• Welcome to Smashboards, the world's largest Super Smash Brothers community! Over 250,000 Smash Bros. fans from around the world have come to discuss these great games in over 19 million posts!

    You are currently viewing our boards as a visitor. Click here to sign up right now and start on your path in the Smash community!

Ask a Question, Get an Answer Thread~[Read Before Asking a Question]

weedwack

Smash Journeyman
Joined
Jun 24, 2011
Messages
365
Location
NJ
I'm looking for data on tech roll speeds, distances, and durations. Anyone know where I can find that stuff?
 

Krynxe

I can't pronounce it either
Joined
Aug 1, 2011
Messages
4,903
Location
Lakewood, WA
3DS FC
4511-0472-1729
Have you already checked this thread?
yep, I had already checked that thread. The descriptions all labelled them as inferior versions of the stock n64 joystick so I was offput

I went ahead and got a nice, relatively underused joystick and replaced mine with it. Also threw in some mechanical grease to help prevent some wear, so hopefully that will do. If there's ever any sort of alternative with a gamecube-like stick, someone please let me know :) I'll wait until after I finish my engineering courses until I go and do hardmods and casework by myself.
 

mixa

Banned via Warnings
Joined
Jun 6, 2012
Messages
2,005
Location
Isle of ゆぅ
I'm looking for data on tech roll speeds, distances, and durations. Anyone know where I can find that stuff?
I have no idea about speed. And I doubt it's gonna be useful to anyone that's just looking to play better.

The distance depends on the character, and more stuff. So I don't know to test that. If I had some good way to test all characters having 0 horizontal momentum and teching at the same spot, I could measure distance.

Duration is the same for every character.
I recorded tech-in-place and tech-roll for the Frame Display. I haven't finished Ness yet, but those techs and hitting a wall will be in the Ness folder. My intention is to show the intangible and vulnerable frames.
I'll probably upload the Ness today, but meanwhile, you can look here:
You're already intangible on frame 1, assuming frame 1 is when you hit the tech.
square with "input" is when you'll be able to input something. In that table I buffered shield, so you know for a tech in place there are two vulnerable frames.




tech-in-place gif:



To be clear: that table works for everyone and every tech (in place, left or right) I tested them all out.
 

Sangoku

Smash Master
Joined
Apr 25, 2010
Messages
3,931
Location
Geneva, Switzerland
The distance depends on the character, and more stuff. So I don't know to test that. If I had some good way to test all characters having 0 horizontal momentum and teching at the same spot, I could measure distance.
If you use Madao's cheat engine table, you can see each character's x-position, so you can compute the difference between the two values and find the distance. Although I agree it's probably completely useless.
 

weedwack

Smash Journeyman
Joined
Jun 24, 2011
Messages
365
Location
NJ
Not useless. I have very good reason to believe ness can react to tech rolls, chase, and dair.

As in I have done it many times. But only to pika. If I know all the tech speeds it will help me determine which characters can be chased and which must be read.

And woah, did not know there was such a huge difference between tech rolling and teching in place. Thanks Mixa.
 

Sangoku

Smash Master
Joined
Apr 25, 2010
Messages
3,931
Location
Geneva, Switzerland
Still pretty sure distance is useless. What if I tell you a tech roll is 400?

And I've already showed Ness could react to tech rolls, with perfect DJC and perfect reaction time. But again, frame by frame theory differs greatly from in game gameplay.
 

weedwack

Smash Journeyman
Joined
Jun 24, 2011
Messages
365
Location
NJ
I would analyze the data on a comparative basis.

This is not theory. Like I've said, it's something I've done many times. It's not even that tough against pikachu.

I find this whole, "you suck and don't know how to use data anyway" attitude pretty disappointing.
 

mixa

Banned via Warnings
Joined
Jun 6, 2012
Messages
2,005
Location
Isle of ゆぅ
I find this whole, "you suck and don't know how to use data anyway" attitude pretty disappointing.
hm, I don't think that's what was happening.



Tech-roll Distance



What the numbers mean: see this pic by Madao.

0 is the middle of DL (where 1P spawns) and
2318 the farthest position while ON stage (roll to the edge).

The way I tested this was simply doing Fox first-hit-of-Uair and then teching.



I recorded both Link (farthest tech-roll) and Kirby (shortest), so you can see it frame by frame: download here. instructions to use the Frame Display program here (and by 'Start' I mean 'Spacebar', and you should click on the first name on the list, Luigi)
Sorry about the camera. I don't know how much time I'd need to spend on it for it to become good.

The speed is just the distance divided by the time, right? Time is always ~40 frames. Now you can do the math.




edit: yuh, my previous post here was deleted too. thanks Sangoku for reminding me of the Madao table, I forgot it worked outside of training mode.
 

weedwack

Smash Journeyman
Joined
Jun 24, 2011
Messages
365
Location
NJ
Exactly what I needed, thanks Mixa.

And yea I get frustrated pretty easily nowadays lol finals stress.
 

Sangoku

Smash Master
Joined
Apr 25, 2010
Messages
3,931
Location
Geneva, Switzerland
weedwack: don't be too sensitive, no personal attack.

Mixa: I saw your first post and I am quite confused with this whole post deletion thing. I have no idea what's going on here, but I'll ask admins if this continues.
 

weedwack

Smash Journeyman
Joined
Jun 24, 2011
Messages
365
Location
NJ
Those numbers are actually really useful to me. I now know Kirby can be chased instead of read, and fox and samus are the next two I should attempt.

I come from a scientific background so when people tell me data is useless to me and I know it isn't, it's a little upsetting. But yea now that I have the data any arguments are totally disincentivized.
 

The Star King

Smash Hero
Joined
Nov 6, 2007
Messages
9,681
weedwack tbh, I kind of doubt you are legit reacting to tech rolls, unless they're next to an edge and it limits their techroll distance or something like that. You probably just think you are, when in reality you are kind of expecting them to take a certain option, and then attacking that option after they tech roll, and it's successful and it feels like you reacted.

It's like when you play the Mario Party minigame at the start of this video when you are supposed to react, and you plan on reacting, but you press A at the wrong time because part of you expected it to be there when one flips over, so you weren't really reacting at all because of your expectations, though it's hard to tell. You know what I mean? I don't know how to put it.

I can't back up what I'm saying though, but yeah. Ideal would be to TAS to see within how many frames you need to react.

edit: and no this is not a "you suck" thing, this is a "humans suck" thing
 

weedwack

Smash Journeyman
Joined
Jun 24, 2011
Messages
365
Location
NJ
I know what you mean, and I've considered the possibility that I'm tricking myself into thinking I'm reacting to it. Chronology does gets blurry over intervals that small. But that seems statistically unlikely at this point, as I've gotten five or six consecutive chases multiple times.

I'll show you next time we play console. Maybe make a little wager on it or something.
 

The Star King

Smash Hero
Joined
Nov 6, 2007
Messages
9,681
That sounds like a good way to back up the claim, I guess:

-I randomly generate (because humans pretty bad at being random) 10 integers 1-3, corresponding to left/right/in place
-You pick Ness, I pick Pikachu
-You pop me up with a dair in the middle of Dreamland so that I can tech, and I tech according to the numbers I generated, and hold the shield button to protect myself ASAP after the tech lag ends (no looking at my controller ya cheater)
-If you successfully punish all 10, the odds of that would be (1/3)^10 -> ~.002%, so that would pretty much support your claim of being able to react

Unless you punish techrolls with grab? In which case maybe I should mash up-b instead of holding shield.
 

mixa

Banned via Warnings
Joined
Jun 6, 2012
Messages
2,005
Location
Isle of ゆぅ
hahahahaah weedwack pulling the hbox and taking a ~peek at the controller

oh man

I had thought of something similar, SK, but I didn't know what kind of tech we were talking about, is he just standing there as Pika techs or is he landing after a Dair? and like you said, we don't know what is the punish: grab or sth else
 

The Star King

Smash Hero
Joined
Nov 6, 2007
Messages
9,681
I'm not sure I understand the question, but I'm saying that he does a landing dair/DJC dair so that he is in a resting position well before the time Pikachu techs, and he can do whatever he wants to punish the tech. Does that answer it?

The punish is up to weedwack I guess since he didn't claim anything specific, just that he can react and punish Pikachu's techs. I just want to know if grab is a possibility to know if I should hold shield or mash up-B.
 

mixa

Banned via Warnings
Joined
Jun 6, 2012
Messages
2,005
Location
Isle of ゆぅ
he does a landing dair/DJC dair so that he is in a resting position well before the time Pikachu techs
My question was kinda directed at weedwack since we don't know what he meant by tech-chase, but yeah, I get your scenario, and it seems like the easiest one too, so it's good.


I was gonna do some frame math beforehand, but screw that. I'll only add that you can't really distinguish a standing tech from a roll until after frame 6 at least.
 

mixa

Banned via Warnings
Joined
Jun 6, 2012
Messages
2,005
Location
Isle of ゆぅ
It's kinda funny how the same people are having the same conversations, just like in the Fox thread.

On what frame do you have to double jump to get the most out of his extended double-jump?
 

Sangoku

Smash Master
Joined
Apr 25, 2010
Messages
3,931
Location
Geneva, Switzerland
It's kinda funny how the same people are having the same conversations, just like in the Fox thread.
Smash 64 community's infamous bad memory. I don't know why it seemed so clear to me that it had been discussed. Maybe because I did the work so it stayed as a more vivid memory.

On what frame do you have to double jump to get the most out of his extended double-jump?
What do you mean by "the most out of"? The problem here is a "good DJC" must be the best balance between good distance (which is ROUGHLY the more you wait, the farther you go) and the speed (which is obviously the more you wait, the slower it gets). So it's hard to say. We could draw something like a ROC curve and determine stuff, but that wouldn't make much sense lol.

Also, this is EXTREMELY complicated for my simple mind. The distance/time (ie speed lol) depends on two variables: when you input your second jump and when you input your dair. The later you input your second jump, the farther you'll go (assuming you wait N frames between the second jump and the dair in both case). Now that's not equal obviously, because since you waited one more frame of first jump in the second case, you could use that frame between second jump and dair in the first case.

So (each - is a frame advance):

1) jump----jump------------dair
2) jump-----jump------------dair : goes farther

And:
3) jump----jump------------dair
4) jump----jump-------------dair: goes farther.

Which one affects the most, that's the difficult question. Difficult because I'm not sure if the answer is always the same or if it depends on the number of frames you consider. In the above example though:
5) jump----jump-------------dair: goes farther
6) jump-----jump------------dair

I'd say intuitively that the closer the jumps, the better it is, but I'm not sure it's always the case.

Also this is for backward jump, because a 3 frame interval between the two jumps makes no jump (not the case forward).

Edit 2: Thanks to Madao's cheat engine table (again lol) we can be more precise.
If Falcon techs backward, Ness must hit within this window:
time: 20 frames position: 1033
time 39 frames position: 1226

Now this is dynamic, so it's even harder to understand. Because Ness must be at one point at one specific frame. So it depends also on when you react (ie when the first jump is input).
 

Annex

Smash Ace
Joined
Jul 20, 2010
Messages
761
Location
Columbia Gorge
I believe it stands for "Self Destruct" meaning you fall off the stage on your own (no opponent gets a KO for it).
Even more technically it means you last touched the stage, not an opponent, before dying.
Like if puff is usmashed in the middle of hyrule at 0% and manages to puff on off the stage without touching the ground it's still a kill because the last thing she touched was the opponent.
 

mixa

Banned via Warnings
Joined
Jun 6, 2012
Messages
2,005
Location
Isle of ゆぅ
Oh wow, it didn't even cross my mind that Ness' DJC was so problematic. My question was pretty dumb.
I remember having read somewhere that if perfectly performed you could go from the interior edge of a side platform on DL to the interior edge of the other, but that's vague too, right.

What came to my mind was: if Ness' DJC really works as that physics experiment in which you spin in circles a ball connected to a wire, and upon cutting the wire, the ball will have its trajectory determined by the tangent line, then you could just see at which point in Ness' double-jump his X-velocity is greater, and from that you determine at which frame you should Dair (cut the wire), hoping that that will make him go farther horizontally.
 

clubbadubba

Smash Master
Joined
Apr 27, 2011
Messages
4,086
Even more technically it means you last touched the stage, not an opponent, before dying.
Like if puff is usmashed in the middle of hyrule at 0% and manages to puff on off the stage without touching the ground it's still a kill because the last thing she touched was the opponent.
Pretty sure if you get off of a wall it's still a kill though, so the definition would be which did you last touch, a floor of the stage, or the opponent (or the respawn cloud)?
 

Sangoku

Smash Master
Joined
Apr 25, 2010
Messages
3,931
Location
Geneva, Switzerland
Oh wow, it didn't even cross my mind that Ness' DJC was so problematic. My question was pretty dumb.
I remember having read somewhere that if perfectly performed you could go from the interior edge of a side platform on DL to the interior edge of the other, but that's vague too, right.

What came to my mind was: if Ness' DJC really works as that physics experiment in which you spin in circles a ball connected to a wire, and upon cutting the wire, the ball will have its trajectory determined by the tangent line, then you could just see at which point in Ness' double-jump his X-velocity is greater, and from that you determine at which frame you should Dair (cut the wire), hoping that that will make him go farther horizontally.
Well if that were the case, you'd have to cut the rope when the tangent line is going fully left. But that's the beginning of the jump! So there's no tangent force yet. We have a growing force on the one hand and a growing suboptimal angle on the other hand. Lol it's always such a mess. But yeah, in terms of pure distance, waiting a lot is better, but it will also make you go higher.
 

Annex

Smash Ace
Joined
Jul 20, 2010
Messages
761
Location
Columbia Gorge
Pretty sure if you get off of a wall it's still a kill though, so the definition would be which did you last touch, a floor of the stage, or the opponent (or the respawn cloud)?
You're right, I totally forgot about walls too.

I also forgot about spiking people into the ground at high percents and having them ricochet off. That doesn't count as touching the ground either.
 

Sangoku

Smash Master
Joined
Apr 25, 2010
Messages
3,931
Location
Geneva, Switzerland
If I'm pika and I'm sent up without dying, but I Up B up by mistake and die off the top, I would say I SD'ed. Didn't know there was a strict definition lol.
 

M!nt

Smash Champion
Joined
Jan 26, 2011
Messages
2,087
What came to my mind was: if Ness' DJC really works as that physics experiment in which you spin in circles a ball connected to a wire, and upon cutting the wire, the ball will have its trajectory determined by the tangent line, then you could just see at which point in Ness' double-jump his X-velocity is greater, and from that you determine at which frame you should Dair (cut the wire), hoping that that will make him go farther horizontally.
I think you made my head explode reading that
 

MrMarbles

Smash Lord
Joined
Apr 4, 2013
Messages
1,381
Location
Orlando, FL
i'm pretty sure its possible for ness to react to pika's techs. i know when playing online Wilde will punish my techs just about every time and i mean EVERY TIME. no matter how hard i try to be unpredictable
 

The Star King

Smash Hero
Joined
Nov 6, 2007
Messages
9,681
i'm pretty sure its possible for ness to react to pika's techs. i know when playing online Wilde will punish my techs just about every time and i mean EVERY TIME. no matter how hard i try to be unpredictable
I am pretty sure Ness can cover every option if you're teching on the left or right edge of a platform (or stage I guess...). In that case he can dair to cover both tech in place and tech into the closer side (because that tech roll won't go anywhere because of the edge of the platform) and then run and cover the last option.

But covering every option when neither of your tech rolls are limited in distance (not on a platform)? I've been playing for years and I've never seen anyone come even close to doing that (and I've played people like weedwack and Wilde who could reportedly do it). So if that really happens, I'm guessing you're failing at being unpredictable. But maybe it's just the former situation I was talking about.
 

weedwack

Smash Journeyman
Joined
Jun 24, 2011
Messages
365
Location
NJ
I only started doing this recently.

I'm not perfect at it yet, but I know in my smash heart it's perfectable. Against pika anyway.

I'll have it up and running by apex.
 
Top Bottom