Quick Question, Quick Answer Thread

[rPSIvysaur]

Knowing animation length also matters in case you're trying to ISJR

 

[KayLo!]

The difference between animation length and interruptability matters because if you interrupt a move with the same move again its staleness is delayed.

Example: If you continually d-tilt with Marth, as long as you interrupt the d-tilt with a d-tilt you are constantly in the d-tilt action. Each hit counts in the stale move queue, but the move's damage isn't decreased to match until you leave the d-tilt action (either the animation ends or you do something else).

Oh? Didn't know that!

 

[phi1ny3]

Yeah, it's the main reason why lucario can't do a bajillion pummels to refresh his moveset (I bet it's the same with other really fast ones like Wolf).

 

[Elyssa Xey Hexen]

I had question on the frame data.

Ike's Bthrow.

Spoiler

Subaction Main0x72
@17640

-----------|START|--------------------------
Throw Attack Collision - ID=0x0, Bone=0x0, Damage=0x4, Trajectory=0x1C, Knockback Growth=0x32, Weight Knockback=0x0, Base Knockback=0x32, Element=0x0, 0.00x, 1.00x, 1.00x, 0x0, 0x0, 0x3, True, True, 0x8, 
Throw Attack Collision - ID=0x1, Bone=0x0, Damage=0x3, Trajectory=0x169, Knockback Growth=0x64, Weight Knockback=0x0, Base Knockback=0x3C, Element=0x0, 0.00x, 1.00x, 1.00x, 0x0, 0x0, 0x3, True, True, 0x8, 
Asynchronous Timer - Frames=19.00x, 
Special Offensive Collision - 0x150000, 0x2, 0x169, 0x96, 0x1E, 5.00x, 0.00x, 2.00x, 0.00x, 0.00x, 1.00x, 1.00x, 0x31830440, 0x0, 0x84FFFC3, 
Special Offensive Collision - 0x140001, 0x2, 0x169, 0x96, 0x1E, 4.00x, 0.00x, 2.00x, 0.00x, 0.00x, 1.00x, 1.00x, 0x31830440, 0x0, 0x84FFFC3, 
Asynchronous Timer - Frames=20.00x, 
Throw Collision - 0x0, 0x49, RA-Basic[2], RA-Basic[3], RA-Basic[4], 
Reverse Direction
Terminate Collisions
Asynchronous Timer - Frames=32.00x, 
Allow Interrupt
----------------------------|END|-----------

I just wanted to know about how to go about reading this fully. Most of the text is fairly obvious. However, what language is the format 0x0, 0x32, 0x3C, called? Most of the time I see it in the format 0x## anyway, so I've come to just treat the whole thing as a value of the hexadecimal of the 2nd combination of numbers and letters.

Also, looking at the first Throw Attack Collision, and after Element=0x0, what does this string of data refer too?

0.00x, 1.00x, 1.00x, 0x0, 0x0, 0x3, True, True, 0x8,
The first 3 terms look like multipliers. But the rest I have no idea what this is referring to.

Also, on a lot of jabs I see
"Bit Variable Set - RA-Bit[16]"

Does this have something to do with inputs like pressing the attack button causing the Jab2 animation to start?

 

[MonkUnit]

Things like 0x2C are hex.

 

[Toomai]

0.00x, 1.00x, 1.00x, 0x0, 0x0, 0x3, True, True, 0x8,
The first 3 terms look like multipliers. But the rest I have no idea what this is referring to.

The 0.0, 1.0, 1.0 matches the pattern of "trip chance, freeze frames mult, SDI mult" used on hitboxes - except that they don't apply to throws. The 3 also looks like an air/ground specifier (again, throws don't apply). Personally I think it's some sort of inheritance shenanigans (i.e. throw objects inherit from a base hitbox class). The 8 at the end I think is "throw armour frames" but that's 100% untested. The other values all have some sort of meaning but I have yet to find a pattern.

Also, on a lot of jabs I see
"Bit Variable Set - RA-Bit[16]"

Does this have something to do with inputs like pressing the attack button causing the Jab2 animation to start?

Yeah, things like jab loops and jab continuation.

 

[Elyssa Xey Hexen]

Thanks.

Also, I just need some help coming up with some way to test something out.
Frame 0: Dash input
Frame 1: Dash animation starts; control stick + attack inputs
Frame 2: dash attack starts; grab input
Frame 3: Dash Grab starts
Supposedly, doing this "boost grab" (DAC grab) gives Falco extra sliding distance similar to the DAC pivot grab, but not nearly as much sliding distance.

Now, when compared to a delayed dash grab
Frame 0: Dash input
Frame 1: Dash animation starts
Frame 2: Dash animation continues; grab input
Frame 3: Dash Grab starts

These two methods seem to cover very similar distances in frame advance visually. So I have hypothesized that Falco does not have a boost grab. Instead the distances Falco covers per frame for either method is identical I think.

Although, I want to actually see if I could confirm this theory by whatever method the game uses for calculating how much a character's bodyframe translates each in-game frame. But, I'm not sure how to go about figuring this out.

In brawlbox, I can select TransN bone and track the numerical translation of this bone for each frame in the DashAttack, but the Dash animation will not translate in brawlbox.

 

[Toomai]

In brawlbox, I can select TransN bone and track the numerical translation of this bone for each frame in the DashAttack, but the Dash animation will not translate in brawlbox.

That's because dashes use attributes for their physics.

Falco has a Dash Init of 1.9 and a Run Speed of 1.432. I'm not 100% sure about this one, but his Dash Deceleration looks like 0.112.

So in theory Falco's speed per frame on dashing is 1.9, 1.788, 1.676, 1.564, 1.452, 1.432.

 

[Elyssa Xey Hexen]

Falco's DashAttack TransN translation is 0 for frame 1, 4.259 for frame 2, and 6.722 for frame 3. This first frame should take its translation from whatever value the Dash was in? Does this explain why every character has a form of a DACUS because the Usmash uses its translation per frame from the dash attack?

 

[rPSIvysaur]

Yes, it also applies for glide tossing.

 

[DeLux]

What would be the easiest way to test for a quantifiable amount of shield push in terms of IGUs?

 

[MonkUnit]

IGU ?

 

[DeLux]

In Game Units. I'm doing a option based deconstruction of matchups. I want to identify using game mechanics which options are punishable on shield via dash grab. But in order to figure that out, I need to have a method of measuring shield push and stuff

 

[Toomai]

Shield push I'm pretty sure is delivered just like knockback is, so you'll have to apply traction to figure out how far you'll slide.

 

[phi1ny3]

So tell me this, this is more of an empirical observation more than anything, but here's the scenario that to me, doesn't make sense:

Scenario A: A character approaches shield, and spaces a move, it isn't powershielded, it connects with shield on the front, it's safe on block (the opponent tries to punish it w/ grab/dash grab/attack, gets punished).

Scenario B: Same person, same attack, but he's doing it on crossup, or behind the opponent on his shield. Opponent wtfs w/ frame data, and does a turn around dash grab, gets the punishment.

Does anyone know what I'm talking about? I'm sure it's just because the opponent gets overconfident and doesn't dodge/attack before the opponent responds, but it happens often enough that I have to ask for clarification.

 

[MonkUnit]

videos of that crossup dash grab thing prease

 

[Toomai]

So tell me this, this is more of an empirical observation more than anything, but here's the scenario that to me, doesn't make sense:

Scenario A: A character approaches shield, and spaces a move, it isn't powershielded, it connects with shield on the front, it's safe on block (the opponent tries to punish it w/ grab/dash grab/attack, gets punished).

Scenario B: Same person, same attack, but he's doing it on crossup, or behind the opponent on his shield. Opponent wtfs w/ frame data, and does a turn around dash grab, gets the punishment.

Does anyone know what I'm talking about? I'm sure it's just because the opponent gets overconfident and doesn't dodge/attack before the opponent responds, but it happens often enough that I have to ask for clarification.

Um what's the question? (Using "opponent" to refer to both players is rather confusing.)

If it's "why do people connect with shieldgrabs more often if their shield is hit in the back", then there's no mechanical reason for it.

 

[MonkUnit]

If it is that question, then I would say that it's because of the animation and when the grabbox starts / where it is positioned.

 

[Dr. Tuen]

If you hit someone's shield and undergo hitstun/lag (sorry, i'm not that great on my terminology), how are those frames assigned? I'll make up an arbitrary example.

Move A:
Start up: 5 Frames
Hits: Frames 6-10
Cool Down: Frames 11-20
Shield stun/lag: 5 Frames

If you hit with Move A on frame 10, do you undergo shield stun/lag [5 frames], then cool down [10 frames]?

===

Also, if you move A is an aerial which has 10 frames of landing lag, what happens when you're experiencing aerial cool down (frames 11-20) and land on frame 15? Do you experience more landing lag (the move now ends on frame 25 because of the added 10 frames) or does it 'autocancel' and end on frame 20 anyways?

===

This should help me assemble a spreadsheet project I've been thinking about doing. I'll try it on a few characters and show you guys what it is soon.

 

[phi1ny3]

Um what's the question? (Using "opponent" to refer to both players is rather confusing.)

If it's "why do people connect with shieldgrabs more often if their shield is hit in the back", then there's no mechanical reason for it.

Oh yeah, whoops, didn't get around to editing it, that would've helped

@Monkunit ok, that makes sense, I've seen esp. w/ D3, Marth, or MK that they seem to get a dash grab despite it not being plausible via frame data (especially considering it takes an extra frame to turn around, and at least one frame more to initiate a dash -> dash grab from the get-go iirc)

@Tuen I think I have the formula around somewhere, if not it's in the lab probably.


Oooooooor if you just wanted to do it without worrying about numbers being fudged, one could test it via frame advance

 

[Stealth Raptor]

At work but real fast if you land with an aerial your aerial animation ends and you enter regular landing lag or aerial landing lag depending on if its before or after autocancel frame

 

[Yikarur]

Ganoncide
what is the factor for him losing?
I heard years ago stuff abough heigth etc but never got a accurate confirmed answer.

 

[Toomai]

I have no testing or science to support this, but it's a theory nevertheless: Because characters are grabbed by XRotN, but they touch blastlines with TopN, it's possible that the grabbed character's TopN is above Ganondorf's during the fall (because they're kind of facedown-ish). So if they don't hit the blastline on the same frame Ganondorf hits first.

 

[MonkUnit]

To add to that, I also have no proof to back this up, but, I think that if Ganondorf and the grabbed foe's TopN are in the same location when hitting the blastzone, then it would come down to port priority to see who would lose a stock.

 

[Yikarur]

no it's either Sudden Death or Ganon loses.

 

[KuroganeHammer]

Just a question on special flags; how do I go about finding some of them?

For example, Zelda's Neutral B special was super easy to find (204FFFC3), but Zero Suit Samus' down smash's special flag doesn't make sense.

Variables:
00000001 0003A980
------------------
00000000 00000000
00000000 0000000B
00000000 00000046
00000000 00000032
00000000 00000028
00000001 00083D60
00000001 00000000
00000001 0002BF20
00000001 000927C0
00000001 00004650
00000001 0000EA60
00000001 0000EA60
00000000 06430E94
00000000 00000000
00000000 014FFFC0
------------------
00000000 00000001
00000000 0000000B
00000000 00000046
00000000 00000032
00000000 00000028
00000001 00083D60
00000001 00000000
00000001 0002BF20
00000001 000F9060
00000001 00004650
00000001 0000EA60
00000001 0000EA60
00000000 06430E94
00000000 00000000
00000000 014FFFC0
------------------
00000001 000493E0
------------------
------------------
00000001 00124F80
------------------
------------------
PSA Commands(For Injection):
00020100 00011BA8
06150F00 00011BB0
06150F00 00011C28
00010100 00011CA0
06040000 00000000
00020100 00011CA8
64000000 00000000
PSA Commands if Params start at: 0x0
00020100 00000000
06150F00 00000008
06150F00 00000080
00010100 000000F8
06040000 FFFFFFFFFFFEE458
00020100 00000100
64000000 FFFFFFFFFFFEE458

I'm not too experienced with OpenSA2, but I'm pretty sure that's the special flags, and when I convert them to binary, it's missing 4 bits.

Can someone point out what I'm doing wrong?

 

[Toomai]

[...] it's missing 4 bits.

Can someone point out what I'm doing wrong?

You're not including the leading zero.

 

[KuroganeHammer]

Hahaha, oh man, I feel so dumb.

To be fair though, it's this hex calculator. It doesn't put the four zero's at the front of the number.

GG.

Thanks Toomai! <3

 

[phi1ny3]

Ok, remind me again, what is the formula for (dis)advantage? I'm getting around to doing aura percent scaling.

Also, do variables going off count towards momentum canceling? Say I have a dair that on frame 1, sets the variable essentially "gravity" to 0, would that mean that you would regain control sooner via the variable going into effect, or only when the hitbox comes around?

 

[Sieguest]

Shield Advantage = Hit Frame - End Frame + Shield Stun + Shield Hitlag - Hitlag

As for the variable thing, I don't really know.

 

[phi1ny3]

I just realized, it probably doesn't, since autocancel has a variable technically as well, and that doesn't necessarily affect MC.

 

[TheReflexWonder]

When footstooling an opponent who is in the air, will the opponent go into the "I've just been footstooled" animation if he is footstooled during the cooldown/endlag of a move?

 

[Pikabunz]

I don't think so. As far as I know, you can only footstool someone when they're not in the middle of any action except for grabs and landing lag.

 

[T-block]

heyo smash lab

how far did we get/what work has been done regarding a universal unit of distance?

i'm thinking about getting something going... inspired by wanting to compare dash speeds with rolls for the purpose of tech chasing. ie- linking frame data with space covered

if there's already a system in place (that's actually good) i'd like to leverage off of that obvs

 

[DeLux]

Are IGU's too intangible?

 

[TheReflexWonder]

I thought DMG and I decided that we would use Squirtles as a standard unit of measurement.

 

[MonkUnit]

I thought the general thing was something like 10 derpmeters is the same length as a stage builder block. That way you can just say something is X derpmeters and actually have a frame of reference.

Or just use character(s) sizes. "I can't believe he hit me from that far away! I was at LEAST 4.5 Ganon heights away."

 

[T-block]

refresh my memory on IGUs?

i was thinking tenths of a stage builder block actually

i need a resolution of at least the distance bowser moves in one frame in a dash for the proposed purpose. if it's reasonable to speak of smaller distances than that, it might be wise to define the unit of distance smaller for future applications.

 

[Toomai]

One Stage Builder block is exactly 20 units. But we really shouldn't be trying to invent some new unit of distance in my opinion because pretty much all the data uses in-game units.

Bowser has a run speed of 1.527 (that's in units per frame), so (assuming that takes priority over whatever TransN's doing) he covers one block in just over 13 frames at 4.581 blocks per second.

 

[T-block]

toomai, i love you, as always

that means we can skip a lot of the data harvesting and just do some number crunching