Damage formula research?

[Gardios]

From what I've seen, Floating Continent has enemies that cause Panic, meaning we can hit each other. Since we know our own stats, we could actually start finding out the damage formula. With that being said, has anyone started taking some numbers yet or does anyone know if the JP players did anything so far?

[SPOnion]

Panic cause you move the opposite direction on a player...

[Enforcer MKV]

Panic cause you move the opposite direction on a player...

Indeed, but we also have the ability to harm one another while we're...'panicked,' (I really just want to say confused...)

There have been a few screenshots showing players harming one another, I'll try to dig them up.

[HFlowen]

I always do very little damage to other players when I purposely try to harm them. Dunno if there's an innate damage reduction on friendly fire or if that's how the stats work out.

[SPOnion]

We can? Thanks for the inform...Now I'll try to get hit by the dragon that appears in E and see what happens...

[Valymer]

It's going to be very hard to collect data on this since it's random who you damage and when, but it's an interesting idea.

[Shadowth117]

Perhaps this could be done. Thing is, generally there's a lot of distractions around when you've become panicked and it doesn't last terribly long. It would be cool if people could figure out a formula from this, but I wouldn't count on it very fast. This would be far easier if there were simply a trap that panicked players.

[Chik'Tikka]

Damage other players while panicked....
+^_^+ it's Afin killing time!!+^_^+

[DoubleCannon]

PVP anyone? lol

[Asdfv]

http://ameblo.jp/yuugenbana/
http://ameblo.jp/mondlaterne/

The above blogs paraphrased, their findings on the damage formula:[spoiler-box]This does not take into account the following:
-Just Attacks' effects
-Target's hitbox multipliers
-Weak Bullet's effect on those hitboxes
-Effects of passive damage multipliers such Weak hit Advance and Standing Snipe
-Effects of Shifta, Shifta Drinks, and Shiftaride

damage equation with rounding steps (round to nearest integer values between [] brackets):
damage =
[(b_atk + a_atk) / 5 * mult]
+ [w_atk * (elem + 100) * random / {5 * (resis + 100)} * mult]
+ [t_atk / 5 * mult]
- [def / 5 * mult]

simplified estimation:
damage = {b_atk + a_atk + w_atk * (elem + 100) / (resis + 100) * random + t_atk - def} / 5 * mult

even more simplified estimation:
damage = (atk - def) / 5 * mult

b_atk, base attack:
Class's base attack & mag's attack, i.e. the portion used for equipment requirements

a_atk, affix attack:
Atk gained from the weapon's and the units' special abilities ("affixes")

w_atk, weapon attack:
The weapon's base atk & grinder enhanced atk, without the atk gained from the weapon's affixes
Only this portion is affected by damage randomization, element, and resistance

t_atk, temporary attack:
Temporary atk gained from skills such as Fury Stance

random:
Randomized value that varies from a value that supposedly depends on the attacker's and the target's ability stats to 1
The highest minimal value is theorized to be 0.9

elem, element:
The visible elemental value of the weapon if there is one

resis, resistance:
The target's resistance to the corresponding element of the attacker

def, defense:
The target's defense

mult, multiplier:
Value that depends on the attack used
(Examples of some estimations:
Fist's regular combo: 0.9, 1.24, 1.23 (1st, 2nd, 3rd attack)
Spear's regular combo: 0.87, 0.85, 0.83
Sword's first attack: 1.53
Rifle's regular attack: 0.24
Launcher's regular attack: 1.19
Gunslash(gun)'s regular attack: 0.6)

Shifta Drinks and Shiftaride boost the b_atk, a_atk, and t_atk portions (1.05/1.1/1.2x from drinks and 1.185x from Shiftaride), but a glitch causes the boost to a_atk to be ignored after changing equipment. Changing areas causes the displayed atk in the stat window to also lose the bonus to a_atk, but this doesn't seem to affect the actual damage inflicted.[/spoiler-box]

[Jonth]

Pretty cool Asdfv. Any information on how weapons derive their attack values? For example, I have a theory (not initially mine) that certain weapons, such as the rod, formulate their damage from two stats. In the case of the Rod, it would be S-ATK and T-ATK. I'm looking for some verification to either prove or disprove this now that the game is in release. I am planning to do some testing, but I currently don't have adequate equipment .

[Asdfv]

The blog authors seem to mainly play ranger and hunter and I couldn't find any mention of that or rod damage on their blogs.

[Zyrusticae]

Huh. Element being a % of weapon damage and ONLY weapon damage actually makes some semblance of sense... But it still seems implausible. I think I'd notice an increase of damage roughly equivalent to 200 more points of ATP. Doesn't seem anywhere near that high in actual play...

[Tanarin]

Hello, new to the site, but had some input on all this.

As far as the damage code is concerned, I am not too shocked it is like this. Seems pretty reasonable the modifiers are all on the weapon and not factored into player side stats. As far as element damage is concerned, 200 damage seems high, but we are missing 2 important parts to the equation.

A) Enemy Defense stats
B) Actual Enemy resistance numbers.

A is obvious, but B may not be too obvious. B needs to be considered because not every game takes weaknesses the same among mobs. One mob may be weak to Ice, but another mob may be WEAKER to the element and thus a higher elemental weakness rating. More testing would be needed to know for sure.

Also, until stuff like weak bullet and Just attack is taken into account, the formula will be inaccurate.

[Gardios]

Awesome stuff, Asdfv! I'd love to test myself how JAs, weak spots and most importantly Ability factors in exactly, but I'm still don't have Floating Continent unlocked. lol


Huh. Element being a % of weapon damage and ONLY weapon damage actually makes some semblance of sense... But it still seems implausible. I think I'd notice an increase of damage roughly equivalent to 200 more points of ATP. Doesn't seem anywhere near that high in actual play...
I haven't played with elemental upgraded weapons myself yet, but the low damage increase makes sense due to the 0.2 multiplier at the end.

[Angelo]

Perhaps this could be done. Thing is, generally there's a lot of distractions around when you've become panicked and it doesn't last terribly long. It would be cool if people could figure out a formula from this, but I wouldn't count on it very fast. This would be far easier if there were simply a trap that panicked players.

Doesn't last terribly long?

The damn crystal dragon had me confused for a good 30 seconds straight yesterday.

[Tanarin]

OK, seems there was some updates to the blog entry posted here: http://ameblo.jp/yuugenbana/entry-11307321219.html

Google translate is failing me by making it unreadable, any idea what the corrections are getting at exactly? From my guess, it is this:

Shifta drinks are applied directly to the attacker's attack value.

Also the other blog here (http://ameblo.jp/mondlaterne/page-2.html) (Which is doing the main amount of work,) is stating that Shifta and shifta drinks are multiplicative in it's effect. Assuming I am reading all this correctly of course.

Also the 2nd blog clearly states that the Shifta Bonuses are only applied to base attack. Doing otherwise would result in much higher damages than they are seeing.

Edit: it also seems as though the formula presented on page 1 is off due to Google translate issues. The formula by my best estimations is as follows:

[(Base Attack + Affix Attack) + {Weapon Attack * (Elemental Bonus + 100) / (Elemental Resistance + 100) + Temp Attack Bonus} - Defense ] / 5 * Multiplier * Random Value

The big change is moving the random value to outside the factoring for weapon based attacks. Not only does this page (http://ameblo.jp/yuugenbana/entry-11306618062.html) seem to support that change, but it just makes more sense if it worked that way in the first place.