Damage Scaling in KOFXIII and potential issues

[nax]

I’d first like to thank Laban` from #SNKPlaymore for helping to record this.
[media=youtube]3zQG7Z5LFB8[/media]

As you can see here, the exact same combo is performed twice. The difference, however, is that both combos have different damages. Before I explain whats going on, I don’t want people to freak out about this yet. Chin has 5 drinks on just to show a bigger difference in the number and performing this isn’t exactly viable as adding to it will incur damage scaling and that type of damage won’t happen.

So what’s going on is that in the second recording, chin uses his neomax, and as soon as he can activated his EX DM which begins dealing damage precisely following the final damage of the neomax. This gives some insight into damage scaling. Now I’m not going to pretend to be some expert on code I’ve never seen, but due to an issue that has been brought to my attention with Kensou incuring damage scaling outside of a combo, there seems to be two “rules” regarding damage scaling:

1. Canceling via HD seems to directly incur damage scaling (This may be iffy. There’s been issues that seem to back this up, but then some other instances that make it seem unlikely)
2. If there is no canceling, then there is a set delay between two different instances of damage

An example of #2 was in the arcade 1.1, you were able to cancel for example close ;c into kensou’s punch DM. If you purposefully delayed the punch DM to come out without it combo’ing, but quick enough that the close c was canceled into the DM, then damage scaling occured as if it was a full combo. (Thank you Parapets)

So, anything that falls under either of those 2 rules incurs damage scaling.

This of course is all theory since I’m going off this (rather odd) instance of chin breaking damage scaling, but hopefully if any other instances like this exist, they can be found and reported to snkplaymore or atlus soon. That is my understanding of how damage scaling works and a problem with the way it works. This does not seem to be a bug so much as an oversight.

[rogueyoshi]

42 vs 43 hits… maybe the neomax has randomness to it

[nax]

there’s no randomness to the neomax, i didnt notice it before but he messed up a bit.

still would you be thinking one extra hit would do +20%?

try it yourself and get them to the same number of hits and it will still happen. the video is just a demonstration.

[kikimaru024]

So… a bit like Just-Frames in Tekken?

[blackgenma]

I don’t understand the given explanation. can anyone simplify it?

[nax]

basically two different instances of damage can avoid being scaled down if you don’t use any canceling and there’s very very little pause between the end of one move and the beginning of another.

chin makes it so he hits the first hit of his DM at the tail end of his neomax, creating a constant stream of damage that, because of no delay being between them, the game thinks is one big set of damage rather than separate instances of damage which it would normally lower incrementally.

[M.D]

Very strange. So late cancels/links are the way to go.

[Dr.Grammar]

Very strange. So late cancels/links are the way to go.

The glitch with the late cancels is that the canceled-into move does not combo, but it scales anyway. It’s not useful.

[phoenixnl]

What seems to happen is that no scaling applies when the opponent is still in the hitstop of a fireball, or if the fireball is still on the screen or something.

It would be interesting to see if this lack of scaling applies when you hit someone with an attack and a fireball at the same time. That should be pretty easy to check by getting the opponent to run towards the corner throwing a slow fireball after them catching up and trying to time the fireball and attack to hit simultaneously, or the attack like 2 or 3 frames after the fireball hitting, and after than like 15 frames later. If there’s a damage difference, it’s the hitstop that somehow makes scaling get ignored.

I doubt it’ll have any practical applications though.

[imfree]

I can envision a setup with Saiki where you hit them with the fireball super and then neomax after he returns to neutral but before the fireball runs out of hits. Then again, that’s like 4 bars for 600 damage or so. Probably not worth it.

[nax]

What seems to happen is that no scaling applies when the opponent is still in the hitstop of a fireball, or if the fireball is still on the screen or something.

It would be interesting to see if this lack of scaling applies when you hit someone with an attack and a fireball at the same time. That should be pretty easy to check by getting the opponent to run towards the corner throwing a slow fireball after them catching up and trying to time the fireball and attack to hit simultaneously, or the attack like 2 or 3 frames after the fireball hitting, and after than like 15 frames later. If there’s a damage difference, it’s the hitstop that somehow makes scaling get ignored.

I doubt it’ll have any practical applications though.

I’m HOPING that’s the case, since chin’s setup is rather unique. If anything, it may allow a small damage boost in SOME combos by neglecting damage scaling where possible. But, in any case, I wanted to bring this to everyone’s attention so if there IS a very practical and even abuse’able application that it gets found quickly and gotten rid of before the tournament seasons really start kicking off.

I’m still playing around with several characters and I believe there may be away to use this damage scaling knowledge to redo some of mature’s combos and make them more efficient. I’ll post any other findings I or my friends come up with for “benificial” (ie chin’s bug) or “harmful” (ie old Kensou’s bug) that either artificially increases or decreases the damage scaling in combos.

[phoenixnl]

Okay, I did my research. This bug can be exploited with Mature as well, and the more I think about it the more applications I can think of.

So what I tested with was Mature EX fireball into qcb,hcf+K.

Turns out, depending on when you activate the super, you can get up to 4 different damage values.

358 damage When you activate super after the three hits
369 damage When you activate super after the first two hits
383 damage When you activate super after the first hit
400 damage When you activate super before the fireball hits at all.

It becomes pretty obvious if you look at the video:

[media=youtube]jlQKHwf44hA[/media]

So let’s do the math. EX fireball is 60+60+60 Super is 0+220. This super scales as if its one hit, so we can just calculate as if its simply 220.

60+60*(1-0.051)+60(1-0.052)+(0+220)(1-0.05*3) = 358
First three hits are scaled, the two hits of super count as one hit

60+60*(1-0.051)+(60+0+220)(1-0.05*2) = 369
The first two hits are scaled normaly, the third hit is taken with the scaling of the super, and all three hits count as 1 hit.

60+(60+60+0+220)(1-0.051) = 383
First hit scales normally, the remaining hits all scale as one hit.

(60+60+60+0+220)(1-0.050) = 400.
All hits are taken as one hit, and thefore don’t scale at all.

---

So who could use this realistically?

Well, Leona’s charge b~f+P into super is a good example, the faster you confirm into v-slasher, the fewer hits of the fireball will continue to scale.

Can anyone else think of practical situations? it pretty much has to be done with multi-hitting projectiles into super.

[Mazinkaiser]

snip

Can anyone else think of practical situations? it pretty much has to be done with multi-hitting projectiles into super.

Good catch and nice explanation about this… i think the guys with slower multihit projectile, like terry and k’ for example. but i think it’s kinda of gimmick not reliable to use in a serious match because to abuse of this (outside of chin -.-,?) you have to do the super using your sixth sense o.O, if you wait to see if the fireball connect and hitconfirm into super there isn’t enough time to connect without scaling… i guess.

[phoenixnl]

Yeah exactly. Oh well, maybe for combo videos.