Jump to content

Damage formulae, weapon skill effects, personal skill effects


Tables
 Share

Recommended Posts

3 hours ago, SourPeridot said:

[Questions]

While I'm not Tables...

Regarding damage reduction, damage being divided by Def/Res does fall in line with my own experience and would explain the following:

1. Why low-level enemies aren't obliterated the instant they touch 240 Mt lightning.
2. Why it took me roughly 3 times as long to kill a high-level enemy with a bronze sword (10 Mt) compared to a brave sword (240 Mt) using Lucina (93 Str) instead of taking me over the 60-minute time limit to kill a single guy.
3. Why, despite endgame enemies packing brave weapons (brave weapons, by the way, greatly outscaling your defenseive stats), Elise and her 42 Def don't take 300 damage from a physical hit.
4. Why, despite having 121 Def, Frederick's damage taken isn't zero from the same physical hit.

Regarding how individual attacks in a combo are factored, I'd wager multiplicative. Being additive means it wouldn't scale very well. Also, different attacks -definitely- don't deal the same amount of damage. *Looks at Takumi's Strong I*

And while you didn't ask this, Astra is not a trigger skill, nor is Luna nor Sol nor Aegis nor Pavise. Aether's not even in the game. :p

Link to comment
Share on other sites

4 hours ago, Technoweirdo said:

While I'm not Tables...

Regarding damage reduction, damage being divided by Def/Res does fall in line with my own experience and would explain the following:

1. Why low-level enemies aren't obliterated the instant they touch 240 Mt lightning.
2. Why it took me roughly 3 times as long to kill a high-level enemy with a bronze sword (10 Mt) compared to a brave sword (240 Mt) using Lucina (93 Str) instead of taking me over the 60-minute time limit to kill a single guy.
3. Why, despite endgame enemies packing brave weapons (brave weapons, by the way, greatly outscaling your defenseive stats), Elise and her 42 Def don't take 300 damage from a physical hit.
4. Why, despite having 121 Def, Frederick's damage taken isn't zero from the same physical hit.

Regarding how individual attacks in a combo are factored, I'd wager multiplicative. Being additive means it wouldn't scale very well. Also, different attacks -definitely- don't deal the same amount of damage. *Looks at Takumi's Strong I*

And while you didn't ask this, Astra is not a trigger skill, nor is Luna nor Sol nor Aegis nor Pavise. Aether's not even in the game. :p

Different damage from different moves in a character's arsenal is likely explained by there being different kind of multipliers assigned to various attacks (along with some attacks being multi-hit).

Link to comment
Share on other sites

@SourPeridot I suppose the simplest explanation to most of your questions is, yeah, strictly speaking the formula should also have a "x Move multiplier term" appended to the end. I didn't really think to add that in, probably should do that though. While things like your basic string seem to usually do the same damage per hit, charges, specials and so on each do considerably more.

On top of that, the formula in its current form gives relative damage, not necessarily absolute amounts. Mostly, that means it's a way to compare similar situations accurately. For example, how much damage do you lose by putting slayer skills onto a 5* Bravestone? This formula helps answer that more accurately. Or how much damage do I gain by adding Pair Up+ and Strength IV? Again, that's the kind of question it helps answer. It's not very good at comparing, say, how much damage will Takumi do against this General compared to Elise, since there's so many values you (correctly assume) probably need to datamine to find accurately.

 

Go ahead and fire your Luna questions as well. I think we've got a decent understanding of how it works at this point.

 

@Brother of MetalI too think that's a possibility as well. It's worth noting that the game highlights which stat a character (topsy-turvy notwithstanding) uses, and it doesn't highlight both Str and Mag for characters like Celica, however her specials along with characters like Sakura could specfically use Mag instead of Str. Might be worth testing in the future.

Link to comment
Share on other sites

I find Sakura's description in the guides to be interesting. "While these [bow] attacks lack in power, her Warrior Specials are quite effective." So I decided to try a crude test. I'd attack an enemy with a Topsy-turvy bow,  then I'd attack the same one with a bow of equal strength and no skills attached. Here's what I got; 

Spoiler

 

DO5OGoQVQAEBvU8.jpg:large

DO5OUMdUEAUFSSJ.jpg:large

 

I can't see any visual differences, oddly enough. This could mean that warriors specials ignore the effects of Topsy-turvy - in which case, we wouldn't know if they're magic or not. I'm going to compare them directly in an image editor just to make sure though.

Edit:

TopsyTurvyComparison.jpg.cca9e73df2ab59d21b026961b858e9f7.jpg

It's hard to see, but there is a tiny difference. The one on the top is with Topsy-Turvy and the one bottom is the one without the Topsy-Turvy effect. If the sniper has higher defense then resistance then the magical attack would be the one with higher damage. This would mean that Sakura's Warrior special is indeed magical. I'm going to test this with a more powerful weapon and see if the difference is any more noticeable, just to be safe.

Another Edit:

Spoiler

DO5T4oAVwAApU0Y.jpg:large

DO5T8wyUQAAHGDf.jpg:large

Okay, yeah, it's much more noticeable with a Brave Bow. Either this sniper has some awesome resistance or Sakura's Warrior Specials are definitely magic. This probably applies to other characters as well.

Edited by Zachmac
Link to comment
Share on other sites

10 hours ago, Zachmac said:

 Either this sniper has some awesome resistance or Sakura's Warrior Specials are definitely magic.

While I wouldn't say that sniper has awesome resistance, if Takumi's the 'base' for archers, then the sniper should have better Res than Def. Could've been confirmed via a full-charge Strong I or any other attack or combo that does an appreciable amount of damage.

That said, I'll spare you the additional testing and say that, after firing off two warrior specials at a Lv. 64 mage (who I know from previous tests has way better Res than Def) at the Lv. 64 arena in the Gaiden map -- once with Topsy-Turvy on Spellbane Yumi, once without -- that Sakura's warrior specials are physical, not magical.
-
Maybe they should've put Topsy-Turvy on Spellbane Yumi from the start to give her a better start. Edit: Actually wait, at 30 Mt until upgraded, she'd generally be better off with a regular silver weapon as you progress. Maybe they should've just made Topsy-Turvy bows available earlier instead. Either way, poor Sakura. :(

Edited by Technoweirdo
Link to comment
Share on other sites

A good thing to do for a more direct comparison with that Sniper is to also compare the damage dealt by some consistent set of basic attacks. For example, hit with a fully charged C1, once with Topsy Turvy, once without. That would be a good way to estimate the Sniper's relative Def/Res, assuming you also recorded her stats at the time.

 

Oh, and also, Armour Break has been tested by other people, Graphyte mostly, and we found that it halves enemy defences just like it halves your own. It stacks multiplicatively with Luna, or in other words it also doubles your damage with Luna or without.

Edited by Tables
Link to comment
Share on other sites

24 minutes ago, Tables said:

Oh, and also, Armour Break has been tested by other people, Graphyte mostly, and we found that it halves enemy defences just like it halves your own. It stacks multiplicatively with Luna, or in other words it also doubles your damage with Luna or without.

It's pretty damn bonkers.

Doesn't work on Velezark, though.

Link to comment
Share on other sites

40 minutes ago, Tables said:

Oh, and also, Armour Break has been tested by other people, Graphyte mostly, and we found that it halves enemy defences just like it halves your own. It stacks multiplicatively with Luna, or in other words it also doubles your damage with Luna or without.

Yikers. Will Armor Break always happen if you get hit with a strong attack?

Anyways, I really must commend the people who went out of their way to test all this.

Link to comment
Share on other sites

I just used Celica to test to see if snipers had lower defense or resistance, since her attack and magic are equal (at least at the level I have her at). Yeah, their resistance is definitely better. Silly me, I just assumed that they had better defense since they're a physical class. 

Just to be safe, I did one last check on a manakete who I know with absolute certainty has significantly higher defense then resistance (I tested this with Celica as well), and Sakura OHKOed with Topsy-Turvy and couldn't even bring it under half health without. Yeah, that's definitive proof that all of her attacks are physical. Topsy-turvy really is the only way to go with her. 

Edited by Zachmac
Link to comment
Share on other sites

  • 1 month later...

Discoveries on using the new bond crystal and arena fights:

9KOs is enough for C and B supports. Each battle takes less than 5 mins, about 3.5-4.5 depending on how long it takes to cycle through the prompts, so better than even a short onslaught.

2 9KOs arena fights was not enough to go from B to A but doing a full arena of 22KOs was. Now it may be possible to get enough champion kills in a short onslaught but I don't know that I can accurately keep track of all my significant kills so I'll let someone else confirm that.

Either way you're looking at ~15-20 mins to get an A support vs what seemed like an hour+ pre bond crystal.

Edit: Nevermind, that all got shot to hell. While all that information may work on characters with no support, if they already have some support things get wonky. Makes me wonder if having it equipped not only increases the rate support is gained but also applies it to all support levels or something.

Edited by CmdrQuartz
Link to comment
Share on other sites

  • 4 weeks later...

It's been a while, but I want to update a few with a few things that have been tested and discovered:

 

Sol: Heals ( Enemy Max HP / 400 ) x Luck

See below for some details on enemy stats

 

Character stats/growth rates: In short, characters have a fixed set of level 1 and level 99 stats, and their growths are just the difference between these /98, then rounded to the nearest whole number. Or in other words, character stats at every level are entirely predictable once you know both level 1 and level 99 stats - and level 1 stats can be derived using a bit of maths even for characters who join above level 99. You can look at the calculator here to see everyone's level 1 and 99 stats as well as work out their stats at various levels.

 

Enemy stats: Generic enemies have some fixed rules for how their stats work.

HP = 60 + ( level - 1 ) x 540/98 + promotion bonus.

Promotion bonus is 50 HP for promoted enemies. 540/98 is roughly 5.5 points per level, for reference.

540/98 seems like a super precise and bizarre value to say with certainty, but it basically comes from the above paragraph on stat growths. Since growths for characters are always x/98, and it became clear enemy HP growths are marginally above 5.5 points per level, the logical thing to test was in that form. 540/98 seemed to work well for this. 

(As an aside, I'm aware 540/98 simplifies to 270/49. I've kept it over 98 as I think that helps make the value seem less arbitrary - and makes it easier to compare to other growths which are of a similar form).

 

Def and Res fall into one of four ranges, which I've creatively named "Very Low", "Low", "Mid" and "High". Most enemies at equal level will have one of four defensive values, dependent on their class. For example, Myrmidons have Mid Def and Mid Res - meaning they take equal damage from an equally powerful physical and magical hit, and also take the same damage as an Archer would take take from magic, as they have Mid Res as well. However, Archers have Low Def, so they take more damage from physical attacks.

The exact values I'm not 100% sure on yet, in particular the scale enemy defensive stats use is something I've yet to work out with certainty, but I have some relative values that seem reasonably accurate:

Very Low: 2 base, 8.5% growth, 20 promotion bonus

Low: 4 base, 17% growth, 20 promotion bonus

Mid: 6 base, 34% growth, 20 promotion bonus

High: 8 base, 68% growth, 20 promotion bonus

You can see a table showing these values for each generic class (as well as character enemies) here.

I did say "most" at equal levels have these values. In some missions, enemies may have raised or reduced defences - these tend to be a scalar multiple of both stats, effectively not changing their relative def/res. Some enemy types (usually mission critical ones like targets you have to kill) also have increased defences, again a scalar multiple. But in most cases, enemies at identical levels behave largely identically.

 

Str/Mag is hard to tell, but one thing I have observed is that enemy weapon appears to be purely cosmetic, or at least that enemy damage can vary significantly more than just based on level, as Def/Res do. For example I took more damage from story mode lunatic enemies with bronze weapons than I did from lower level History mode enemies with steel.

 

Skl, Spd and Luk have no real reason to exist for enemies. Even if they do exist, there's no way short of hacking to measure them, and they don't do anything anyway.

 

Character enemies seem to have identical stats to their player character self, or at least a scale multiple of them defensively. Tests with Sol found that their max HP was exactly as one would expect, while tests against their Def/Res found them to be in roughly the correct ratio. Validar (lvl. 30) was briefly tested, and found to be using generic dark mage stats - albeit with scaled up def/res, as a character enemy often would. It is likely he would use promoted stats in higher level missions.

 

Lethality does need a bit more testing, but early observations suggest its trigger rate is perhaps something like ( Luck /200 ) %. Its damage multiplier is huge, perhaps in the 2x to 3x damage range when it triggers.

Edit: Lethality very likely has a (Luck / 200) % activation rate or something quite close to that. It seems to be a +400% damage bonus when it triggers.

 

Fury Builder is +5% damage (additive) per 100 KOs. Each character has their own separate kill count, and it only increases with player kills. The max limit seems pretty high, you can get to at least a +100% bonus and perhaps more from Fury Builder.

 

Breaker skills give +75% damage (additive) against their specific weapon type.

Edited by Tables
Link to comment
Share on other sites

  • 4 months later...

Haven't been paying this game much myself of late, but I've been watching several others (friends/family) playing the game of late, and have noticed that PC mages seem to have more trouble dealing damage (with equal weapons) against high-res enemies like mages than PC physical units do against high-def enemies like knights. The mechanics suggest this shouldn't be the case. Is there a reason I'm missing or are we just crazy? I know that fighters have an easier time getting WTA (e.g. just send Frederick/Camilla/etc. against the knights) which helps but doesn't even in neutral cases this seems to be the case.

Thanks so much for doing this, once again.

Link to comment
Share on other sites

On 5/27/2018 at 11:37 AM, Dark Holy Elf said:

Haven't been paying this game much myself of late, but I've been watching several others (friends/family) playing the game of late, and have noticed that PC mages seem to have more trouble dealing damage (with equal weapons) against high-res enemies like mages than PC physical units do against high-def enemies like knights. The mechanics suggest this shouldn't be the case. Is there a reason I'm missing or are we just crazy? I know that fighters have an easier time getting WTA (e.g. just send Frederick/Camilla/etc. against the knights) which helps but doesn't even in neutral cases this seems to be the case.

Thanks so much for doing this, once again.

Don't know if playable mages have weaker attacks than pegasi et. al. or if enemy mages just have that much more Res than knights do Def -- best resource on that afaik would be this which just gives approximations Will say though that attack speed and stun mechanics are definite factors, mages attacking more slowly than most and having to charge up their C1 before they can get a stun. Trade-off is that mages get to stun anyone and, last I checked, they can charge their C1 really quickly. Well, the cavalry mages at least.

Link to comment
Share on other sites

Hm, I never really noticed much of a difference with mages vs. mages or physical vs. Knights personally. I do know that every different weapon type has different base power, which could be a factor, as @Technoweirdo said. I know for example that Pegasus Knights are nautrally lower damage than Wyverns, even with identical stats and skills, for example. It wouldn't surprise me too massively if other mages are naturally a tad weaker too.

 

Maybe I'll test this at some point. Got some spare time this week. I can also write up a few other things I've found since I last updated - notably several other skills including the new ones.

Link to comment
Share on other sites

Mages' attacks probably have a lower power than physical weapon classes simply because of how much spread and range their attacks have.

Archers are generally weaker than their fellow physical weapon types, too.

Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

 Share

  • Recently Browsing   0 members

    • No registered users viewing this page.
×
×
  • Create New...