Debuff overwriting confirmed to be bugged in English version

[sanahtlig]

Update: This bug has been fixed in the English version as of 1/5/18.

Defense down debuff fails to correctly overwrite weaker debuff of the same type in Kamihime Project R
Wpmz, Dr. SlickFenix^, sanahtlig*

* Corresponding author, Principal investigator
^ No intellectual contribution, and mainly just distracted the others

This research was conducted by the Godtamers Union Research Division (https://discord.gg/7BjHCRG).

Introduction: Debuffs failed to overwrite correctly prior to 1/20/17 in the DMM version, resulting in stronger effects failing to apply as expected [1 (http://xn--hckqz0e9cygq471ahu9b.xn--wiki-4i9hs14f.com/index.php?%E3%83%90%E3%83%95%E3%83%BB%E3%83%87%E3% 83%90%E3%83%95%E5%8A%B9%E6%9E%9C#ca99a8ff)]. Until now, it was unconfirmed (or at least not widely known) whether the problem persisted in the English version, which was launched a month later.

Methods: Nergal's type A DEF down debuff (-10%) was applied, damage from several rounds of normal attacks was recorded, the debuff was overwritten with type A DEF down debuff Ambush (-20%), and several additional rounds of normal attacks were recorded. This was performed with and without a type B -20% ATK/DEF debuff from Amaterasu. The initial experiment was performed on the Fire Disaster (Normal), and the results were replicated on the Thunder Disaster (Normal), and Yatagarasu (Ultimate, boss only) encounters. Encounters were video recorded, damage numbers transcribed into a spreadsheet, and the normal attack damage for all characters before and after Ambush application were compared for a statistically significant difference using Student's t-test.

Results: Ambush failed to provide the expected 10-16% increase to normal attack damage in subsequent rounds predicted by Sanahtlig's Kamihime Project Toolbox (https://docs.google.com/spreadsheets/d/1PhfOAQk6EyK5l33ux_C4T-cexVIQp6D-_Hbd-OO1OmQ/edit#gid=848984711) Damage Calculator. Intriguingly, it appeared to increase normal attack damage in the SAME ROUND it was applied, but this bonus did not carry over to subsequent rounds, suggesting a strange system glitch rather than a design oversight.

Results spreadsheet with video links (https://docs.google.com/spreadsheets/d/1dcfA1PV3TtkUKL4lN5COStKhfT74sFoLtKP5V4AVFxo/edit?usp=sharing)


https://youtu.be/WGVsd1-PeeY

Conclusion: Weak debuffs should not be applied when a stronger debuff of the same type is available, or will soon be available. The weak debuff should be applied afterwards instead (resulting in a 'miss'), or saved in case the stronger debuff misses. This experiment only tests a specific set of conditions, but since it replicates a known issue in the DMM version, a general defect in debuff overwriting should be assumed. Mordred-users in particular should pay attention to the order of debuff application, as the type B ATK/DEF applied by Outrage is the weakest of its type.

Moreover, a user has reported (http://harem-battle.club/kamihime-project/3703-debuff-overwriting-confirmed-bugged-english-version-2.html#post83337) that Regen doesn't update either, implying that stronger buffs may also fail to replace weaker buffs.

[Wpmz]

This is Wpmz who did the verification. If anybody has any questions regarding the videos or spreadsheets feel free to contact me on discord at Wpmz#6239 or via the Godtamers discord.

[Rhenosa]

i hope it didnt happen to the buff mechanics. since buffs will overlap alot especially in raid

[sanahtlig]

i hope it didnt happen to the buff mechanics. since buffs will overlap alot especially in raid
Debuffs generally have stronger effects than buffs. And the potential for weak debuffs blocking stronger debuffs in Raids is much greater due to the large variety of same-type debuffs (that affect the entire Raid) and their ~10x longer duration.

[Shieun]

How do we know if a debuff is stronger/weaker?

Is it the +/++ symbol?

What about some character like Tyr who only has * powered up?

[Aidoru]

It varies but ++ will most likely be better than +.

I say it varies since the game doesn't specify the exact values. You might find skills with 2 debuffs with the same/similar effect description but one is -12% and the other is -15%. You can find the values listed on the JP wiki.

http://xn--hckqz0e9cygq471ahu9b.xn--wiki-4i9hs14f.com/index.php?%E3%83%90%E3%83%95%E3%83%BB%E3%83%87%E3% 83%90%E3%83%95%E5%8A%B9%E6%9E%9C

[Shieun]

took me a while to understand google translate result... but I suppose I kinda get it...

Also I presume the debuff value stacks by straight addition (e.g. 10%+15%+15% = 40%)?

[sanahtlig]

Debuffs generally fall into 5 types: A, B, C, eidolon/Burst effect, stacking (not many of these). Debuffs of different classes stack additively, while debuffs of the same class are supposed to apply the strongest one. Except we verified that this isn't the case; a weaker debuff will remain and be renewed if overwritten with a stronger one. The stronger debuff will be active for the turn that it was applied, but in the big scheme of things that doesn't really help.

[Aidoru]

took me a while to understand google translate result... but I suppose I kinda get it...

Also I presume the debuff value stacks by straight addition (e.g. 10%+15%+15% = 40%)?

Yes and caps at 50%

[SlickFenix]

took me a while to understand google translate result... but I suppose I kinda get it...

Also I presume the debuff value stacks by straight addition (e.g. 10%+15%+15% = 40%)?

Yes they do stack by straight Addition, however not every debuff stacks. They have to be in different Classes or "Frames" as Google Translate says from the DMM Wiki.

Below is the link to the debuffs, just remove the enter between http and the actual address. Only way I can post links.

http
://xn--hckqz0e9cygq471ahu9b.xn--wiki-4i9hs14f.com/index.php?%E3%83%90%E3%83%95%E3%83%BB%E3%83%80%E3% 83%A1%E3%83%BC%E3%82%B8%E3%82%A2%E3%83%83%E3%83%97 %E7%B3%BB/%E3%83%86%E3%83%BC%E3%83%96%E3%83%AB#r984489a

[Wpmz]

took me a while to understand google translate result... but I suppose I kinda get it...

Also I presume the debuff value stacks by straight addition (e.g. 10%+15%+15% = 40%)?

Okay so debuffs have what are called "frames" (it's kind of a bad term, they're more like categories). Nergal's debuff and ambush are examples of A frame debuffs. You can only have 1 A frame debuff applied at a time. Sniper shot is an example of a B frame debuff. You can have 1 A frame and 1 B frame debuff applied simultaneously, and indeed they do stack by straight addition (say ambush - A frame 20% and sniper - B frame 20% would indeed be 20% + 20% = 40%).

However what we're saying here is not about that. It is about overwriting debuffs. Since you cannot have 2 A frame debuffs at the same time, if you apply a weak one and then you apply a stronger one, the stronger one should overwrite the weaker one. However that is not what is happening. The stronger debuff overwrites the weaker debuff but only for 1 turn. The next turn it goes back to the weaker debuff.

[sanahtlig]

Yes and caps at 50%
We have unconfirmed reports that this is not the case. Rather, debuffs appear to be currently uncapped, which will lead to a game-breaking bug when a certain kamihime is released in about a month, allowing a division by 0 to occur in the game formula (and thus infinite damage).

[Aidoru]

What are the reports for that based on? Tests on atk down? I don't think there's enough def down debuffs to reach 50% yet.

[sanahtlig]

What are the reports for that based on? Tests on atk down? I don't think there's enough def down debuffs to reach 50% yet.
Yes, ATK down. Sniper Shot, Jeanne's skill, type C from Sol or that wind R, Jorm or Yatagarasu. That's -65-70% ATK down.

[SlickFenix]

What are the reports for that based on? Tests on atk down? I don't think there's enough def down debuffs to reach 50% yet.

The Distracting one is testing it......

[jazz154]

So... it will take 4-5 months before it gets fixed, when this gets to the version that fixed it, right? Because I doubt that they will fix this themselves.

[sanahtlig]

So... it will take 4-5 months before it gets fixed, when this gets to the version that fixed it, right? Because I doubt that they will fix this themselves.
That seems probable. The bug was fixed prior to the English release, so if Nutaku were at all motivated to fix bugs in earlier versions of the game engine, they would've fixed it prior to release.

[Unregistered]

I don't think you can overwrite buffs at all, unless there is a different phase for HoTs and DoTs that is separate from the attack phase?

If I use Orlean Call (from Joan) and then I use Divine Solar Shield+ (from Amaterasu), my regeneration for that turn stays at 250 instead of increasing to 400.

[Wpmz]

I don't think you can overwrite buffs at all, unless there is a different phase for HoTs and DoTs that is separate from the attack phase?

If I use Orlean Call (from Joan) and then I use Divine Solar Shield+ (from Amaterasu), my regeneration for that turn stays at 250 instead of increasing to 400.

This thread is about debuff overwriting. As for overwriting buffs, regeneration is a weird case, I'm not really sure how that works. However you can overwrite some buffs. For example if you use fenrir active (that grants increased ability damage) and then oceanus ability damage increase spell, it will overwrite.

[sanahtlig]

It looks like the debuff overwriting bug has been fixed. Sniper Shot now properly overwrites Outrage and persistently increases damage, and the second Outrage after Sniper Shot now misses as it should. It's unclear how this affects other anomalies such as Regen overwriting.

The most immediate effects of this are that Auto-skill with Mordred and Sniper Shot will work better, and Raids won't be permanently gimped by new players using weak debuffs early in the fight and blocking veterans from applying the stronger versions. That'll make Ragnarok Raids much easier.

I've updated the Buff and Debuff Stacking (https://docs.google.com/document/d/19K_dU3rCief_ajPGS9YBuVdxXBYFyCGGXAxYYyK0jzg/edit#heading=h.mpzxayk350pi) section of my guide accordingly.

A video of my 2nd test will be made available here (https://youtu.be/Aay_cCFT59E) shortly.

[Slashley]

Was this fixed in today's maintenance? Since I'm pretty sure that a few days ago, whenever I used Sniper Shot after Outrage, I was still not doing the -40% Def amount of damage.

[Kureru]

Still not fixed.

Actually I went and tested again just to confirm and something strange happened. The first attack after the good buffs did the damage I'd expect, but then the second was back down to Outrage levels of damage. It's a difference of ~8000 so it's not just variance, not sure what happened there.

[Slashley]

Still not fixed.

Actually I went and tested again just to confirm and something strange happened. The first attack after the good buffs did the damage I'd expect, but then the second was back down to Outrage levels of damage. It's a difference of ~8000 so it's not just variance, not sure what happened there.Well, as the first post says:
-- Intriguingly, it appeared to increase normal attack damage in the SAME ROUND it was applied, but this bonus did not carry over to subsequent rounds, suggesting a strange system glitch rather than a design oversight.--

[Kureru]

Bleh, that's what I get for not reading I guess. The second attack was fatal so I couldn't test any further, I'll try again later but it's looking like it's unfixed

[sanahtlig]

Still not fixed.

Actually I went and tested again just to confirm and something strange happened. The first attack after the good buffs did the damage I'd expect, but then the second was back down to Outrage levels of damage. It's a difference of ~8000 so it's not just variance, not sure what happened there.
That's odd. What did you test on? I tested on the Expert Raid boss. It's... possible that the new event bosses operate on different system rules than other content. That would be truly erratic.

Also, I performed my confirmation testing after receiving 2 other reports about this. So it's repeatable.

[Kureru]

I tested on light disaster. I went and tried again and got full damage, maybe I just had a brainfart and saw 28 as 20 when I tested.

[sanahtlig]

I have a theory about what's going on with heals causing damage during Raid battles. It could be that only 1 heal effect is allowed per turn, and that subsequent heals are actually OVERWRITING the effect of prior heals. Thus when you get a heal from another party's Andromeda, then heal yourself, if your heal is less than that of Andromeda's then it actually reduces your party's HP. Unconfirmed, but it explains what I just observed with my team.

[Kureru]

I thought it just displayed a health gain without actually healing, meaning it disappeared the moment your health was updated again