 

E.g.: (napkin math):

At 0% attack, 10% attack would increase your damage by 10% (not really 10%, it will be less due to the way damage is calculated in DA:I, but for the sake of it, assume 10%).

 

If you already have 50% attack from other sources (not an unreasonable value), the extra 10% attack would be a roughly 6% increase.

 

At 100% bonus attack, the extra 10% would be 5% (roughly), and so on.

 

You have to factor that into the calculations, as you cant compare them head to head. I think it's highly unlikely you already have armor pen on gear, thus the value of armor pen stays the same, while the value of attack becomes more diluted the more attack you already have. 

 

My guess is the armor value at which armor pen becomes better than attack will be less than 62 for you. Easiest way to check would updating the values in your formula with your current attack from promotions/gear and redo the calculations

[Texasmotiv]

This is true, but without any indication on how much Armor enemies have, it can become a very moot point.

 

It doesn't matter if Armor Penetration is 100% when Armor is at 10, but it would be incredibly powerful if Armor was at 1000.

 

Without those Armor numbers, it's actually impossible to determine the actual moment where the formula switch from "more beneficial going for Attack" to "more beneficial going for Armor Penetration".

 

My guess is that Armor Penetration could be very powerful, but unless you play only Nightmare/Heartbreaker, it ain't gonna matter.

 

It might be useful to the community for us to make a wiki thread with enemy armor values in different game modes. Next time I play nightmare, I will write some of the armor values down as I will have plenty of time in the fade to check these...

[TheThirdRace]

Updated the original post to include my "findings" and add 2 questions.

[SkillSpray]

Updated the original post to include my "findings" and add 2 questions.

 

Tried to answer question 1 in the above post.

I wont answer question 2 as I have no solid data about it. My guess is - not many.

[LearnedHand]

@ThirdRace, I don't see why you would ever choose an attack % upgrade unless you had zero or little willpower promotions.

[TheThirdRace]

@ThirdRace, I don't see why you would ever choose an attack % upgrade unless you had zero or little willpower promotions.

I don't see why I'd ever choose armor penetration upgrade unless it makes mathematical sense?

On Metals, all you have is Attack, Armor Penetration and Stagger on Hit. Not much "choices"...

[LearnedHand]

I don't see why I'd ever choose armor penetration upgrade unless it makes mathematical sense?

On Metals, all you have is Attack, Armor Penetration and Stagger on Hit. Not much "choices"...

 

Ah, I must have forgotten one hand weapons don't have leather offense options?

[capn233]

It might be useful to the community for us to make a wiki thread with enemy armor values in different game modes. Next time I play nightmare, I will write some of the armor values down as I will have plenty of time in the fade to check these...

 

Yes.  A small team of people could knock out generating a spreadsheet with basic values relatively quickly.

[Drasca]

Ah, I must have forgotten one hand weapons don't have leather offense options?

 

Single Blade Daggers T3 upgrades, not 1H weapons in general.

Dual Blade Daggers have lots of offense leather.

[TheThirdRace]

There's a thing you missed in your calculations.
The more attack you have (promotions, weapons, etc), the less valuable that 10.5% attack on the dagger becomes.
 
E.g.: (napkin math):
At 0% attack, 10% attack would increase your damage by 10% (not really 10%, it will be less due to the way damage is calculated in DA:I, but for the sake of it, assume 10%).
 
If you already have 50% attack from other sources (not an unreasonable value), the extra 10% attack would be a roughly 6% increase.
 
At 100% bonus attack, the extra 10% would be 5% (roughly), and so on.
 
You have to factor that into the calculations, as you cant compare them head to head. I think it's highly unlikely you already have armor pen on gear, thus the value of armor pen stays the same, while the value of attack becomes more diluted the more attack you already have. 
 
My guess is the armor value at which armor pen becomes better than attack will be less than 62 for you. Easiest way to check would updating the values in your formula with your current attack from promotions/gear and redo the calculations

I don't understand where you come from for Attack. I'll need a bit more math than just conclusions to understand how you got there.

My guess is you have a different perspective than I do. The difference between 50% and 60% can be 10% (60 - 50 = 10) if you see it as an absolute difference between the 2 numbers, but it can be 20% (60 / 50 = 0.2) or 16.666% (50 / 60 = 0.833) if you see it as a relative difference between the 2 numbers.

When I read the damage formula, it says "*(1 + attack_bonus)", I see it as an absolute difference between the 2 numbers, not a relative difference. In that regard, 10% more Attack will always be 10% more damage because at that specific point in the formula, it does add 10% to the current damage value. Overall, it might not give a 10% increase in damage because of other variables, but that's a relative comparison.

If I'm completely off the track, please advise as I just wanna know the truth.

Edit: Changed some terms to better reflect what I mean (absolute vs relative) - Thanks Drasca for pointing it out.

Modifié par TheThirdRace, 14 juillet 2015 - 07:09 .

[Drasca]

It might be useful to the community for us to make a wiki thread with enemy armor values in different game modes. Next time I play nightmare, I will write some of the armor values down as I will have plenty of time in the fade to check these...

 

You can take screenshots with print screen on PC. I hear one of the consoles has a screenshot function too.

 

If I'm completely off the track, please advise as I just wanna know the truth.

 

He's talking about relative increases of damage vs absolute ones.

[DragonRacer]

You can take screenshots with print screen on PC. I hear one of the consoles has a screenshot function too.

 

 

I believe XBox One has a feature for that, possibly. I don't know 100% since I don't own one.

 

PS4 definitely does. In the Share button, you can set up very easily to take screenshots (or save video) and either save them to the PS4 (and put on a flash drive later for easy transferring) or directly upload them to Facebook, Twitter, or - in the case of videos - YouTube.

[SkillSpray]

I don't understand where you come from for Attack. I'll need a bit more math than just conclusions to understand how you got there.

My guess is you have a different perspective than I do. The difference between 50% and 60% can be 10% (60 - 50 = 10) if you see it as a pure difference between the 2 numbers, but it can be 20% (60 / 50 = 0.2) or 16.666% (50 / 60 = 0.833) if you see it as a comparison between the 2 numbers.

When I read the damage formula, it says "*(1 + attack_bonus)", I see it as a pure difference between the 2 numbers, not a comparison. In that regard, 10% more Attack will always be 10% more damage because at that specific point in the formula, it does add 10% to the current damage value. Overall, it might not give a 10% increase in damage because of other variables, but that's a comparison.

If I'm completely off the track, please advise as I just wanna know the truth.

 

 

Okay. 

 

Basically when you put your numbers into the formula, you're simply calculating the difference between attack/arp on the dagger. The calculations that you did are correct, but they do not reflect the whole picture. What you did is an absolute comparison between the 2 options, as Drasca pointed out.

 

In order to get an actual idea what's happening, you have to also consider your other stats, which in this case would be the values of attack and arp that you already currently have on your gear, e.g. a relative comparison of these 2 to your current stats.

 

When the formula states (1 + attack_bonus), it implies you need to add all the % attack you already have from gear (rings, weapons, willpower on gear, etc) and promotions (willpower), not just the stats on your weapon. The other stats you currently have also affect the final damage you do.

Same applies for armor pen already on your gear.

 

Example: Let's assume you have 20% attack from gear, and another 30% attack from 60 willpower (promotions), for a total of 50% attack before taking the new grip into consideration.

 

The other bit of formula in this case is irrelevant, as we are comparing the attack gain vs arp gain.

 

final_damage = (base_damage * rand(0.95 to 1.05) - armor * (1 - armor_penetration))

    * (ability_multiplier)

    * (1 + critical_damage_bonus + flanking_bonus)

    * (1 + attack_bonus + damage_multiplier + type_bonus)

    * (1 - magic_resistance)

 

I made the relevant parts of this bold. Thus the formula is:

final_damage = ((base_damage- armor * (1 - armor_penetration))* (1 + attack_bonus)

 

With zero attack from gear/promotions (impossible, as you have 10 willpower by default), attack_bonus would be equal to whatever you want to put into your grip on the dagger.

 

With the stats i provided above, the attack_bonus becomes 0.105 (new grip) + 0.2 (gear)+0.3(promotions)=0.605

Therefore the final formula would look like this:

Assuming 62 armor.

For attack on grip:

final_damage = (base_damage- armor * (1 - armor_penetration)* (1 + attack_bonus)

final_damage = ((167                    - 62     * (1-  0                              ))*(1+ 0.605)

                       =168.52

 

For arp on grip:

final_damage = (base_damage- armor * (1 - armor_penetration)* (1 + attack_bonus)

final_damage = ((167                    - 62     * (1-  0.18                              ))*(1+ 0.5)

                       =174.24

 

Redo the calculations until you find where these 2 results are equal. That would be the armor value that causes arp to become better than attack.

But again, this is an unrealistic calculation. Attack loses value even more when you factor in the rest of the formula. Attack would be just another multiplier, while arp affects the base damage, which is more important in this case.

 

[yarpenthemad21]

I don't understand where you come from for Attack. I'll need a bit more math than just conclusions to understand how you got there.

My guess is you have a different perspective than I do. The difference between 50% and 60% can be 10% (60 - 50 = 10) if you see it as an absolute difference between the 2 numbers, but it can be 20% (60 / 50 = 0.2) or 16.666% (50 / 60 = 0.833) if you see it as a relative difference between the 2 numbers.

When I read the damage formula, it says "*(1 + attack_bonus)", I see it as an absolute difference between the 2 numbers, not a relative difference. In that regard, 10% more Attack will always be 10% more damage because at that specific point in the formula, it does add 10% to the current damage value. Overall, it might not give a 10% increase in damage because of other variables, but that's a relative comparison.

If I'm completely off the track, please advise as I just wanna know the truth.

Edit: Changed some terms to better reflect what I mean (absolute vs relative) - Thanks Drasca for pointing it out.

Look at formula closely. (1+ attack_bonus) means that 10% more attack (so adding +10%) gives in relative way 10% more damage only if current attack_bonus is 0, which isn't a points ever.
Most of the time weapons + passives alone give you around 40% attack stat. Next willpower with whatever value.
In that case +10% attack can give almost 0 damage increase in relative sense. For longshot for example it would give around 1%.

Armor penetration works on base damage. This is different. Relative increase or decrease in base damage always effect relative increase or decrease in overall damage.

You should recalculate your data. You assumed situation in which you have zero promotions and zero +attack stats and zero skills and passives. In that case which does not exists at all in game added 10.5% attack just gave 10.5% damage increase.
If would change bonus damage to more "realistic" value around 40-50% it wouldn't be a case, and probably even at 30-40 armor value penetration would win.

[kmeeg]

It could also be interesting to add a 40%(basic)+100%(dagger) critical damage multiplier to the equation, since the reference point were a 'Hakkon dagger".

[SkillSpray]

It could also be interesting to add a 40%(basic)+100%(dagger) critical damage multiplier to the equation, since the reference point were a 'Hakkon dagger".

I did that with critical damage in an earlier post.

 

I'll pass the honor of doing the same for attack to someone else.

[TheThirdRace]

You should recalculate your data. You assumed situation in which you have zero promotions and zero +attack stats and zero skills and passives. In that case which does not exists at all in game added 10.5% attack just gave 10.5% damage increase.
If would change bonus damage to more "realistic" value around 40-50% it wouldn't be a case, and probably even at 30-40 armor value penetration would win.

I did a quick calculation with 40% Attack as a base, meaning I add either 10.5% Attack to that or 18% Armor Penetration.

The result was that 18% Armor Penetration became better than 10.5% Attack at 50 Armor (maybe 49, I didn't go into details) instead of the original 62 I calculated. This means the more Attack you already have before adding Armor Penetration, the more Armor Penetration benefits from it and become stronger.

More Critical Damage, Flanking, Ability Multiplier, etc., also helps lower the Armor value where Armor Penetration becomes better than Attack, but not as much as I would have anticipated. I'll need to do more precise calculus with my real Attack % before drawing any definitive conclusion...

In any case, we still don't have a document that list Armor values, health, etc., for enemies. Any ideas what are the range of Armor values for Perilous, I'd like to spend my time elsewhere than in the Fade to get those numbers

[Drasca]

In any case, we still don't have a document that list Armor values, health, etc., for enemies. Any ideas what are the range of Armor values for Perilous, I'd like to spend my time elsewhere than in the Fade to get those numbers

 

0 - 75 ish for Perilous on most mobs.-2 to +10 armor tolerance. Either way won't affect calculations much.

 

75 being the upper end for higher armored mobs like Revenant, High Dragon, etc

 

Demons have the least armor, Pride Demons having the most (iirc), DC having Zero

Venatori next, with Brutes having the most amongst them.

Red Templars, Red Knights / RTC having the most, while Archers having minimal (and I think Horrors too)

 

We really do need people recording the armor values.

[DanakV]

We really do need people recording the armor values.

 

Oh hell yes.  Finally an excuse to justify to my friends playing my Hunter on Nightmare!

[yarpenthemad21]

I've done few test around Avvar
Setup was:
Avvar
Nightmare diffuculty
Hunting for red templar...
Weapon used - inquisition 2h sword (58 base)
Skill tested - korth might, lady wraith and crashing leap.

What I was testing - Do Avvar elemental damage abilities ignore armor.

Test results - Yes they do. For 99% when 1% is me being blind. I can see a difference on red templar horror without barrier using basic version of korth might - massive 5 damage
and cold version - around 100+
Aoe effect of LW was dealing around 50 damage to zealots, crashing leap around 160. I've ignored demon faction and focused on either zealots or horrors (for which I know has quite high armor).

So it officially in my opinion call Skywatcher a mage

[kmeeg]

Oh where's kmeeg? He's way better at articulating this stuff than me.

 

Oh 3'rd did ask me, but I could not give any decent answers (I had not done any calculations)  But 3'rd knew that there were some people on the forums that were quite skilled in math and whom might able to give much more accurate answers, hence this post.... and he was right! 

 

Offhand I just figured +attack for perilous (and below), and +armor penetration for nightmare.

 

 

New player advice: staffs benefit absolutely nothing from armor penetration.

[Archaeologist]

How much effective "uptime" does ArP have in NM? As in what's the ratio of monsters with 62+ armor to ones without?

[Archaeologist]

New player advice: staffs benefit absolutely nothing from armor penetration.

Even for those rare few AWs using Spirit Blade?

[TheThirdRace]

Even for those rare few AWs using Spirit Blade?

Magic damage isn't affected by armor, only physical damage is.

Spirit Blade is Spirit damage, thus magic damage, thus isn't affected by armor.

[TheThirdRace]

How much effective "uptime" does ArP have in NM? As in what's the ratio of monsters with 62+ armor to ones without?

Just updated the original post to give a follow up about that.

While I don't know the specific answer to the "what is the ratio of monster" question, having 40% Attack before adding either 18% AP or 10.5% ATT makes a huge difference as it lowered the armor "target" from 62+ to 40+.

We still need to confirm what is the ratio of monster that are 40+ armor, but I'm guessing it's more than 50% on Perilous and higher. For me it was a no brainer to go Armor Penetration after seeing a 40 armor value here as anything under 40 armor dies pretty fast anyway so we might as well kill those over 40 armor faster too...