Jump to content
Sign in to follow this  
Kruro

Acid Demonstration Formula

Recommended Posts

23 hours ago, Kruro said:

So much conflicting info :< Is there a definitive resource you can point me to @Pokemon? Or is that a result of testing?

I did some testing myself by swapping between Mad Bunny/Immune Shield, both with KK card, just to emulate the single loss of 5% atk (which converts to +5% Class All Bonus in @battlestats). Without the 5% atk, I was doing 98% of the damage that I deal usually with the mad bunny on, but I only did 10 ADs with each shield, so I can't really trust my data here.

Additionally, if +%atk (aka Class All Bonus) wouldn't work, that'd mean Abysmal Knight shouldn't work either, since those boni stack up additively in @battlestats.

 

The ATK to MATK ratio is useful though, that definitely clears up my question about the armor slot! :) 

Renewal Acid Demonstration is one of those nebulous formulas that have no good answer to because the official game has it act so bizarrely. As a result, there's a lot of conflicting information as people try to get closer to approximating it correctly.

 

Personally I refer to rAthena when I need a formula check, though this doesn't necessarily mean it's exact to Nova. Some bugs still exist in the rAthena implementation which Nova fixed.

 

Though now I'm pretty confused on how Min got the results he did regarding ATK% and the impact of ATK vs MATK.

When I checked there's no flag to ignore % modifiers in the skill database : https://github.com/rathena/rathena/blob/master/db/re/skill_db.txt#L718

And the formula has ATK and MATK in equal parts: https://github.com/rathena/rathena/blob/master/src/map/battle.cpp#L6575

 

And then rAthena has these outstanding issues related to it:
Creator Acid Bomb/AoE Bomb formula is wrong and bypasses alot of things: https://github.com/rathena/rathena/issues/803

  • Notably for this one, ATK is capped while MATK has variance...I suppose this is where Min got his 1 ATK = 1.25 MATK?

Acid Demonstration should be neutral (Nova fixed this ages ago): https://github.com/rathena/rathena/issues/2025

And this one, most confusingly of all, bonus errors: https://github.com/rathena/rathena/issues/2496

  • Reporter claims ranged damage bonus, ATK% and MATK% should all affect it, though commenters are then talking about some strange unique behaviours as to what bonuses should affect it.

And then there's this kRO guy (http://www.inven.co.kr/board/ro/1940/10719) who wrote and maintains a character simulator who says the formula should be updated entirely from the one currently in rAthena (and likely in use by Nova) into:
[([(ATK + MATK) * 0.7 *  enemy_VIT] - [(enemy_hardDEF + enemy_softDEF) / 2 + (enemy_hardMDEF + enemy_softMDEF) / 2]) / 10]

 

 

 

tldr; it's nuts.

But yeah, ATK% should affect it currently in Nova, and assuming we also have the MATK variance issue, ATK is better to stack than MATK for the time being until that gets fixed.

Share this post


Link to post
Share on other sites

Thanks for that write-up, I really appreciate that you did all this digging.

 

I wanted to check rathena sources myself at some point but was unsure since nova follows kro so closely. And yea AD is really one of those skills were the actual formula changes heavily depending on where you look..

I learned to trust @battlestats more than stuff like irowiki, because of those differences. The thing with AK card adding with All Class Bonus, makes it very likely that +%atk works imho.

Atk being worth more than Matk is something I heard from many different players, so it might really be a thing, but Cart Cannon alone is a good reason to chose Atk over Matk anyway :) 


4 

Share this post


Link to post
Share on other sites

Pendant of maelstrom gives atk% n mtk% thus it boost the A.D. dmg or only the atk%? If its only the atk% . It would be better to use sheriff badge sinces it gves 3% atk n 3% ranges and much cheaper.

Edited by GeneticEX

 

5.png

Share this post


Link to post
Share on other sites

The way I see it now, both +%atk and +%matk should boost AD, which is why PoM is pretty much the best damage-accessory for now.

However, I currently do use sheriff's left badge as my phen accessory when mvp hunting, it's a nice and cheap alternative and shouldn't do much worse than PoM. Another free alternative could be the Medal of Honor if you have BG badges flying around, which also brings some other neat bonuses at the cost of not having a card slot.


4 

Share this post


Link to post
Share on other sites

Bump, no rest for the wicked, I guess :P

 

So it turns out, against anything discussed before, that a full atk Excellion Suit does beat Flattery Robe: https://clips.twitch.tv/SlipperySuspiciousWalletGrammarKing

Is this just a fluke regarding high-/low-rolls or is matk really that much worse than atk? Now that Flattery Robe in the above clip is still missing a Porcellio for maximum output. Does anyone happen to have both flying around and is willing to test? I kinda find it hard to believe that +95atk from excellion beats +25atk/+150matk from flattery, which would make matk not just slightly worse (previously estimated at 1:1.5 atk/matk ratio), but pretty much abysmal...

 

Edit:

Alright, was able to borrow an imperfect excellion suit[Atk,Str,empty].

Average AD dmg vs. bradium golem with that was: 165,060

Average AD dmg vs. bradium golem with Flattery: 172,410

 

Now I think the missing Attack Supplement might only even it out, not beat it, but can't say for sure yet.

 

Edit2:

Okay, borrowed an atk/atk/str excellion suit, average AD dmg vs. bradium golem was: 168,322

Now with flattery (very similar to above), i had average AD dmg vs. bradium golem: 172,368

 

Not sure what was going on in the above clip, but Flattery[Porcellio] definitely beats the full-atk Excellion suit for AD.

Edited by Kruro

4 

Share this post


Link to post
Share on other sites

INT adds MATK equal to STR's ATK but... (not sure on exact formula)

Personally i believe Cast time being a factor id rather have INT


Lu7ky

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
Sign in to follow this  

×
×
  • Create New...