12 réponses à ce sujet

[Azrus]

I spent a few hours this afternoon running some tests to answer a few questions I had.  Figured I'd share the results with you all.
 
Dual Wield:

 
Your offhand weapon does not appear to influence your "damage" bar in the inventory screen.  Does it influence actual DPS?
 
Basic Attack:
Test details:

Spoiler

Character used:  Lvl 12 Alchemist.
Weapons used in testing:
High/Low test:
-259 DPS double-blade dagger.  94 AoE damage + 10 per strike from frost rune.
-154 DPS double-blade dagger.  52 AoE damage + 10 per strike from frost rune.
 
High/High test:
-259 DPS double-blade dagger.  94 AoE damage + 10 per strike from frost rune.
-259 DPS double-blade dagger.  94 AoE damage + 10 per strike from fire rune.
 
Testing conditions: 2-3 zones worth of Demonic enemies on routine.  Strikes were non-flanking strikes on Wraith enemies.  Under weakened status effect for most strikes, may cause a little inconsistency with overall damage numbers, but should not affect weapon comparison.
 
High/Low observations: Damage ranged from 95-110 on first strike of combo, and 50-60 on second strike of combo.  Alternated based on weapon striking enemy.
 
High/High observations: Damage ranged from 95-110 on first strike of combo and 95-110 on second strike of combo.

Results: Yes, the damage you deal per strike is dependent on the weapon that strikes your opponent.  Damage does not appear to be averaged across equipped weapons.
 
Twin Fang: (Multi-strike ability)
Test details:

Spoiler

Character used: Lvl 9 Assassin.  Twin Fangs w/ Ripping Fangs perk, Ambush, and First Blood passives.
Weapons used:
High/Low test:
-259 DPS double-blade dagger.  94 AoE damage + 10 per strike from frost rune. +5 Cunning +29% Flank damage
-154 DPS double-blade dagger.  52 AoE damage + 10 per strike from frost rune. +2% ArPen +2% Crit +15% Flank damage
 
High/High test:
-259 DPS double-blade dagger.  94 AoE damage + 10 per strike from frost rune. +5 Cunning +29% Flank damage
-259 DPS double-blade dagger.  94 AoE damage + 10 per strike from fire rune. +5 Cunning +26% Flank damage
 
Testing Conditions: 4-5 zones per test, damage figures were observed on stealth Twin Fang strikes while flanking.  Enemies targeted were Venatori archers.
 
High/Low observations: 900-1000 damage on first strike, 500-600 on second strike.
 
High/High observations: 1000-1100 damage on first strike, 1000-1100 on second strike.

Results: Same as basic attacks.  Damage for each Twin Fang strike appears to be calculated by the dagger in that hand.  Max damage difference in High/Low and High/High likely due to 11% higher flank damage bonus on off-hand dagger.
 
Shadow Strike (Single-strike ability)

Test Details:

Spoiler

Character used: Lvl 11 Assassin.  Shadow Strike w/o upgrade, Ambush, and First Blood passives.

Weapons used:

High/Low test:

-259 DPS double-blade dagger.  94 AoE damage + 10 per strike from frost rune. +5 Cunning +29% Flank damage

-154 DPS double-blade dagger.  52 AoE damage + 10 per strike from frost rune. +2% ArPen +2% Crit +15% Flank damage

 

High/High test:

-259 DPS double-blade dagger.  94 AoE damage + 10 per strike from frost rune. +5 Cunning +29% Flank damage

-259 DPS double-blade dagger.  94 AoE damage + 10 per strike from fire rune. +5 Cunning +26% Flank damage

 

Testing Conditions: 4-5 zones per test, damage figures were observed from frontal stealth attacks to negate flanking bonus.  Targets were Templar Marksmen (I think that's what they're called.  The archers at any rate) on routine difficulty.

 

High/Low observations: 500-600 damage.

 

High/High observations: 500-600 damage.

Results: Damage appears to be calculated using main hand dagger only.  Swapping to a lower DPS off-hand dagger resulted in no DPS loss from Shadow Strike. 

 
Armor Rating:

 
Alright, so the big question.  What exactly does armor rating do?  My tests have led me to conclude that armor rating is a flat damage.  57 Armor Rating should reduce incoming damage by 57.  That said, there are still other factors at play.  Resistances and armor penetration will both influence the damage you take.
 
How'd I get to this conclusion?
I ran a series of tests using T1, T1Alt, T2, and T2Alt Archer armors with no mods applied, all against the same enemy (Lesser Shade) on routine difficulty and recorded the incoming damage.  Once I had that data, I re-ran the T1 and T2Alt armor tests against a Lesser Shade on Threatening difficulty.  The result was roughly a 110 damage per strike increase (37 damage up to 150), yet damage reduction was still within 1 point of damage to the damage reduction recorded on routine difficulty.  If armor rating granted a percentile damage reduction based off of some sort of scale, then I would expect to see much higher damage reduction when upgrading armor.  Furthermore, my observations in the first experiment indicated that damage reduction scaled linearly in relation to armor rating increases.
 
That's great, but where's your data bro?
Right here.  Enjoy.
 
What was that about Venatori Gladiator ArPen being 75-80%?
Overestimated it.  After looking at my barrier/non barrier tests against the Gladiator, I'd say they're at 50% ArPen.  My bad.
 
Soo... that's cool and all, but what's it actually mean?
It means that Armor Rating is most effective against weaker enemies.  You'll see the most benefit out of T2 alternate armors on routine difficulty.  Damage scales so sharply on threatening/perilous that you probably won't get much benefit out of it.  I'd recommend saving those melee/missile defense mods and boosting defensive attributes (Con, Cunning, Will) if you want to be tankier.
 
Hey, this isn't what was here earlier.  You trying to take back your previous statements?
Yes.  No, not really.  If you really care, you can find my original comments below.  Don't know why you would, but there it is.
Original comments:

Spoiler

There's been a lot of questions about how exactly armor rating works, and whether it's worthwhile to upgrade your armor.  So I ran a series of tests to help determine the effect of armor.  However, I'm currently pretty tired, so I'll be editing this post after I've slept off the turkey to include all of my test data (which is more substantial than the previous tests.  I was feeling lazy for the DW tests and wasn't particularly concerned with the actual figures so much as the comparison between weapon strikes).  The edit will include all my test data and exact testing conditions so that anyone interested can evaluate my data themselves.
 
Until then though, I'll say this.  From my observations it appears that Armor Rating is a flat damage reduction and not scaling percentage value.  
 
So upgrading from the basic Archer's Tunic at 48 Armor Rating to the Peerless Archer's Tunic (Obsidian) at 67 Armor Rating should net a damage reduction of 19 damage per strike vs enemies with low armor penetration.  I was seeing damage received from Lesser Shades reduced from ~38 damage per strike, to ~20 damage per strike with the upgrade.  The tests were performed with T1, T1 Alt, T2, and T2 Alt armors and damage reduction progression was linear.
 
That said, I did find that some enemies have extremely high armor penetration.  Venatori Gladiators (the guys with the tower shields) appear to have ~75-80% armor penetration, and appears to be the reason they hit so much harder than the standard Venatori Zealots.
 
[Edit] Upon reviewing my data, I'd say Gladiator ArPen is actually lower than I was estimating, though not by much.  They appear to have 50% ArPen
 
Anyway, like I said, I'll get back to this post in a bit and update it with the actual data.  For now... nap time.

 

Weapon Properties:

 

"On Kill: Target explodes for 75% weapon damage" property: This does not appear to be based on weapon damage like it states.  Instead, damage appears to be based on the strike that killed the target.  Killing a target with a ~750-1000 damage Shadow Strike resulted in explosions dealing ~750 damage.  Killing a target with a basic attack resulted in explosions around 100 damage.

 

So yeah, the next big question is whether or not Armor Rating or Resistances are applied first.  Depending on how motivated I get, I'll probably end up running some tests to try to sort that one out here in the next few days.  Hope this info helps.

[Azrus]

Updated with more coherent information and actual test data.

[Klokos]

So I assume skills that deals percentage weapon damage without actually being a "strike" (like poison cloud or traps) only take into consideration the damage from your main hand ?

[Azrus]

Unknown.  I plan to test that here shortly.  I'm thinking either it uses main-hand DPS or it averages the two.

 

Personally, I'm hoping it's based off of main-hand DPS.  That'd open up the possibility of using off-hand daggers with special effects even if they're not the highest DPS dagger you've got.  If you build around single-strike abilities, then you'd only really be seeing the DPS loss in your basic attacks, which could be worth it depending on the special effect on the dagger.

[Azrus]

Updated with shadow strike and "On Kill: Target explodes for 75% weapon damage" test results.  It appears Shadow Strike calculates its damage based solely on the main-hand dagger.  On Kill: Target explodes property appears to calculate its damage based on the strike that killed the target, not the weapon damage, as such higher damage hits will result in higher damage explosions.

[Mindtraveller]

Thanks for the info.

 

It could be helpful to know for anyone who only has one good dagger to cancel the attack animation after every hit by clicking sprint, that way you'll never basic-attack with your crappy offhand dagger. Though this might get fixed at some point considering it's probably unintended and an exploit.

[Red Leader]

Maker bless you, thank you for your work.

[Azrus]

Not a problem.  Hope you guys find it useful.

[gay_wardens]

Thanks, this was really helpful. Kind of frustrating to learn that T2 armor is essentially useless until you get better armor upgrades to replace the ones on your T1, but at least now I know.

[EVILFLUFFMONSTER]

Thanks dude. Its always great to know this stuff. Its people like you that really helped in MEMP because there were lots of things left poorly explained. Cheers!

[gay_wardens]

Actually I don't know, I'm a little skeptical that armor with 20 higher rating won't be as good as the default legionnaire armor with just a couple points in constitution and something other stat.

[gay_wardens]

^ I mean, just because I don't have upgrades on the higher tier

[Azrus]

Yeah, well it's been mentioned by the powers that be that armor was intended to be more of a new skin than an actual defensive upgrade, so I suppose it makes sense.  On the one hand, I'm kind of happy to hear that, as it means I won't be stuck wearing ugly armor just so I can survive.  On the other hand though, I wish there were better options for defensive upgrades.  You can get lucky and find a lvl 20+ armor mod with +Ranged/Melee D, but those are slim odds and it'd be nice to have a little more control over it than that.