Quick Question, Quick Answer Thread

[Indigo Jeans]

Sizes are in thinkaman's doc, the one with aerial landing lag, etc.

 

[Kinzer]

Can anybody tell me whether or not a character will suffer more shieldstun if their shield is depleted any more than a good/healthy shield, or it doesn't matter?

 

[Luxor]

Shield health doesn't affect shieldstun, no. Only damage dealt is a factor, specifically shieldstun = floor(damage/3).

 

[JOE!]

that doc did not help me...it said Diddy and DK were the same size

 

[Yikarur]

Shield health doesn't affect shieldstun, no. Only damage dealt is a factor, specifically shieldstun = floor(damage/3).

mh..
8,4% = 2 Frames (2.8)
13,65% = 4 frames (4.55)
14,7% = 5 frames (4.9)
16,8% = 5 frames (5.6)
17,85% = 6 frames (5.95)
33.6% = 11 frames (11.2)
23.1% = 7 frames (7.7)

a few test percentages to look into your formula.

so under which conditions will be rounded up?
14,7% = 5 frames (4.9)
17,85% = 6 frames (5.95)
11.7% = 4 frames (3.9)
8.76% = 3 frames (2.89)
-
14.4% = 4 frames (4.8)
2.43% = 1 frame (0.81)
2.4% = 1 frame = (0.8)
5.4% = 1 frame = (1.8)

the border between "rounded up" and "rounded down" seems to be about Y/3 = X.8

 

[Luxor]

Were those moves fresh or in training mode?

 

[Yikarur]

decimal places only apply in Vs. So in Training Mode it would be 13% in Vs. 13.65%
I've used versus.

 

[KayLo!]

For some reason, Zelda's dair sweetspot gives sourspot damage/knockback on grounded opponents, but it keeps them in shield stun for 5 frames as if it's truly sweetspotting. Any idea why this is/if this happens with other similar moves? I'm guessing the shield somehow negates the fact that they're grounded?

Just curious really.

 

[phi1ny3]

I'm curious if it's because the shield itself is above ground level and therefore registers like so. Did you try this on Yoshi's shield (although it's going to have severe disadvantage either way, fhl ) and note the advantage (by seeing what frame his spotdodge or shield grab appears)? Or maybe trying angling shields? idk, sounds interesting, and from my limited playing with Zelda I too have noticed this.

Also, I'm curious as to the plausibility of recording advantage for aerials, don't things like FF and incomplete aerials, among other things, create factors that make finding a true advantage hard/impractical, or is this actually possible?

 

[rPSIvysaur]

Go in frame advance. Tell me what frame it hits.

 

[KayLo!]

The sweetspot hits on frame 14. Sourspot from 15 til whenever. The sourspot gives 1 frame of shield stun like it's supposed to..... I just figured the sweetspot would give 1 as well since her dair is dumb and can't sweetspot grounded people on hit.

I'll try doing it on Yoshi just for kicks. I tested it on Bowser earlier because someone requested it.

 

[rPSIvysaur]

What frame is the d-air hitting shield in frame advance.

 

[KayLo!]

............Frame 14 for the sweetspot. 15 for the sourspot. x.o I tested both.

 

[rPSIvysaur]

What frame did it hit shield? (I can't use the wii atm)

 

[KayLo!]

Okay .

 

[KayLo!]

@Phil: Yoshi was the same. Why'd you think he might be different?

 

[rPSIvysaur]

Yoshi's shield has some weird properties... who knows?

 

[Luxor]

decimal places only apply in Vs. So in Training Mode it would be 13% in Vs. 13.65%
I've used versus.

Were they fresh or used once? If they were totally fresh they would be off by that amount due to fresh move bonus.

 

[Yikarur]

varies. I use a damage formula excel which gives me fresh + stale moves 1-9 usage and vice versa.
I used the moves+staleness I needed to get the result I tried to confirm.

The Shield Stun formula doesn't work if you always round down. But it seems to work like I said Y/3 = X.8 = rounded up. everything below = rounded down.
but you have to use decimal places so that it's really exactly.
14% does fresh 14.7%, 14.7/3 = 4.9 = rounded up = does 5 frames shield stun.
22% does fresh 23.1%, 23.1/3 = 7.7 = rounded down = 7 frames shield stun
4% does 2.4% with the first four stale queue entrys = 2.4%/3 = 0.8 = rounded up = 1 frame shield stun.
this seems to be correct. (constant results after a lot of testing :/ )

 

[KayLo!]

I'll go update the shield stun formula. Probably means that the rest of the Floor equations aren't exactly right either.

This is why frame advance will always be better than formulas and crap. <3 Thanks Yika.

 

[phi1ny3]

Yoshi's shield has some weird properties... who knows?

This mainly. Also had the silly idea that it might be registering shields as in the air since they do create a different hurtbox (sort of). I wanted to see what would happen with a shield that stays on the ground.

Kind of a silly idea in retrospect

 

[KayLo!]

Well, on the plus side, it's much easier to tell when shield stun is over with Yoshi (since he begins to pop out of his egg). I'll probably keep using him as a test dummy in the future. =X

 

[phi1ny3]

about vertical kill moves:
I know this isn't usually applicable when one is in the air (and when usually these kind of kill moves hit you) because the hitlag or distance is too far for SDI to take care of, but several times while I'm on the ground, I've been able to SDI into the ground moves like Snake utilt or Marth usmash (usually by accident), and just get knocked into the ground instead of dying, despite people saying that this isn't possible (at least when someone mentioned this in early 09 lol). Anyone know why this hasn't been looked as much, aside from the situations that it doesn't work as often?

@Kaylo, I forgot about that, that's a good point

 

[T-block]

I'd like to see that answered as well. My understanding of it would be as long as you take the hit in the air so that you're allowed to SDI down, you should be able to SDI into the ground and tech, but I've heard that past a certain percent it becomes impossible to do that. Why is that?

 

[Yikarur]

well the technique's name is ASDI. This means on first frame of knockback you get warped in the direction you're holding. BUt that warp is not much and contrarys with your true knockback.
That means if you SDI into the ground you have to hold down until the knockback starts, if the first frame of vertical knockback is not as much as the distance you're gaining through ASDI you hit the ground.
Thats why it doesn't really work at high% with vertical moves but can work a long time with horizontal moves
like here: http://www.youtube.com/watch?v=WbAYGJTzHMU&feature=mfu_in_order&list=UL

 

[Luxor]

Yika, it doesn't have to be ASDI, and actually ASDI will rarely be enough- QCDI works better, ASDI is really only for multihit moves. According to one of Veril's threads, you can SDI into the ground whenever an attack pops you into the air DURING hitlag- this is probably the ase with Marth's Usmash and Snake's Utilt.

Frame advance testing would be good, I'll do it tomorrow if nobody has yet.

 

[phi1ny3]

^That would explain it.

 

[Yikarur]

Yika, it doesn't have to be ASDI, and actually ASDI will rarely be enough- QCDI works better, ASDI is really only for multihit moves. According to one of Veril's threads, you can SDI into the ground whenever an attack pops you into the air DURING hitlag- this is probably the ase with Marth's Usmash and Snake's Utilt.

Frame advance testing would be good, I'll do it tomorrow if nobody has yet.

I back up everything I say on Frame advance testing.
So my rule I posted in another thread is still valid

"1. Yika is always right (about brawl mechanique)
2. Yika is always right
3. see rule 1 and 2"

xP

QCDI is normal SDI.
I just explained how that whole thing actually works.
it IS ASDI in the end but you have to, of course, SDI as low as possible.
ASDI Warps you in the direction you're holding in the first frame of knockback, it's almost unnoticebly on ordinary moves but it really occurs.
and you can only SDI down if you get hit in the air. If you're grounded you're not allowed to SDI down at all.
that's why it almost always works with multi hit moves, the first hit knocks you automatically in the air, every hit afterwards will hit you in the air so you can DI down.

summary: if you get hit, in the air, close to the ground, SDI as down as possible and HOLD down until the knockback starts and you'll hit the ground when the first frame of knockback is less then the distance you get through ASDI.
The condition "in the air" is met with almost every non DI't multi hit move from the second hit.
The Knockback needs a certain strange through (example: a weak uptilt (Snake; not fresh) doesn't have enough knockback to get you collide with the floor, like a spike at low%.)

 

[rPSIvysaur]

Hey Yika, do you think you can write up a a public release thread about "God-DI" and letting them know how it happens and what they really should be calling it?

 

[Yikarur]

Yeah now the question: what should we call it? we don't have a true name for it.

E: I don't find Verils Thread about that, could you give me the link to it? D:

 

[Luxor]

Yeah Yika, I know you're right; I was clarifying. I posted the link to the thread somewhere, check the last page.

God DI sounds fine to me. No need to overcomplicate things.

 

[Indigo Jeans]

^ I love this guy's modesty.

QCDI is optimal DI because of the way DI was coded in Brawl. It was coded for the Wiimote. Because the Wiimote only has positions

*8
456
*2

available, DI was coded for that. However, since the GCC has all positions available,

789
456
123

An input in either 7, 9, 1, or 3 is read as an input in 4/8, 6/8, 4/2, or 6/2, respectively. Sure it's SDI, but it's special SDI, hence it deserves its own name

Wtf internet getting me triple ninja'd.

 

[Yikarur]

lol nice, I didn't know this but you're right (I test everything I read instantly =D)
how did you know/discover/whatever that? o:

 

[rPSIvysaur]

Dantwizard I assume.

 

[Luxor]

No, it was in swordgard's originial "How2DI" thread. I think swordgard just guessed correctly and it entered the communal consciousness. That's how I found out about DI/SDI/QCDI/ASDI in the first place, anyway.

 

[Remzi]

why does this work?

http://www.youtube.com/watch?v=PUtUo7jOGGU&feature=player_embedded

 

[Yikarur]

http://www.smashboards.com/showpost.php?p=11609863&postcount=48

 

[KayLo!]

No, it was in swordgard's originial "How2DI" thread. I think swordgard just guessed correctly and it entered the communal consciousness. That's how I found out about DI/SDI/QCDI/ASDI in the first place, anyway.

Do you have a link for that by any chance?

Also, can someone give a brief rundown on buffering in general? I know the basics of it, but, for instance, I didn't know you could input that many things into one buffer until I saw the DDD video. I thought it was limited to one move per buffer.

I'm also a little fuzzy on what exactly you can buffer/what you can buffer from.

 

[Yikarur]

you can buffer a lot of stuff but that Dedede thing was really extreme lol
you can buffer from everything (with exceptions)
and the main rule is that lasting actions overwrite buffer inputs like.. if you hold left/right you will walk instantly, if you hold down you'll crouch.
You can't buffer from ledgejumps, ledge hangs (you can buffer ledge options but not the drop), that is what comes in my mind first.. but I think there's more.
you can buffer for example dash, jump, aerial.
there is no limit of buffering, just the 10 time window as long as the actions merge each other. (like dash jump aerial)

The Dedede one was really complicated. He buffers back dash (which results in a turnarround only because it's always turnarround -> dash), crouch (to cancel the turnarround) and another dash so that he dashs in place in the opposite direction on the first frame.

 

[Indigo Jeans]

He doesn't dash in place in the opposite direction with the second dash (if I'm understanding you right). He inputs the second dash after the down input to, well, dash. However, once he inputs opposite direction + Z, he is essentially cancelling the dash on frame 0, its input frame, because of buffering. This is how BDACUS works. And since he didn't gain anything from the dash apart from making the grab a pivot grab, he pivot dash grabs in place instead of doing it moving backwards, which allows multiple grabs.

tl;dr facing right
first dash input to the left is to register turn around
crouch input to allow a second dash input in the same direction
second dash input to register "start dash w/o turnaround this time"
right input + grab to cancel dash momentum and allow a pivot grab

imo:bacon: