Skill damage mechanics

[lamabbio]

First of all, sorry for my english. Secondarily, I've done some tests and maths to try to understand skill damage mechanincs...

Let's start with an Alchemist example: level 10 Ember Lightnings "depends on 100% magic", the damage is 20% weapon +218-515 electric damage.
Alchemist starts with 10 magic.
Let's talk about the minimum damage, 218.
Each point spent in magic adds 1% of 218 to the 218 base minimum damage, so if I put 1 point in magic the minimum damage becomes 218+(1/100)*218~220. Exactly, ~0,908% fits my data better than 1%. The same works for the maximum damage.
So if I put 100 points in magic (ndr. my alchy would have 110 total magic) then level 10 Ember Lightnings becomes 20% weapon +416-983 electric (218+0,00908*218*100~416 and 515+0,00908*515*100~983).

Some other Alchemist skills "depends only on 50% magic", like Ember Lance or Ember Bolt. The damage of level 10 Ember Lance is 8% weapon +159-397 fire damage.
Let's talk about the maximum damage, 397.
Each point spent in magic adds 0,5% of 397 to the 397 base maximum damage, so if I punt 1 point in magic the maximum damage becomes 397+(0,5/100)*397~399. Exactly, ~0,476% fits my data better than 0,5%. The same works for the minimum damage.
So if I put 150 points in magic then level 10 Ember Lance becomes 8% weapon +273-680 fire damage (159+0,00476*159*150~273 and 397+0,00476*397*150~680).

Ember Shield depends 50% on defense and 50% on magic. There's nothing new here: each point spent in magic or defence adds 0,476% of base damage to base damage (to be accurate, absorbed damage).
Similar reasoning applies to Ember Phase, that depends 50% on strenght and 50% on magic.

Finally I can say that Iceshot, Pyre and Ember Strike depend on 100% magic, even if its not listed in the skills description. So if I put 5 points on magic then level 10 Ember Strike maximum fire damage will become 764+0,00908*764*5~799.

I can also confirm that Vanquisher's skill "Arrow Hail" is 50% based on Dexterity.

My open questions are:
-> How does Crit works?
-> What is output damage formula?
-> How is calculated the damage on the right of the particular stat? For example, how is calculated magic damage from magic stat?
-> Monster resistances are % based or not?

About the last one, I noticed with pyre that if the damage listed is about 600 then the damage to lvl50 monsters is ~300, if pyre listed damage is 900 then same monsters got ~450. That should demonstrate that monsters resistance are % based, the higher the level the higher the resistance.

I'll continue my tests, and I hope we can talk about 'em

Your contribute:
-> Corseth performed many tests and reassumed his findings this way:
"So to repeat. If you have zero magic stat (or close to zero) and virtually no weapon damage, then the skills do the predicted amount of damage. However as you add magic and weapon damage, the skill damage goes up slower than the damage formula suggests, to the same level 1 monsters."
-> Ctuchik noticed unexplainable high critical hits (me too)

[Anon]

Ah, now i understand it better. The game really needs to explain it's skills.
Very nice, thx alot!

[yoob]

Finally I can say that Iceshot, Pyre and Ember Strike depend on 100% magic, even if its not listed in the skills description. So if I put 5 points on magic then level 10 Ember Strike maximum fire damage will become 764+0,00908*764*5~799

you sure? that would be great news as i haven't really tried those skills seriously yet

as for "weapon damage" in regard to alchie skills, its the true right-hand weapon DPS. "true", as in after stats/buffs are taken into account. ember shock works great with a slow, high damaging weapon

[lamabbio]

@Anon
I'm happy I was of help

@yoob
Yes, I'm pretty sure

Update on damage mechanics in main topic!

[Melianor]

I was already wondering what that 50% Attribute marks meant. Now its pretty clear and i think i will have to redo my alchemist from scratch, to support for both Lightning and Ember Lance.

I will see if i can find out anything else new along the way. Great post and nice to see someone doing the math behind the pics.

The added damage from items and attribute does not show display with the basic weapon damage on the respective skill tab, right?

[Bremze]

The Magic bonus seems correct on Ember Shock(there is a tiny error margin, but thats expected). Are you sure its weapon not weapon damage on strike instead of dps? Because it seems to me that slower weapons with less dps but more damage on strike add more.

I think the base crit damage is twice the regular damage. I have maxed the Critical Strike skill and 80% to crit damage comes from items so a total of 200% to critical damage. I have never seen anything higher than 4x of my usual damage. Also I might be biased, but I think the game somehow round down crits. I do around 2-4k damage per hit but I have never seen me crit higher than 12k. Plenty of 9k and 10k though.

[lamabbio]

The added damage from items and attribute does not show display with the basic weapon damage on the respective skill tab, right?

Sorry I didnt understand this, u mean the tab on the right of magic, after pressing C? (Sorry I'm not english, sometimes sentences are not so clear )

Are you sure its weapon not weapon damage on strike instead of dps? Because it seems to me that slower weapons with less dps but more damage on strike add more.

I'm really unsure about the weapon effect, that's why my tests are based almost on Ember Lance whose weapon effect (8%) is minimal. I'll do some more tests on Ember Strike tomorrow and post here to confirm or not your impression

I think the base crit damage is twice the regular damage. I have maxed the Critical Strike skill and 80% to crit damage comes from items so a total of 200% to critical damage. I have never seen anything higher than 4x of my usual damage. Also I might be biased, but I think the game somehow round down crits. I do around 2-4k damage per hit but I have never seen me crit higher than 12k. Plenty of 9k and 10k though.

I have the same crit% on my alchy.
I've seen very very few crits of 5x my usual damage, the most part being around 2-3x.
IMHO that's why the damage inflicted depends also on the monster resistances: I noticed that Pyre does often half of the listed damage, so I suppose that monster's fire res is about 50%... and that explains the apparent 2-3x of crits.

Thank you all for partecipating, I hope that togheter we can understand a bit more of how TL works

[TiberiusAudley]

Then Critical Strike applies. After other tests I think that, without spending points in the Critical Strike skill, CS triplicates the base damage. So, in formulas, for each successfull Critical Strike you got
CRIT.DMG. (SKILL 0) = DMG * 3
If you spend 10 points in the skill you have 120% Bonus to Critical Damage (and +10 Critical Chance, whose meaning seems pretty obvius), so what should happen is
CRIT.DMG (SKILL 10) = DMG * (3+120/100)

Hope someone else has made some tests too, so we can talk about it

I'm pretty sure the crit, by default is either +50% or +100% damage. If you're seeing it deal triple your usual damage, there are a number of things that can be happening.

Let's say for example your base damage is 50-100 per hit, and your target has 30 armor.

You deal 20-70 armor per hit. On a crit, (assuming +100%, just for good measure), you'll deal 100-200, and thus after armor, see 70-170... This is about 2.5-3.5x damage from your base. The reason it works this way is that armor and resistances are a flat reduction of damage in this game rather than a percentage (I'm pretty sure about this, but could still be wrong). Your 'triple' display could also be coming from getting low damage range on your default attacks and high damage range when you crit, showing a further gap than normal.

[lamabbio]

I'm pretty sure the crit, by default is either +50% or +100% damage. If you're seeing it deal triple your usual damage, there are a number of things that can be happening.

Let's say for example your base damage is 50-100 per hit, and your target has 30 armor.

You deal 20-70 armor per hit. On a crit, (assuming +100%, just for good measure), you'll deal 100-200, and thus after armor, see 70-170... This is about 2.5-3.5x damage from your base. The reason it works this way is that armor and resistances are a flat reduction of damage in this game rather than a percentage (I'm pretty sure about this, but could still be wrong). Your 'triple' display could also be coming from getting low damage range on your default attacks and high damage range when you crit, showing a further gap than normal.

Think you're right, I've just seen a crit too high for my model. So I guess Crit doesnt work as I wrote :/

Another question is: how is calculated the damage on the right of the particular stat? For example, how is calculated magic damage from magic stat?

[Dzheph]

It's an untested theory (working on a test), but I'm guessing there's a distinction between casting skills and attacking skills. Among other things, one difference I predict is that casting skills will not transfer +Elemental damage from gear (outside of weapon DPS), while attacking skills will. Some skills specifically mention the use of weapons, and I'm guessing that makes them eligible for various benefits (leech? Knockback?).

Skills that appear to be attack based:
-Slash Attack
-Devastate
-Ricochet
-Frost Pilum
-Explosive Shot

Unclear whether they are attack based:
-Spectral Echo
-Stab

It could also be that skills that use weapon DPS will also use on-strike powers--but I somehow doubt that ember strike is giving any kind of leech, nor carrying any +frost damage from my rings.

I'll report findings if I figure out a way to test this stuff--unless someone already has concrete answers.

Thanks! (and great work on this)
-Dz

[lamabbio]

I somehow doubt that ember strike is giving any kind of leech, nor carrying any +frost damage from my rings.

True, I have a +ice damage ring but ember lance doesn't freeze. And, If i remember well, doesn't leech too. But correctly knockbacks.

[Dzheph]

After some unscientific testing with a +poison amulet and no other gear (aside from a level 1 gun), I can report the following:

- Ricochet seems to benefit from elemental damage
- Seeking Shot does not seem to benefit
- Wind of Justice pretty conclusively does not benefit (it's rank 1 damage is small and consistent enough that variation would be noticeable)

[in case devs are watching] Figuring this stuff out through dungeon trials is sort of interesting, but it's also pretty frustrating. It's like we have most of the variables but no equation to solve. And for those of us who don't get off on algebra, a skill cast damage preview would also solve the problem, or at least let us do the math without having to also run for our lives.

[Twotricks]

Question

If skill says 10% of weapon damage.

What happens when you dual wield ?

Is sum of both weapon damage calculated ?
If not, is it right or left hand ?
Or is it the highest damaging weapon that is used ?

Also another question:

Are elemental damage bonuses like 10-20 fire damage (or other bonuses) also calculated ?

Also question:
What is is calculated. Weapon DPS or pure damage ?

[Ethereal]

I 've found the damage formula of all direct damage skill( based on fixed damage, weapon DPS, and weapon damage ).
I'll post them up. But the formulas are really complex ,and my English isn't so good, so the procedure maybe long. Please be patient .

[lamabbio]

I 've found the damage formula of all direct damage skill( based on fixed damage, weapon DPS, and weapon damage ).
I'll post them up. But the formulas are really complex ,and my English isn't so good, so the procedure maybe long. Please be patient .

I'm patient Any news?

[Corseth]

Skill scaling formula are very obscure in this game. Nothing works like you'd expect it to.

By the way, in the above formula, 397+(0,5/100)*397~399 -> x + (y/100)*x, you can simplify that to (1+ y/100)*x where x is the damage range and y is your magic score times the multiplier for the skill.

Testing with a magic score of 186. Weapon is a plain white wand with 126-251 fire damage (resulting in 386-771 Weapon Elemental Damage). No other gear was equipped.

Example, level 1 Ember Lightning. Magic 100%, 140-331 Lightning Damage, weapon damage 20%
186 magic score,100% modifier should result in 186% increased damage

I go to floor 1 and shoot a level 1 monster. I hit for 182 damage. 291 is the highest I ever see it hit for outside of a crit. That's about right (remember the 140-331 is after my magic score's multiplier since the improvement to base damage is on the skill sheet; starting a level 1 alchemist lists level 1 ember lightning as something like +40-110 damage or so)

Now, let's equip that weapon.
I SHOULD be doing 140-331 + 20% of 383-765, or 216 - 484. However I routinely see hits in the 140-200 range, a few in the 200-260 range, and never one 270 or more. Now, lets look at the numbers if we use the weapon's DPS (238) there. I'd expect 193 - 384. This is still too high for the damage I'm dealing. So let's use the weapon's base damage range of 125-249. This gives us 165 - 380.

Repeated zapping through level 1-5 enemies was yielding a minimum of 150 damage and a max of 319 This isn't right by any of the formula, though it's closest the third. And the high is awfully low, in almost all of these cases I should be seeing more hits over 300 than just the few I did. The average hit was around 200. Normally I'd say that minimum of 150 is just due to a resist (though these are awfully low level enemies to have a lot of resistance), but I tried rank 1 ember bolt.

18-26 poison damage, 10% weapon damage.
So we'd expect either either 38+18 to 76 + 26 = 56 to 102 if it's the damage shown on the character sheet as weapon elemental damage, 24 + 18 to 24 + 26 = 42 to 50 if it's weapon DPS, or 12 + 18 to 25 + 26 = 30 to 51 if it's the weapon's damage range.

Repeated bolting yielded a low of 38 and a high of 73, with most hits in the 50's (again, ignoring crits). So we've got an observed range of 38 to 73. the 38 is lower than any of these possibilities, so might could be a resist, but the high of 73 only fits in with the first option. However, it'd be an extremely low high for that range and I should not be seeing nearly so many hits around 50 damage (which was by far the most common). However, while it almost always hit for 50-60 with the occasional low in the 40's and high (non-crit) in the 70's, so either level 1 monsters are running around with 25-30% resists against magic or something is really borked, as none of the formula match the observed results very well.

Just to make things more "fun" I maxed out Magical Weapons Expertise (+40% weapon elemental damage) and Infuse (+75% ranged and melee damage for 75 seconds). Going back to rank 1 Ember Bolt, I saw a non-crit high of 111, and my average went up from 50 to 70. But I STILL saw a non-insignificant number of hits in the 40's.

So I dunno what the formula really is, but it seems pretty messed up. The damage done by spells in practice doesn't easily match anything the 'theorycraft' puts out.

No animals were harmed during the making of this post. (Several hundred Ratlins, however, were sacrificed in the name of science)

[outlander78]

I'm not sure if your testing took this into account, but creatures absorb damage, so the final damage done isn't just affected by your stats. "Glancing blows" occur when they resist more than you do (ie 0 damage - just learned this myself).

[Corseth]

I mentioned resists might be possible, BUT

I'm a level 44 Alchemist, and I'm testing on the FIRST floor of the dungeon. These are level 1-2 creatures, and in order for resists to be the culprit they'd have to be walking around with 50+% resistance in several cases. I made a new Alch with his piddly gear and 10 points of Magic and rank 1 Ember bolt and ran in, and Ember Bolt hit for almost exactly what the tooltip said. If the monsters had 50%+ resistance it woulda been very noticeable at that point. It's not the monsters, it's the scaling.

Resists don't explain non-crit maximums ABOVE the projected damage max, either.

[lamabbio]

@Corseth Ty for nice post

Now, I have two hope-not-wrong hints for you

First, I think monsters resistances are not % based... i.e. I think that 10 fire res reduces by 10 the fire damage taken. So, maybe that the different effect you got from Ember Bolt depends on the fact that is a poison damage skill, while Ember Lightnings is lightning ofc. If you tested both skills against same monster, maybe that that particular monster have "+15 min +50 max lighning res" while "-15 min -50 max poison res". If that's true Ember Bolt would work exactly as Ember Lightnings.

Secondarily, Ember Bolt is only 50% based on magic: the starting damage should be 14-20 instead of 18-26 (after applying the "magic bonus"). This could match better the first hint with your data.

Sounds forced, isnt it? Well, its just a try... lemme know what u think about it!

May I finally suggest you to try with much higher skill levels against low level monsters? This way their low res effect would be minimal, I guess

[Corseth]

Yes, I did. And Ember Bolt already lists the damage post-magic stat on the tooltip (the damage you see in the +x poison damage is the damage WITH your magic stat already calculated in; compare a high level alch to a low level alch but the skill the same level, and you'll see the high level alch's does a lot more damage).

Also, I tested both skills in the same level (floor 1). Ember Bolt was doing about 40 damage less than predicted average and Ember Lightning about 100 less. So either the level 1 monsters have 50% resistance (rediculous, because level 1 characters with no scaling power do the predicted damage), reduce poison damage by 40 and lightning damage by 100 (also rediculous, because then a level 1 character could never hurt them with poison or lightning and that's also not the case), or the skill is not scaling the way it says it does.

Rank 1 and Rank 10 saw the same problem. And how could a new level 1 character have rank 10 ember bolt anyhow? They'd be using rank 1.

Given that 2 of these 3 would make it impossible to kill any monster in the game at level 1, and I'm pretty sure that's not the case I'm quite sure skill scaling is not how it's supposed to be right now.

So to repeat. If you have zero magic stat (or close to zero) and virtually no weapon damage, then the skills do the predicted amount of damage. However as you add magic and weapon damage, the skill damage goes up slower than the damage formula suggests, to the same level 1 monsters.