Official Ask Anyone Frame Things Thread

[Bones0]

I need the shield stun formal or stun from 3% to 11%

Search "Kirby Kaze frame advantage on block"

 

[tauKhan]

@the muted smasher
shield stun formula: floor{ (dmg (staled) +4,45)/2,235 }

@ schmooblidon Doesn't smash turn out of dash set your velocity at zero, that's what it looks like at least?

 

[schmooblidon]

@ T tauKhan Unfortunately not, it decreases at roughly: previousvelocity * 0.2 = newvelocity
That's why smash turn PC drops work.

Edit:
I think I just figured it out.

previousvelocity*0.25 - traction = newvelocity ; minimum 0

with that you could work out everyone's maximum smash turn distance for pc drops
(maxdashvelocity*0.25 - traction)*2 = smash turn distance

(might be a bit different for characters with higher inital dash velocities, as they may be able to smash turn at a higher velocity then their max, samus definitely comes to mind)

It suggests character's with a higher dash velocity and a lower traction will move the most distance.

Still gotta work out frame 2 dash though.

 

[schmooblidon]

K so I think i figured everything out, and it's rather interesting.

It seems that every character has an initial dash velocity value on frame 1, but you do not move on frame 1. So your frame 2 velocity is actually ( frame 1 velocity + 1 frame of dash acceleration ) if max velocity is higher, or (frame 1 velocity - 1 frame of traction ) if max velocity is lower.

You would never even notice this with some characters. Peach for example has a maximum dash velocity of 1.30. Your velocity on frame 2 of dash is also 1.30. But her frame 1 velocity value is 1.2. Her acceleration is 0.12, so it only takes 1 frame for her velocity to reach the maximum.

The only reason I could even determine this was because when you smash turn during a dash, you have that small amount of velocity in that direction. So it would take that velocity + maximum velocity, to reach maximum velocity in the opposite direction, which is enough of a difference to see peach for example gaining 1.32 velocity. (frame 1 initial dash velocity + 1 frame acceleration).

Finally I can create a dash dance equation!

 

[King Mob]

I need the shield stun formal or stun from 3% to 11%

http://smashboards.com/threads/tool...dstun-calculation-v1-11.324878/#post-14651043

 

[dude it's raining]

I asked this indirectly in the official FAQ/Q&A thread, but does wavedashing happen during jumpsquat or after jumpsquat? My gut tells me it's after, but I can't test it myself without a 'cube. And btw, my question in the faq/q&a thread was if it is after, would falling speed affect the wavedash?

 

[schmooblidon]

I asked this indirectly in the official FAQ/Q&A thread, but does wavedashing happen during jumpsquat or after jumpsquat? My gut tells me it's after, but I can't test it myself without a 'cube. And btw, my question in the faq/q&a thread was if it is after, would falling speed affect the wavedash?

After. Only the airdodge angle and traction will affect it.

 

[BSeeD]

Hey there.
I've been looking for some data on boost grabbing (especially with Sheik), and couldn't find anything relevant by searching for half an hour.

Can someone help me here ?

 

[schmooblidon]

Hey there.
I've been looking for some data on boost grabbing (especially with Sheik), and couldn't find anything relevant by searching for half an hour.

Can someone help me here ?

Is there anything specific you would like to know?

Sheik's max run/dash speed is 1.8. If you perform a dash grab you will start with velocity 1.72 and drop by 0.08 each frame.

Dash attack has a velocity of 4.58 on frame 1, 4.79 on frame 2, 3.39 on frame 3. (This is very very fast) When you start a grab you go straight back to 1.72 and drop by 0.08 each frame. So the longer you stay in dash attack (the latest you can grab is frame 4), you more time you are moving at super speed and thus the further you reach.

 

[schmooblidon]

Today I found something interesting with wavedashes, I'm not sure if this was previously known or to what extent.

Sheik
Marth
Link
DK
Roy
Bowser

When performing a frame perfect wavedash, where you airdodge on the first frame airborne and land on that frame too, these characters only move half the distance they should on the first frame. Their velocity is still supposedly 2.5 or whatever, and on the next frame it will drop by the traction depending on the character stats like normal, and gain the appropiate distance. But the first frame you only get half of it. This does not happen when you airdodge on the second frame airborne, and land on that frame, and also when perfect wavelanding.

I tested the implications on this a little with Marth and Sheik. For Marth, this only really affects his smaller wavedashes, and after a certain angle (slightly above the notch) performing a second airborne frame wavedash will gain more short term distance in the same amount of time. You sacrifice 1 frame before you can act though.

For Sheik, it is less effective, as she jumps much higher on the first frame and much be in airdodge for longer.

For all of them though, this is bit of a kick in the nuts and unnecessarily hinders their movement. Wavelanding is better, whatever the angle.

 

[Stride]

I heard Scar say that Falcon's reverse up-B grabs the ledge differently in PAL than in NTSC; more specifically he said that it was "worse" in PAL.

I'm very skeptical of this claim, especially considering it doesn't seem to have been objectively/rigorously tested. Is there any way this is true?

Here's the video where he says it after seeing Rocky whiff a reverse up-B recovery (with a transcript below):
https://www.youtube.com/watch?v=qxxA9OIlYko&t=10m39s

So, the other thing that I wanted to say is that, so obviously I went to B.E.A.S.T. in Sweden and I played Falcon in PAL. Falcon's reverse up-B recovery is definitely worse in PAL. And that's not one of the thing we have, like, in our little patch notes, which I think like, lots of US players, or NTSC players, should probably play PAL so that we actually get the full, like, changelist. But I'll bet any amount of money that reverse up-B has a way worse like, "grab the ledge box" in PAL.

 

[tauKhan]

I'd not put my money on that. Scar also claimed on stream that sdi is easier to do in PAL which is obviously false.

 

[Meru]

So in the smash wiki, it says:

"While the damage reduction caused by stale-move negation is about the same in Brawl as in Melee, the calculation of knockback in Melee ignores it (for non-projectiles), while Brawl does not. As the damage of a hitbox is a main component in the knockback formula this results in a much more noticeable drop in the knockback of a stale attack"

So a non-staled Falcon Punch (from the air) will send a Jigglypuff (with no A/S/DI) this far:
http://puu.sh/gwkxk/929607cf9e.jpg

while a fully staled Falcon Punch (from the air) will send a Jigglypuff (with no A/S/DI) this far:
http://puu.sh/gwkII/309fb21831.jpg

Basically, it doesn't send people as far. Obviously. Which makes sense intuitively, and works experimentally. But theoretically (that is, from working through the formula given by StrongBad) the if the victim has 0% damage, then the amount of damage a move inflicts, does not matter.* So do we have any way to explain the knockback? Am I just interpreting the equation incorrectly?*
Am I crazy and stale-move negation simply is a multiplier for total knockback and not just the damage the move inflicts?

*
{{[(0* 0.1) + (Damage * 0 * 0.05)] * VictimKBMultB * 1.4 * [200 / (Weight + 100)] + 18} * (KBG * 0.01)} + BKB
=
{{ [(0)+(0)]*(1)*(1.4)*[1]+18} * (KBG * 0.01)} + BKB
=
{{18} * (KBG * 0.01)} + BKB

QED when "Total Damage" = 0, staleness is a nonfactor. But it is.
Also, this is odd because the equation implies that KBG is still a factor in moves even when "Total Damage" is 0. Which I guess isn't super weird, but whatever.

**
{{[(Total Damage * 0.1) + (Damage * Total Damage * 0.05)] * VictimKBMultB * 1.4 * [200 / (Weight + 100)] + 18} * (KBG * 0.01)} + BKB
I'm interpreting "Total Damage" as the amount of damage the character already has before the move (as in, you spawn in the stage with Total Damage of 0%)
And I'm interpreting "Damage" as the amount of damage that a move does to a target (E.G. Marth's fresh, uncharged tipper fsmash does 20%)

TL;DR
So I'm calculating knockback using the knockback equation, and it says that when total damage is 0, move staleness is a non-factor. But this statement is experimentally (and intuitively) false. Is there a way I can get around this discrepancy?

 

[tauKhan]

@ Meru I believe the "Total Damage" is victim damage after hit, so staling reduces a bit off kb that way in melee too.

 

[Kadano]

@ Meru yeah, as @ T tauKhan said it uses percent after hit. Damage is used as unstaled, though. Examine the excel sheets provided by Strong Bad for the exact math going on.

 

[Meru]

Lol what a simple solution/answer for what has been a massive issue in the calculations I've been running :b
Got it all fixed up and running fine. Thanks @ T tauKhan and @ Kadano , especially for answering so quickly. I kinda can't believe I never thought of looking at the excel sheet for the equations for Knockback.

 

[Phez_]

Is it possible for Young Link to do a 100% invincible haxdash / ledgestall?

edit: thanks @ schmooblidon

 

[schmooblidon]

Is it possible for Young Link to do a 100% invincible haxdash / ledgestall?

Yeah he can haxdash, it has a decent window for lenience too. Fastfall, and doublejump towards the stage on the 6th frame airborne. That's puts you at a nice position above the stage to waveland. You can also doublejumpledgestall on yoshis by fastfalling then doublejump on frame 9 or 10 airborne. On every other stage if you doublejump on frame 11, you will be tangible for 1 frame.

 

[Bones0]

Yeah he can haxdash, it has a decent window for lenience too. Fastfall, and doublejump towards the stage on the 6th frame airborne. That's puts you at a nice position above the stage to waveland. You can also doublejumpledgestall on yoshis by fastfalling then doublejump on frame 9 or 10 airborne. On every other stage if you doublejump on frame 11, you will be tangible for 1 frame.

Why is it different on YS?

 

[schmooblidon]

Because the ledge is on a slant, so your ledgesnapbox will touch it at a higher point than if it were flat.

 

[Stratocaster]

Does anyone know how which way you are knocked down is determined? That is whether they are stomach up or stomach down when lying down in the knockdown state? I have wondered this for a long time but I've never known. I just know more often than not they land stomach up. AFAIK shine always knocks down stomach up and Falcon's down throw will make them land stomach up until they are at very high percent (for spacies at least). Does it have to do with ECB and is it something that is actually predictable in normal gameplay? (without testing every individual case)

@ Kadano
@ schmooblidon
@Magus420

 

[Phez_]

Is there anywhere I can find IASA frames for every move?? In the 'hitbox and frame data' threads it doesn't have IASA frames

 

[Deleted member 212841]

gcyftftdtfyxycyvj

 

[IC-Rambler]

How do I get used to playing humans when I all have are CPUs?

 

[dude it's raining]

How do I get used to playing humans when I all have are CPUs?

M2K has said this:
Don't play against high level CPU's b/c they will/can do things that human ppl will not/can't do.
You can play against lvl 1-3 cpu's. They will do things that ppl won't do, but they won't do things that ppl can't do.
All you have to do is pretend that you don't know that they're dumb, and you can practice covering all options.
Source: http://clashtournaments.com/covering-options/

But even better than that, there's a guide on this site for how to start your own smash scene.
http://smashboards.com/threads/if-i-can-you-can-a-journey-to-start-a-smash-scene.394437/

 

[vZakat]

Not sure if this is the right place to ask but I didn't want to create a new thread just for one question. I was under the impression that you could do any standing action after a powershield but after practicing I can't seem to get anything other than oos options to come out. Is this because I'm powershielding a projectile as opposed to a physical attack? Also how many frames would you have to act after the powershield (assuming it can be done after a physical attack)?

 

[Bones0]

Not sure if this is the right place to ask but I didn't want to create a new thread just for one question. I was under the impression that you could do any standing action after a powershield but after practicing I can't seem to get anything other than oos options to come out. Is this because I'm powershielding a projectile as opposed to a physical attack? Also how many frames would you have to act after the powershield (assuming it can be done after a physical attack)?

http://smashboards.com/threads/offi...-questions-here.230020/page-106#post-16309106

 

[Kyu Puff]

The frame data for Ice Climbers' up tilt seem to suggest that a non-staled up tilt would keep a shielding opponent in constant shieldstun. However, when I test it myself, there seem to be frames during which the shielding opponent can buffer a roll or jump out of the (active, fresh) up tilt.

U-Tilt

Total: 49
Hit: 8-9, 11-12, 14-15, 17-18, 20-21,
23-24, 26
IASA: 42

The weaker hits each do 1% fresh, so they each have 2 frames of shieldstun, which should be enough to keep the opponent in shieldstun until the next hitbox comes out. What am I overlooking here?

 

[schmooblidon]

I think what's happening is, they get hit by 8, and are in shieldstun through 9-10, then they roll on 11 and the attack misses their hurtbox. Problem also with this is, if they get out of shield and get hit by one of the early ones, they will have considerable frame advantage on you.

 

[Kyu Puff]

I think what's happening is, they get hit by 8, and are in shieldstun through 9-10, then they roll on 11 and the attack misses their hurtbox. Problem also with this is, if they get out of shield and get hit by one of the early ones, they will have considerable frame advantage on you.

So the game will read a roll input even when they are in shield stun?

 

[Strong Badam]

Not sure if this is the right place to ask but I didn't want to create a new thread just for one question. I was under the impression that you could do any standing action after a powershield but after practicing I can't seem to get anything other than oos options to come out. Is this because I'm powershielding a projectile as opposed to a physical attack? Also how many frames would you have to act after the powershield (assuming it can be done after a physical attack)?

yes. you can only do grounded options out of powershield when powershielding an attack. powershielding a projectile functions as if you were never hit, so you do not suffer shieldstun and you're in your shield.
you still suffer shieldstun when powershielding a regular attack, so how many frames you have to wait depends on what you got hit by, but can interrupt the shield release afterward with any grounded A or B move.

 

[schmooblidon]

So the game will read a roll input even when they are in shield stun?

Well depends how you look at it. On uptilt frame 11 I'm not in shieldstun, so if I input roll or jump at the beginning of the frame, I'll roll/jump on that frame. I don't know about c-stick buffering, I've never been 100% on that as I never have c-stick in debug.

 

[Kyu Puff]

Well depends how you look at it. On uptilt frame 11 I'm not in shieldstun, so if I input roll or jump at the beginning of the frame, I'll roll/jump on that frame. I don't know about c-stick buffering, I've never been 100% on that as I never have c-stick in debug.

Alright, I guess I was unclear on how inputs are read frame by frame. I was assuming the hitbox on frame 11 would keep them in shieldstun. Is there an order/priority to the way things happen each frame (i.e., first inputs are read, then actions are initiated, then hitboxes take effect)?

 

[Stratocaster]

Alright, I guess I was unclear on how inputs are read frame by frame. I was assuming the hitbox on frame 11 would keep them in shieldstun. Is there an order/priority to the way things happen each frame (i.e., first inputs are read, then actions are initiated, then hitboxes take effect)?

The best way I can describe it is the game is reading ahead. I pulled up Magus' mod which can display shieldstun and here's what it does:

Frame 8: ~2.44 frames of shieldstun left
Frame 9: ~1.44 frames of shieldstun left
Frame 10: ~0.44 frames of shieldstun left (inputs here will take effect immediately next frame)
Frame 11: actionable

The game knows you will be actionable on frame 11 because it knows that 0.44 - 1 is less than 1, so there will be no shield stun. So if you input an action on frame 10, it will happen on frame 11. Popo's next hitbox comes out on frame 11 too, but the player's input from frame 10 applies to frame 11 and so on frame 11 they can be out of shield. For the same reason if they don't hold shield through the attack they will be hit by it, in contrast to Fox's drill which has enough shield stun to force a player to stay in shield whether they hold shield or not until the attack ends.

 

[tauKhan]

Alright, I guess I was unclear on how inputs are read frame by frame. I was assuming the hitbox on frame 11 would keep them in shieldstun. Is there an order/priority to the way things happen each frame (i.e., first inputs are read, then actions are initiated, then hitboxes take effect)?

Yeah, that's precisely the case. I'm not sure of the exact order things happen during one frame, but I think hitboxes are applied very late in the frame. Actions are done at the beginning, and movement is applied after that I think, but before hitboxes are applied. Even grabbing the edge happens before hitboxes, which is crucial for many intangible stalls.

 

[schmooblidon]

Anyone know exactly how the control stick position affects dash/run acceleration? I feel like there is a simple equation but I'm not seeing the pattern.

 

[Varist]

Riddle me this

why is it that when I set a 20xx 2-Player to hold shield and shieldgrab when attacked,

continuing,

and I fall down on top of him with a dair,

still following me

and then shine,

that a powershield animation happens. wtf. then he obviously gets hit by shine or sometimes he doesn't

 

[The Phenom]

Riddle me this

why is it that when I set a 20xx 2-Player to hold shield and shieldgrab when attacked,

continuing,

and I fall down on top of him with a dair,

still following me

and then shine,

that a powershield animation happens. wtf. then he obviously gets hit by shine or sometimes he doesn't

https://youtu.be/jHepfZTehKY double powershield, you can see it in this video @10second mark. I'll post more links

Edit: don't take all of what I wrote in the video comment as 100%correct Pls/thanks Since it was from 3 months ago I first believed shield could double powershield kinda like multi hit moves.

 

[Comet7]

Are instant wall jumps from the ledge only possible from when one reaches the peak of up an up special or jump a result of how the ECB reacts to those? Does the ECB change depending on the character, or is it the same for everyone? If it does, that begs this question to me as a Pichu user: would he be able to do an instant wall jump on any stages from even a wavedash or from a fastfalling state at the time when he grabbed the ledge? Are there any stages where he can do the wall jump with only a non fastfalled jump? I'm also curious about how spacies can do it; I know Fox can wall jump on FD from the normal double jump and I think illusion, but can Falco do the same with phantasm?

 

[Kyu Puff]

ICs have this weird mechanic when you wavedash to the ledge; as soon as you grab the ledge, you lose control of Nana and she double jumps. Sometimes she jumps and drifts back onto the stage, but sometimes she drifts off the stage and kills herself. I've been testing a bunch of factors to see what influences her behavior, and so far I've narrowed it down to distance from the ledge and angle of wavedash. I have a feeling it has something to do with her horizontal velocity and distance from the stage as she begins her double jump, and that there are several different variables that influence the final result.

In testing, I noticed something weird. I start from a savestate where ICs are facing away from the ledge. I use a series of TAS inputs to reproduce the same wavedash every time (jump, and airdodge on the first possible frame with 40X 40Y for control stick inputs). Depending on how long I wait before starting the series of inputs, the wavedash takes a different number of frames to reach the edge of the stage. It takes them anywhere from 8 to 10 frames to slide off from this particular starting position, even though each wavedash should be exactly the same. What could account for this variable amount of time?

@ schmooblidon I have a feeling you know the answer to this one