Data Zelda Hitbox Visualization

[Furil]

Smash 4 hitboxes are interpolated, so imagine a line between a hitbox's current position and it's last. It hits that area too.
Example: http://i.imgur.com/wEBKNpI.gif

There are potentially bugs in any of these, so let me know if you notice anything particuarly off and I'll check. There have been instances of weird hitboxes that ended up being a bug.

Spoiler: JAB 1

Spoiler: DASH ATTACK

Spoiler: FTILT

Spoiler: FTILT HIGH

Spoiler: FTILT LOW

Spoiler: DTILT

Spoiler: UTILT

Spoiler: FSMASH

Spoiler: DSMASH

Spoiler: USMASH

Spoiler: NAIR

Spoiler: FAIR

Spoiler: BAIR

Spoiler: DAIR

Spoiler: UAIR

Spoiler: Nayru's Love

Spoiler: Farore's Wind (Hit 1)

Spoiler: Farore's Wind (Hit 2)

Spoiler: Farore's Wind (Hit 1, Aerial)

Spoiler: Farore's Wind (Hit 2, Aerial)

Edit:
Grabs may be off, so consider them unreliable for now.

Spoiler: Standing Grab

Spoiler: Dash Grab

Mini-FAQ:
Q: How did you make these?
A: I made a parser for animcmd commands and applied it to the model and animation.

Q: Can you do hurtboxes?
A: Not yet

Q: These don't look like the quality of the Wii U models, why?
A: These are 3DS models (and data). There isn't a difference in frame data between Wii U and 3DS.

Q: This attack looks wrong, is this actually correct?
A: Let me know which one and I'll check. It's possible for a bug or two to mess up stuff.

Q: The image isn't playing.
A: There are play, reverse, and frame advance buttons under the images. They may be hard to see.

Q: Can I use these for my guide?
A: Yes.

Thank you

 

[Alacion]

Been waiting for this. Awesome work and thank you!

 

[MOI-ARI]

Thanks! Happy you're still at it!~
ZeldaBoards is thankful

 

[kew]

Thank you so much for this!

 

[PGH_Chrispy]

Important things I've noticed so far that will affect my gameplay:

Utilt hits around the same height on the front and back, do not use on low profile opponents.

Ftilt high actually seems like it could have some serious use.

Dsmash will hit slightly below ledge, but only if spaced properly. On a related note, dsmash has a blindspot under Zelda that shouldn't affect too many interactions.

Fsmash should not be used to try and catch someone in the air - it's too low to the ground. Also has a blindspot in front of zelda.

Usmash hits on both sides all the time (silly me, at least I can now do pivot JC Usmashes relatively fast)

Zelda's body is a hitbox for the last hit of nair - wonder whether it takes the damage of the back or front hit.

Sweetspot position data!

Nayru's Love has surprising horizontal reach.

Grounded Farore's Wind 1 will not hit aerial opponents for crap.


The Gorls thank you will all our buff prayers!

 

[JosePollo]

I see you, Zelda, with them hitboxes



Would it be possible for you to get the hitboxes on the various charge stages of Phantom Slash?

 

[KlicKlac]

Would it be possible for you to get the hitboxes on the various charge stages of Phantom Slash?

I'd love to see this too, if possible.

 

[Lavani]

Zelda's body is a hitbox for the last hit of nair - wonder whether it takes the damage of the back or front hit.

The last hit works differently from the rest of the move; all hitboxes do 5%, with the arm hitboxes having more knockback than the body hitboxes. The arm hitboxes have priority if the opponent overlaps both.

 

[SmashXX]

Grounded Farore's hitbox is a lot smaller than I thought. It'd be nice if it covered more above her.

Thanks for posting

 

[BJN39]

YAAAS. Thank you so much Furil ! These are great.

Some of these moves' hitboxes actually look surpringly solid after being able to see them visualized. Others...actually look less amazing than they feel. Weird.

I expected NAir to look good, but I mean it looks really good. I love how the devs forgot to increase the final hit outer hitboxes sizes along with the loop hits in 1.0.8.

Dsmash's range (especially the back) looks particularly surprising.

One question: Jab's final hit has 3 hitboxes. Is the middle hitbox just fully eclipsed by the others and thus unseeable?

 

[Furil]

One question: Jab's final hit has 3 hitboxes. Is the middle hitbox just fully eclipsed by the others and thus unseeable?

Mostly, you can actually see the second one in between them, but just barely.

 

[BJN39]

I think I'll sticky this thread~ Knowing hitboxes is really important for knowing the character, and no doubt people will want to be able to find this thread often. So now let's do it without having to sift through the board's general threads.

Also Furil would it be possible to get grab hitboxes as well?

 

[Lavince]

Also I'm pretty sure this was @Zolda 's request, but would it be possible to get the hitbox data for ledge canceled farore's wind? I hear it's supposedly really large and doesnt have a sourspot but I'm curious to see just how large it is lol

 

[BJN39]

Also I'm pretty sure this was @Zolda 's request, but would it be possible to get the hitbox data for ledge canceled farore's wind? I hear it's supposedly really large and doesnt have a sourspot but I'm curious to see just how large it is lol

Zolda is misinformed. There is no unique ledge cancelled variation.

LC FW uses the grounded reappear hitbox. The sweet-spot can clip ledge hangers.

 

[Furil]

There was a bug with F-tilt, I'm fixing them now.
edit: fixed

Also Furil would it be possible to get grab hitboxes as well?

Eh, maybe. Grabs seem to work differently than regular hitboxes, as some have much more range than the data implies. It may only be a pivot grab thing, though, so I'll get standing and dash for now.

And if someone could verify their accuracy, that would be nice.

 

[BJN39]

Eh, maybe. Grabs seem to work differently than regular hitboxes, as some have much more range than the data implies. It may only be a pivot grab thing, though, so I'll get standing and dash for now.

And if someone could verify their accuracy, that would be nice.

That surprises me. More than than the data implies? I mean, pivots grabs are just longer, that's how Sakurai intentionally designed it this time. Do you just mean the software doesn't portray the data correctly, or that there's something about the data in particular?

 

[Lavani]

That surprises me. More than than the data implies? I mean, pivots grabs are just longer, that's how Sakurai intentionally designed it this time. Do you just mean the software doesn't portray the data correctly, or that there's something about the data in particular?

Grab data is just...off, somehow. If you remember that old research thread we had a couple months after the game's release, I brought this up in that; Wii Fit and Marth have the same grabbox size and horizontal reach with no bone attachment, but Marth has a ton more grab range. The second grab parameter should be the bone (Toon Link has values set there for the hookshot chain/handle I believe) so I dunno what causes the inaccuracy.

 

[BJN39]

I mean, you'd think it would have to do with where the character's central axis is at the time of the grab right? (Because coordinates w/bone 0 start from a character's xyz0) Marth hunches forward, while Wii Fit's body remains in the exact same horizontal placement for the whole grab. It would be odd if Furil's hitbox program didn't catch this though. Does Marth's central axis temporarily shift forward when he grabs? Are there any other odd instances of this happening?

 

[Lavani]

Marth hunches forward, while Wii Fit's body remains in the exact same horizontal placement for the whole grab.

I...totally didn't even think of this, lol. Marth does move his center forward significantly, that would explain that instance.

It doesn't line up for everything though. Bowser's pivot grab even with hitbubbles drawn doesn't remotely match the actual range:

But...yeah, maybe it is just a pivot grab thing then.

 

[BJN39]

I...totally didn't even think of this, lol. Marth does move his center forward significantly, that would explain that instance.

It doesn't line up for everything though. Bowser's pivot grab even with hitbubbles drawn doesn't remotely match the actual range:

But...yeah, maybe it is just a pivot grab thing then.

I guess then, does pivot grab have any data params/uniqueness that the other grabs don't have? That's one possibility.

Like, maybe they secretly param to a different location relative to the character.

Ooh, maybe they are relative to the 'expected' starting position. So like, at a full run, where you'll slide while pivot grabbing, they would somehow lineup with a location offset to the central axis to counterbalance the slide and provide appropriate range. (But not that they change range every usage, but just an 'expected range' offset that is used for every usage) IDK, seems a bit confusing but I'm just tossing out ideas.

 

[buzzard]

Thanks a lot!

 

[Kider]

The data is beautiful, thank you!

 

[BJN39]

Hey Furil I had a question about Ftilt. Would your program be able to tell exactly what the coordinates of Ftilt's hitboxes would be relative to Zelda's xyz=0 if they were placed exactly where they are (when they are fully extended in front of her) but not attached to a bone?

 

[Furil]

I think I understand the grab thing, at least the wii fit marth part.
They do have the same range, it's just misleading when measuring this way, since wii fit trainer's base node can be much closer to the edge than Marth here.
(Note Marth is probably a bit too large here, it's an estimate)

With that said, I think standing and dash grab data is actually correct.

Pivot grabs, though... I have a theory.
Here is what the data would currently imply..

Now here's if we make the points relative to the pivot.
It lines up! It would help to know a few more pivot grab ranges (such as Duck Hunts) so this theory can be better tested.

 

[RidiculousFalcon]

I love Nayru's horizontal disjoint. I've used it to shut down Bayo's grounded side B spam in neutral, but I didn't know it would hit from that far out.

 

[KlicKlac]

Wait a minute, what determines whether farore's wind reappear uses the grounder or aerial hitbox?

I am confused. If you start farore's wind on the ground and reappear in the air, what reappear hitbox does it use (grounded or aerial)? Same if you start farore's wind in the air and reappear on the ground.

Also, when you ledge cancel farore's wind, do you technically reappear on the ground or in the air?

 

[BJN39]

Wait a minute, what determines whether farore's wind reappear uses the grounder or aerial hitbox?

I am confused. If you start farore's wind on the ground and reappear in the air, what reappear hitbox does it use (grounded or aerial)? Same if you start farore's wind in the air and reappear on the ground.

Also, when you ledge cancel farore's wind, do you technically reappear on the ground or in the air?

NO MATTER WHERE YOU DISAPPEAR:

Aerial reappearance uses aerial version hitbox

Ground reappearance uses ground version hitbox

A successful ledge cancel involves a grounded reappear being cancelled on the edge >>> grounded reappear hitbox

If you got an aerial reappear hitbox trying to LC, you're also getting a SD lol

It doesn't matter where you disappeared.

I'm being perfectly honest @ any Zelda player: if you didn't know this, please let me know. Ask questions. I just assume that people know these things, but if many people don't, I can start compiling these kinds of things in the Fairy Fountain.

 

[KlicKlac]

NO MATTER WHERE YOU DISAPPEAR:

Aerial reappearance uses aerial version hitbox

Ground reappearance uses ground version hitbox

A successful ledge cancel involves a grounded reappear being cancelled on the edge >>> grounded reappear hitbox

If you got an aerial reappear hitbox trying to LC, you're also getting a SD lol

It doesn't matter where you disappeared.

I'm being perfectly honest @ any Zelda player: if you didn't know this, please let me know. Ask questions. I just assume that people know these things, but if many people don't, I can start compiling these kinds of things in the Fairy Fountain.

Thanks for the quick response. I had just assumed this, but never verified it with anyone to double check.

So basically, you don't get a bigger hitbox if you do a short hop right before you teleport onto the ground lol that would have been neat.

 

[Deleted member]

Thanks for this!

Well at least now we can see a few of her hitbox issues. Some of the dumbest things noticeable:

- F-tilt's blindspot because she leans so far forward that her hitbox-less body passes through the opponent. She really needs a decent sized hitbox on her body. The same applies to her F-smash.

- Up-Tilt's hitbox starts slightly later than when her arm first begins the arcing motion and ends a LOT sooner than when she finally brings it all the way down to the end of the arc. This is why she can't hit small/crouching characters. If it just matched the animation better, the move would be a lot more useful.

- Farore's Wind's grounded first hitbox only covers the bottom half of her body...wtf?

- Nair's final hit actually has LESS horizontal range than the multi-hits because the hitboxes at her hands shrink. Again, wtf? It would also be really nice if the multi-hits had those two nicely sized additional hitboxes (one on her head and the other extending down her body) that the final hit has. It would make it easier to hit grounded opponents with the move to start combos.

Please dev team, fix!

 

[BJN39]

Thanks for this!

Well at least now we can see a few of her hitbox issues. Some of the dumbest things noticeable:

-F-tilt's blindspot because she leans so far forward that her hitbox-less body passes through the opponent. She really needs a decent sized hitbox on her body. The same applies to her F-smash.

-Up-Tilt's hitbox starts slightly later than when her arm first begins the arcing motion and ends a LOT sooner than when she finally brings it all the way down to the end of the arc. This is why she can't hit small/crouching characters. If it just matched the animation better, the move would be a lot more useful.

-Farore's Wind's grounded first hitbox only covers the bottom half of her body...wtf?

-Nair's final hit actually has LESS horizontal range than the multi-hits because the hitboxes at her hands are smaller. Again, wtf? It would also be really nice if the multi-hits had those two nicely sized additional hitboxes (one on her head and the other extending down her body) that the final hit has. It would make it easier to hit grounded opponents with the move to start combos.

Please dev team, fix!

Fsmash actually doesn't have blindspot issues. Her body + hitboxes interact differently, and they last much longer, so it never point blank whiffs thank GOODNESS.

Ftilt is potentially such an easy fix that IDK why they still haven't figured it out. The blindspot is WIDER THAN MARIO.

I did some data thinking and concluded that if the innermost hitbox had a stretch coordinate somewhere farther backwards, it should cover the issue. (It currently is at z=-1 relative to the bone connection. If it had a stretch to like, z=-3.5 or z=-4, that could be enough.) Hey, Furil I've actually been meaning to ask if you could make a GIF/image of what this would look like, if you have time.

NAir does have that weirdness yeah. They kinda...forgot to resize it along with the multihit hitboxes in 1.0.8, lol. Surprisingly, it has NO negative effect on its ability to connect.

And ofc NAir got almost 2 whole units more range in every direction~

Ground FW is dumb agreed. I knew it was built differently than just a sphere, but until this I never realized how LOW it is. It's like it was designed specifically so Dtilt>Elevator would be hard/impossible smh. On the flip side, it has 2 whole units of below ground range...I wonder if it could hit ledge hangers....

 

[Lord Renning]

If they put the last hit box of upsmash at the start to pop them into it, the move would be so much easier to connect with.

 

[Rickster]

My favorite thing about Usmash is the fact that the final hit's hitbox isn't even close to her hand.

We're talking Marcina Usmash levels of dumbness here.

 

[BJN39]

Furil random question about FW2 aerial. Does the GIF show the entire frame duration of the aerial reappear, (Up to where special fall begins) or does it cut off early at all?

 

[ZombieBran]

I always knew Din's Fire didn't even have a hitbox at all

 

[KlicKlac]

I always knew Din's Fire didn't even have a hitbox at all

Nothing gets me madder then when I successfully get a read with din's fire and hit someone but they just happened to be throwing out a hitbox and it beats out din's fire.

 

[ZombieBran]

Nothing gets me madder then when I successfully get a read with din's fire and hit someone but they just happened to be throwing out a hitbox and it beats out din's fire.

Even worse is somehow successfully landing Din's but then you remember Sakurai nerfed the worst move in the game so that it's almost all a sourspot and so you do approximately 1.5% of damage.

 

[Lord Renning]

Whenever I get a Din's fire kill I feel like the opponent died to a glitch.

 

[BJN39]

It astounds me how much people respect the move in smash 4 even today.

It's like they've never learned the fact that its hitbox is less than half the size of the visual.

I see people airdodge it when even Brawl Din's might not have hit. Best part is that Zelda can't punish that If Zelda doesn't directly aim for you, you don't even have to airdodge

Better yet just throw out a sex kick and then you have din's immunity and a hitbox too.

 

[Rickster]

I like to steal items from people (metal blade, fruit, etc) and camp with Din's and Phantom.

It works for some reason lol
WHY

 

[MOI-ARI]

One point of reference (Width-wise i guess ) for me is that it actually Strong-hits three wiifit trainers standing side by side. Im afraid the idle pose they all were doing could have had to do with it, especially for the far left Wiifit, but even then it helps with my approximation.