[Work In Progress] Step Dash Distance Rankings

[Radical Larry]

So we know about the new tech that has been going around, called "Step Dash", but the thing that people, especially that of My Smash Corner, has left out is always the distance. However, I have a way that I propose that everyone can try finding out the distances of all the characters, very effectively actually, and this is where I want help from labbers to compile the list of the distances.

So Step Dashing is a new tech that people have been discussing about lately, but if you aren't someone who's seen it yet, here is the video below, courtesy of My Smash Corner:

How To TEST Distances
It's actually very easy to test how far each character will slide. Simply pick Omega Suzaku Castle and perform the input from the edge of the stage, side doesn't matter, but always keep to one side for every character, and left is preferred.

How To RANK Distances
On Omega Suzaku Castle, there are a total of 21 different planks that are all symmetrical to one another, leading to the perfect testing of just how far a Step Dash will go. Ranking will be simple, as told below:

-If a character passes a plank or goes in between planks, it counts as a 1.
-If a character is still above a plank and in the direct middle or right of the plank, it counts as 1.
-If a character is still above a plank and just before the direct middle of the plank, it counts as 0.5

Character Rankings (Work In Progress)
Below will be all of the character rankings for the Step Dash Distances when all have come out:

Ranking|Character|Distance
1||7
2-3||6
4-6||5
7-13||4
14-26||3
27-36||2
37-41||1
42-44||0.5
Undetermined||Undetermined

Characters beside rankings are listed by alphabetical order.

Pixel Method
So Smash Highlights has made a video recently about Step Dashes and, while he doesn't have it completely, he has made a secondary methodology on how to perform Step Dashes by testing out the pixels. Not only that, but he also solved why Step Dashes go various distances; it's due to frame perfection and frame 7 is, mroe likely than not, the best frame to perform Step Dashes. I'm going to include a new table just for the pixel method of Step Dashing, just to provide a secondary method.

This is utilizing his maximum distances on Step Dashes, and is still an early rough draft, like in his video.

Ranking|Character|Maximum Distance (Pixels)
1||343
2||193
3||167
4||166
5||136
6||135
7||134
8||127
9||123
10-11||109

Credits
Credits for people who help out with the rankings will go down below:
T TheGameszocker
Chaos from the Ryu Discord on Smashcords
PK Bash
QualityQ
MattPerren / Smash Highlights

Let's dig up what we can, guys. Any help will count, as I'm even getting the discords to aid here.

 

[TheGameszocker]

Here is what I got:
Mario: 1.5
Green Mario: 3
Peach: 2.5
Bowser: 0.5
Yoshi: 2
Rosalind: 1.5
Bowser Jr.: 3.5
Wario: 1
Donkey Kong: 1.5
Diddy Kong: 2
Mr. Game&Watch: 0.5
Little Mac: 0.5
Edit: I've realised that I ****ed up so the data is probably wrong

 

[TheGameszocker]

OK second try:
Mario: 4.5
Luigi: 6
Peach: 2
Bowser: 1
Yoshi: 4
Rosalina: 3
Bowser Jr.: 2.5
Wario: 2
Donkey Kong: 3.5
Diddy Kong: 3.5
Mr. Game&Watch: 1
Little Mac: 2.5
Link: 3
Zelda: 2
Shiek: 4.5
Ganondorf: 1.5
Toon Link: 3
Metroid: 3
Zss: 1.5 (This one surprised me)
Pit: 4
Palutena: 1
Marth: 3
Ike: 1.5
Robin: 0.5 (Extremely hard to pull of or impossible.)
Duck hunt: 1.5
Kirby: 2.5
DDD: 0.5
Meta Knight: 6
Fox: 3
Falco: 3.5
Pikachu: 2.5
i ran out of time here. Also I'm not very good at this so feel free to correct me.

 

[Radical Larry]

Well this could be decent information. If there is more people who can confirm or deny it, then most certainly I'll update the list. For now, let me update the box.

 

[Game7a1]

I've been trying to measure Puff's step dash distance. I want to say 1 plank, but I might be miscounting and it may be 1.5 Planks (it's not 2 planks, but it's also not half a plank as well).
Could someone else test and confirm her step dash distance?

 

[#HBC | Red Ryu]

What determines the distance is something I am interested in.

Is it Dash speed or something of the sort?

 

[Fenny]

What determines the distance is something I am interested in.

Is it Dash speed or something of the sort?

I'm guessing he's using the planks on Omega Pirate Ship as a measuring unit?

 

[Horseketchup]

This is a great idea for a thread, though I think it'd also be useful to test safer tilts/jabs for each character too since it's possible to have a good empty step dash but also not have good moves that continue to carry your momentum. All I gotta say is that I really really hope Mewtwo doesn't have a good step dash into dtilt, and I hope that Diddy's dtilt doesn't carry momentum out of his either.

If even third of the cast gets solid use out of this though it could be a game changer. I'm curious to see the extent that this can continue to be pushed though. I'm also curious if you're able to pull off a 1 frame turnaround while still sliding, sort of like a perfect pivot version of this. I'm not really in a spot where I have access to the game right now though, I can't wait to try this out!!

 

[Ffamran]

What determines the distance is something I am interested in.

Is it Dash speed or something of the sort?

If it was dash speed, then it would have been Fox (2.4), Little Mac (2.05), Falco (1.9), Samus (1.86), Cloud, Corrin, Lucario, and Pikachu (1.8), Meta Knight (1.75), and a bunch of other characters down the line of 1.7, 1.6, etc. dash speeds, instead of Luigi (1.28) and Meta Knight being at the top of the list. Answer I was trying to figure out in my ramblings.

Spoiler: Ponderings of a madman

Anyway, I'm curious, outside of why this works in the first place, as to why Meta Knight has the longest or one of the longest "step dashes". Nothing about his walk specs seem peculiar, slightly above initial walk velocity, 0.16, regular 0.1 walk acceleration, etc. It probably isn't even his traction, 0.055, which is actually above-average since Mario's is 0.045, but that could have a factor since as people saw in the video, Falco and Fox barely moved anywhere compared to Meta Knight (and Bayonetta). It looks like Falco gets equal or less distance from using "step dash" compared to his normal dash... At the same time, in the same section where he showed Falco's "step dash", Link was shown before him and traveled even more than Falco and Link only has a 0.01 difference in traction to Falco at 0.064.

One thing I might factor in, but I'm hesitant on considering is dash specs since it doesn't seem like dashes are part of "step dash". "Step dash" seems to only be a two-step walk and a crouch. You never dash for it which... once again, I kind of wished it wasn't called "step dash" because it could get confusing like how dash is lumped together with running when they're separate things and repeated dashes are called foxtrots when... whatever. It could have been called "step slide" or something. Issues with naming aside, if dash specs were a factor for whatever reason, then two things could be considered: dash speed itself and dash length. For Meta Knight, his peculiarities with dash is his sixth-highest dash speed, 1.75, under Lucario and Pikachu (1.8), Samus (1.86), Falco (1.9), Little Mac (2.05), and Fox (2.4). Under him would be Captain Falcon, Diddy Kong, Sheik, and ZSS all with 1.7 dash speeds. That wouldn't entirely explain it since if dash speed was a major factor, then everyone around and above him would have a really good "step dash", especially Fox with his stupid 2.4 dash speed. An aside: caught an error to my Smash 4 movement value compilation: I wrote down 1.3 for Zelda's dash speed instead of 1.6. Whoops.

The other thing would be his dash length not dash total frames -- once again, confusing names. Dash total frames we have data on:

Spoiler: Stuff on dash total frames

Most characters have dashes that have 11 total frames. Meta Knight's is 13 which itself isn't limited to him as there's a group of other characters with 13 total frame dashes. The only real "oddballs" are Corrin and Sheik (8), Mii Gunner, Mr. G&W, and Samus (9), Falco (12), and Yoshi (15). There's a majority of 11 total frame dashes, groups of 13, 14, 16, and most of the Fire Emblem characters minus Corrin and Ike in the 17 total frame dash group. I doubt anyone's gone through their animations to check specifically what they do for their dashes, but a 17 frame dash isn't bad as Street Fighter V's Nash can attest to; Nash's dash essentially has 8 frames of startup and moves during its remaining 11 frames.

Spoiler: Nash's dash

There might be characters who have dashes like this, but it might not make a difference since one, Smash and Street Fighter are different games, and two, SFV's dash total frames are much higher than in Smash's where the lowest is 8 frames and the highest is 17 while SFV's lowest is 15, Chun-Li and Rashid's and the highest being 25 frames, Zangief's. Note: unknown about the DLC characters. Also, Rashid's the only character to have a normal run in SFV -- Ken can run, but it's activated through V-trigger. Anyway, I digress.

Now, dash length... I don't think we have data on. As far as I know, nobody's gone out and plotted out how far each character's dash goes. It probably not as simple as looking at all the dash values and saying, "Fox has the highest since his dash speed is 2.4." Fox does have a long dash, but average total frames for his dash. Little Mac's is also good, but I'm no sure, but then you go to Falco and Samus and theirs goes almost nowhere compared to Fox and Little Mac's. Then you look at Marth and Roy's and theirs travel quite far. Maybe each character has a value for how far their dashes goes. I don't know.

The last possible thing I can think of is just the difference in how characters move during their walk. Did you notice that when Meta Knight starts walking, he slides a bit? If you take someone like Falco and Fox, they don't slide during their first few steps. Falco does this weird tip-toe thing while Fox power walks. Or someone like Marth where there's this noticeable delay before he begins moving which is probably his walk acceleration. That being said, if this where the case, then Roy should have an insane "step dash" since he slides far during his first few steps of walking. Then there's the whole pivot-walking thing which if you do it with Falco, he does slide around a bit and it's been a thing for Falco since Melee. So, sliding during a walk...

Basically, I have no idea why "step dashes" work.

I mean, you could look at everyone's movement values and try to figure it out or maybe it has to do with walk animations or something. I don't know. Here's most of them: http://smashboards.com/threads/smash-4-running-walking-speed-rankings.371564/page-2#post-18647333. Here's the rest that needs to be completed: https://docs.google.com/spreadsheets/d/13bO7oaM4HLH0xgq3CmGiRTUF2Ftm_00A6u-g7ow220g/pubhtml#. I'm lazy.

 

[AyyLmaoJK]

Thought I'd throw this here: https://twitter.com/MySmashCorner/status/764236681507733504

Doesn't determine lengths, but the results should be similar to this.

 

[Radical Larry]

The stage that is being used to measure this isn't Omega Pirate Ship, it's:
Omega Suzaku Castle

It's the most visibly even Omega stage in the game, so the measurements should be relatively easy.

Thought I'd throw this here: https://twitter.com/MySmashCorner/status/764236681507733504

Doesn't determine lengths, but the results should be similar to this.

Except that's a fallacious list meant to throw people off, as there's evidence that this is primarily wrong. Most notably with Mac and Ganondorf, who can make some distance out of it.

And as a reminder to everyone, I redid how you measure the distances! If it's just before the midway point of the plank, that's a half; if it's in direct middle or just after it, it counts as one!

 

[Ambrose]

Thought I'd throw this here: https://twitter.com/MySmashCorner/status/764236681507733504

Doesn't determine lengths, but the results should be similar to this.

Ok, just want to clarify. I was preoccupied, so I only glanced at this list and saw some characters I tested were in relatively the right sections, which is why I said some what accurate. If you guys are planning to make this thread the official list, best not to follow any other lists.

It's going to take many trial and errors for this list to be as accurate as possible. To achieve an optimal step dash distance, you need to cancel the first step at a specific frame with a second step, then cancel the second step at a specific frame with a crouch/action. Not entirely sure about this, but I suspect the speed of the steps also plays a factor. If this is true, then we are dealing with two variables. Would be best if whoever is testing can upload a clip of the longest distance they get with the character they're testing with too.

 

[Radical Larry]

Actually, that'd work quite well.

 

[Fenny]

It's going to take many trial and errors for this list to be as accurate as possible. To achieve an optimal step dash distance, you need to cancel the first step at a specific frame with a second step, then cancel the second step at a specific frame with a crouch/action.

Aaaaaahhhhhhhh.

Should be able to do it more consistently now. Was trying to do it off of one step before.

 

[Radical Larry]

Does anyone else have their own distances to share? If it's at all possible, please record all of the distances in a compilation on YouTube (that is, if My Smash Corner or Smash Highlights don't help first).

 

[PK Bash]

A slightly easier method of getting the step dash is to get the control stick in the lower corner so the character crouches, then rock it to 90 degrees in whatever direction you're stepping in. For the purposes of testing is that acceptable? What would be good is if you could get a TAS or something (do they exist for S4 yet? Not sure, there's only certain parts of the forums I bother to read) to ensure consistency although that is not a foolproof method for guaranteeing max distance, but perhaps worth looking into to give rough benchmark figures.

Since I'm actually posting now and probably won't get many chances to post again in the next few days and weeks, I may as well share some very rough tests I took as per this criteria on a few characters which are of interest to me for various reasons, most of which are already in the table. Basically I tested each character 30 times from the leftmost side of the stage and recorded the distances, I'll just stick the biggest ones here. However I am still very inconsistent at this so get these properly verified, but hopefully it's helpful at least a little.

I am measuring from the character's back foot, but note that from the starting position the back foot is actually standing on air, not the plank. Trouble is it is very difficult to find a consistent part of the model to use as the measurement - feet is all I can use really as they generally stay still and little room for error when reading the dtstance due to being on the plank itself. I also have to ensure that the characters are as far back as possible when starting to keep the test as fair as I can.

1 plank(approximate position after the step dash finishes and Ness stands up: back foot is one the line between plank 1 and 2)
1 plank(approx position very similar to Ness. Although Lucas looks like he goes further during the crouch, I put that down to the crouch animations being different.)
3 planks (now this one I am really unsure about. It's not easy to gauge when Mewtwo has taken a single step because he levitates, moves his feet behind him and leans at 45 degrees rather than lifting his leg. Really unhelpful.)
2 planks

Yeah I know it's not many but most of the characters I did are already in the table and confirmed I guess? Definitely make sure you confirm these distances though. I've only had a couple of hours to test this so far... including learning how to actually do it.

So TLDR easier method of testing this without putting the stick back to neutral might be worth doing for more efficient testing, and if you can TAS this or something, that might be worth looking into. That's the main gist of this long ass post.
-----
Sidenote, for what it's worth, I struggled with a little bit with your method due to the planks being quite thin (can't tell 0.5 very well even with camera set to zoom) and because the foot of the characters drops of the edge but not quite far enough to count as a plank I don't think. Although I dunno how to fix that whilst keeping the test fair. [[Personally I use Omega Onett when testing distances and stuff because the grey squares on the roof where the characters stand and the windows on the side of the stage are good for measuring imo. When the foot drops off the stage, it's about an extra two small roof squares. Easier to gauge then Suzaku planks. But since you've started on Suzaku you should stick with it. Everyone uses a different stage and scale when labbing stuff like this anyway tbh]] This is something to be aware of when collecting data because it can produce anomalous results. Or maybe I'm just being stupid, I am pretty sleep-deprived atm. w/e.

Cool project, hope it works out.

 

[Radical Larry]

I believe you should look at how long it takes him to actually lift his leg up (Mewtwo's), as that'll gauge a step. If it's mid-way, I believe that can be a step. And I'm well aware that not all is to be truly confirmed, but I'll at least try finding someone who can confirm or deconfirm these.

And as for my method, I believe there can be certain character attacks that can tell you where they're at, if it's not the shadow. Examples being many of the characters' Down Aerials (e.g. Link, Toon Link, Mewtwo, Rosalina). Depending on where they land should determine the character's distance. If not that, then use the leg that's closest to the background as another example, though D-Air would be the most optimal on many characters (except Sheik, Sonic, ZSS, Fox and Falco, either for tilted D-Airs or stall-and-fall D-Airs that'll move them).

And thanks for the testing.

 

[QualityQ]

I've been toying with this ever since the My Smash Corner video came out. I'll post some labbing notes that may help others, my methods, and then the distances.

First, note that there are three distinct walking animations per character, to my knowledge. ROB is the best example of this because they are distinct based on the angle of his head. Based on my results, the distance you get is dependent on which animation occurs; the slowest, medium or fastest walking speed. It just so happens that the fastest animation results in the farthest 'step dash,' and the fastest walking animation speed comes from tilting the joystick the farthest distance (without dashing).

This means that the timing is less important (as in, you don't have to stop the first step at a specific point), and how far the joystick is during your first step (the walking animation) determines which step dash distance occurs. This is separate from the step-to-crouch part, which also plays a factor but I've learned less about. So overall, to get the longest step dash you need 1) step in fastest walk animation; then 2) the step-to-crouch slide.

I've used the methods as listed in the OP with some notes; I set speed to 1/2 (for consistency) and camera to zoomed out (which allows you to see planks more clearly). For all step dashes I started in crouch to step then the step-to-crouch slide. It seems to me 0.5 distances are irrelevant since you can get a longer distance from crouching at max walking speed, but I kept those numbers for convenience. Also, I made sure I could get the distance at least 3-4 times before I considered it accurate. I tried to choose characters not on the list yet.

Farthest step-dash distances:

7.5
3.5 (may be shorter +/- .5)
6
0.5
2
5 (may be even longer +/- .5)
3
4

Hope this helps, this seems like a very powerful tool for characters with a longer distance.

 

[Radical Larry]

Alright, the OP is updated, but there is going to have to be something special done with Shulk, as is the case in many techs, and it's that someone will have to measure his Step Dash with Speed and Shield Monado Arts as well as his normal state. They do shift his grounded speed and may affect his Step Dash distance.

 

[firstbones]

hi, i usually just lurk but eh i guess any time is good to start posting. i was wondering - woudn't omega windy hill zone make it easier to determine step dash distances? i remember ambrose's perfect pivot vid and he used that method.

 

[Radical Larry]

No it wouldn't. For one, there are blockages surrounding the little squares, making them quite inaccurate to measure. Plus, it would take a lot more observations to determine just where the character landed by comparison to Suzaku, which shouldn't take much observation at all. Plus, I've decided to use Suzaku Castle before putting up the thread.

TL;DR It's easier to do Suzaku.

 

[Horseketchup]

Would you say that getting 2+ planks allows the step dash to be reasonably useful for that character (excluding which moves carry momentum or not)? Even if they don't have good moves which carry momentum, one could always wait and then input the move after the slide, although it's as optimal.

The planks are actually somewhat wide, so 2 planks seems like it would exceed average perfect pivot distance which I think could be pretty significant.

 

[Radical Larry]

Well in my honest opinion, the usage of step dash depends on various aspects of a character, including mobility or even initial dash distance. An example of never to use step dash is Marth and Lucina, who's step dashes, while being in the 2+ plank category, are not their best mobility tactic. Their best mobility tactic is their own extended initial dash, which covers twice the distance of their on step dash, and due to their initial dashes ending early, they can hit the opponent out of it better than using a step dash.

However, on characters with not-so-lucky mobility or initial dashes, like Link, Ganondorf or Falco, I'd recommend you use step dash when it's applicable. Falco has an amazing step dash for his poor mobility, so that's a very easy option to approach opponents when mastered.

So to answer your question; yes and no. It can be reasonably useful for some characters, but many others have better options than Step Dash that make Step Dash somewhat negligible. Again, Marth and Lucina make an example of that. And an even better example is Little Mac, who's extended initial dash/extended foxtrot can be used three to four times just to get from one end of Final Destination to the other.

 

[MattPerren]

Hey this is Smash Highlights account, I never use SmashBoards but I just found your message! Your comment was under spam so it only just reached me...

I posted a video on Step Dashes yesterday and I'll be posting part 2 with community submitted footage soon.
Here's my video: https://youtu.be/p_ynxsZmqh8
Here's my reddit post: https://www.reddit.com/r/smashbros/comments/538kzf/ive_spent_weeks_compiling_a_step_dash_tier_list/
I read this thread a couple of weeks ago so I'll credit you for helping to rank characters.

Sorry for responding so slow! Do you have a Twitter?

 

[Radical Larry]

Hey this is Smash Highlights account, I never use SmashBoards but I just found your message! Your comment was under spam so it only just reached me...

I posted a video on Step Dashes yesterday and I'll be posting part 2 with community submitted footage soon.
Here's my video: https://youtu.be/p_ynxsZmqh8
Here's my reddit post: https://www.reddit.com/r/smashbros/comments/538kzf/ive_spent_weeks_compiling_a_step_dash_tier_list/
I read this thread a couple of weeks ago so I'll credit you for helping to rank characters.

Sorry for responding so slow! Do you have a Twitter?

Holy hell, it's awesome seeing you here, dude! Of course YouTube would screw up the comment system. That's just a given.

Though I saw the video and I really like it. The frame dependency on Step Dash is definitely something I haven't realized can occur in characters, so this definitely put in some mixed results here and there and whatnot. I'll definitely update my list with the new Step Dash rankings alongside the current system.