Juri move data (including AE changes)

juri

#1

I started poking through the SSF4/AE data recently, and I’ve found some interesting things about Juri. Here’s what I found…

Note
There seems to be 12 different entries for standing LP.
[LIST]
[]5LP (close standing LP)
[
]5LPF (far standing LP)
[]R5LP (light-chain close standing LP)
[
]R5LPF (light-chain far standing LP)
[]5LP_UC_ONESIDE (FSE-chain close standing LP)
[
]5LPF_UC_ONESIDE (FSE-chain far standing LP)
[]R5LP_UC_ONESIDE (unused?)
[
]R5LPF_UC_ONESIDE (unused?)
[]5LP_UC (FSE close standing LP)
[
]5LPF_UC (FSE far standing LP)
[]R5LP_UC (FSE light-chain close standing LP)
[
]R5LPF_UC (FSE light-chain far standing LP)
[/LIST]
The vulnerable hitboxes are different for R5LP_UC_ONESIDE and R5LPF_UC_ONESIDE. (This was not changed in AE.)


5LP/HURTBOX (AE)
S|E|   X|   Y|Width|Height
==========================
0|7|   0|-0.4|  0.3|   0.8
0|4|   0|   0|  0.2|   0.2
4|7|0.05|   0| 0.35|   0.2

R5LP_UC_ONESIDE/HURTBOX
no data

R5LPF_UC_ONESIDE/HURTBOX
S|E|   X|   Y|Width|Height
==========================
0|7|   0|-0.4|  0.3|   0.8
0|7|   0|   0|  0.4|   0.3

Juri’s normal cl.MP animation has the active frames starting at frame 7. However, part of this animation is run at 2x speed, making the effective startup 5. The FSE version of the move does not have this speed multiplier.


5MP/SPEED
 S| E|Multiplier
================
 0| 1|         2
 4| 5|         1
11|12|       1.5
28|29|         1

5MP_UC/SPEED
 S| E|Multiplier
================
11|12|       1.5
28|29|         1

cl.MK has a similar speed multiplier that is not applied for the FSE version.


5MK/SPEED
S|E|Multiplier
==============
0|1|         2
2|3|         1

5MK_UC/SPEED
no data

Juri’s active hitbox for 5MP_UC and 5MP_UC_ONESIDE is different from the one on 5MP. If this weren’t the case, standing MP might not whiff in FSE chains at certain ranges. (I might test this later.)


5MP/HITBOX
S| E|   X|   Y|Width|Height
===========================
7|11|0.25|0.35|  0.3|   0.2

5MP_UC/HITBOX
S| E|   X|   Y|Width|Height
===========================
7|11| 0.2| 0.3|  0.3|  0.15

Far standing HK has a different active hitbox for the FSE version.


5HKF/HITBOX
 S| E|   X|    Y|Width|Height
=============================
19|23|0.65| -0.1|  0.6|  0.25

5HKF_UC/HITBOX
 S| E|   X|    Y|Width|Height
=============================
19|23|0.65|-0.05|  0.6|   0.2

AE Change
Standing LP active hitbox is increased slightly and vulnerable hitbox is reduced during active frames.


5LP/HITBOX (SSF4)
S|E|   X|  Y|Width|Height
=========================
4|7|0.55|0.4|  0.3|  0.15

5LP/HITBOX (AE)
S|E|   X|  Y|Width|Height
=========================
4|7|0.55|0.4| 0.25|  0.15

5LP/HURTBOX (SSF4)
S|E|   X|   Y|Width|Height
==========================
0|7|   0|-0.4|  0.3|   0.8
0|4|   0|   0|  0.2|   0.2
4|7| 0.1|   0|  0.4|   0.2

5LP/HURTBOX (AE)
S|E|   X|   Y|Width|Height
==========================
0|7|   0|-0.4|  0.3|   0.8
0|4|   0|   0|  0.2|   0.2
4|7|0.05|   0| 0.35|   0.2

5LPF - same changes
R5LP - same changes
5LP_UC - same changes
5LPF_UC - same changes
R5LPF - NOT CHANGED
5LP_UC_ONESIDE - NOT CHANGED
5LPF_UC_ONESIDE - NOT CHANGED

Crouching LP vulnerable hitbox is reduced during startup and active frames.


2LP/HURTBOX (SSF4)
S|E|  X|    Y|Width|Height
==========================
0|7|0.1|-0.05| 0.45|   0.5
0|7|  0|    0| 0.25|  0.25

2LP/HURTBOX (AE)
S|E|  X|    Y|Width|Height
==========================
0|7|  0|-0.05| 0.45|   0.5
0|7|  0|    0| 0.25|  0.25

j.MP lost its ability to fake cross-ups and cause unblockables.


9MP/HITBOX (SSF4)
S| E|   X|    Y|Width|Height
============================
5|10|0.45|-0.45| 0.35|  0.15

9MP/HITBOX (AE)
S| E|   X|    Y|Width|Height
============================
5|10|0.65|-0.45| 0.25|  0.15

Kasatushi now has a hard-coded vulnerable hitbox during startup frames.


ATEMI_L/HURTBOX (AE)
S|E|  X|    Y|Width|Height
==========================
0|6|  0|-0.35| 0.35|  0.75

ATEMI_M - same change
ATEMI_H - same change

That’s everything I’ve found that I thought was of interest. Any other things you think may have changed in AE are probably either a result of other characters’ changes or Juri’s walk speed change. I might also make some modifications to this stuff for fun later.

This data was found using Dantarion’s bac/bcm editor Ono! (http://watissf.dantarion.com/ono_builds/)

Edit: I modified the chart format, so hitbox data now says ‘SEXY’ across the top.


#2

Nice finds man! I especially like the hard data on FSE, as it seems to point that the FSE wonkiness might actually be accidental and not intentionally made. I used to think many FSE moves are modified to prevent creation of FSE loops, but maybe some things are just things they changed in 1 place but forgot to change the other o_O

Edit: I actually thought at one point it would be interesting to modify SSF4AE PC version to make a “Juri Koryuu” edition, and test different buffs out in practice. Would be interesting to see how it would turn out if you would eg. fix the FSE hitbox differences, would any new combos arise?


#3

Oh, I found something else. FSE far MP is cancelable into specials (AKA SUPER MOVE) only when chained into.


5MPF_UC/CANCELS
S| E|  Type|     Cancel
=======================
1|11|ON_HIT|SUPER COMBO
7|11|  STOP|SUPER COMBO
1|11|ON_HIT|    CHAIN_M
7|11|  STOP|    CHAIN_M
1|11|ON_HIT|   CHAIN_MP
7|11|  STOP|   CHAIN_MP

5MPF_UC_ONESIDE/CANCELS
S| E|  Type|     Cancel
=======================
1|11|ON_HIT|SUPER COMBO
7|11|  STOP|SUPER COMBO
1|11|ON_HIT|    CHAIN_M
7|11|  STOP|    CHAIN_M
1|11|ON_HIT| SUPER MOVE
7|11|  STOP| SUPER MOVE

I’m amazed that the programmers thought it would be okay to just copy/paste this stuff 3-10 times and then tweak it later without expecting some bugs to pop up.


#4

There seems to be some confusion on the specifics of Juri’s dash properties (in other threads), so I’ll share some info on that. The last 2 frames are cancelable into REVERSAL, which includes all of Juri’s grounded specials, supers, ultras, and focus attack. These can also be inputted early and still cancel the dash as soon as possible.


DASH/CANCELS
 S| E| Type|  Cancel
====================
12|18|START|REVERSAL
16|18| STOP|REVERSAL


"Mmm... You look like you're in good shape": Juri Q&A and General Discussion Thread
#5

What does this part mean stated in a simpler way? Also, was this always the case in Super or was it added/slightly enhanced for AE?

For example, I find that FB, fadc Ultra1 has become easier in AE.Yet still not the easiest thing to do.

But while practicing the timing for the followup after activation without a dash, I find it as hard to do as it was in Super. For example after blocking Guy’s run+slide using a reversal Ultra1.
(a good way to practice FSE+ follow up, he’s -10)


#6

There’s some kind of input leniency, similar to the reversal window on wakeup. I’m not sure exactly, but I think you can input anything (except normals) from around 10-12 frames into the dash, and it will come out as soon as possible.

Edit: Nothing was changed in AE as far as I can tell.


#7

I think the reversal window was made larger on the forward dash cz something like that was mentioned on a dev blog.


#8

Hey Dr. Grammar. Have you looked at the data for forward throw? If so, did you see any funny business going on in there, like some kind of buffer at the end? The reason i ask is I was trying to figure out what the frame advantage is like and Im getting different results depending if I do nothing and then meaty, vs dashing once then do nothing and then meaty.

Seems to me that you can dash before you can jump (aka move). Im probably just confusing myself somewhere but maybe there is something to it.

edit: Eh I think there is a buffer, but Im thinking the dash comes out at the same time that you would have been able to move. I think the buffer is up to 6 frames, and it has to be “made up” before doing the meaty if used. Still interested if you see anything confirming this or something else.


#9

Forward throw recovery isn’t cancelable with anything, so it’s probably just the input buffer throwing you off.


#10

Well something is going on for sure. If its just an input buffer, thats a huge input buffer. Heres what the frames looked like.

Throw straight to meaty

  1. 2 frames fwd throw (where the buttons were pressed held for 2 frames)
  2. 109 frames (the amount of frames you must wait until you can move) read: throw animation duration
  3. 82 frames (the amount of frames before inputting the meaty)
  4. 2 frames cr. LK (hit meaty +8)

Throw, dash fwd, wait, meaty

  1. 2 frames throw (where the buttons were pressed)
  2. 97 frames (least amount of time to wait and still dash fwd)
  3. 2 frames (fwd input)
  4. 2 frames (stick neutral)
  5. 2 frames(fwd input, steps 3,4,5 are the dash input) steps 2+3+4+5 = 103 frames
  6. 82 frames (same amount used in other example) dash frames included
  7. 6 frames (the needed “padding” that needs to be made up) 103+6=109 frames
  8. 2 frames cr.LK (hit meaty +8)

Anyways, the conclusion I came up w/ is +88 after fwd throw, but that doesnt add up when compared to fwd throw, dashx4 into meaty.

edit: nevermind I just tested the fwd dash x4 and its the same. When done perfectly you have to wait 6 frames in order to do the meaty otherwise it will be early. So 19+19+19+19+6+6=88. I dont know the 6 frames is because of the buffer or what but at least it adds up.

So after fwd throw Juri is @ +88. After sweep, +44. After back throw +47. She has some normals that set up some good stuff like perfectly meaty cr.MP and perfectly meaty st.LP, almost perfect meaty throw, etc.

eh IDK, throw, dashx4 into perfectly meaty cr.MP is possible which would indicate +86. LOL okay Im gonna stop, im just confusing myself. Shes somewhere between 86-88.

final edit: I tested the frame advantage again but this time I didnt include a dash like I did before.It seems in all the scenarios, depending on if there is a dash or not, makes a 2 frame difference to my numbers. So my final conclusion is
sweep +44/+46
fwd throw +86/+88
didnt test the backthrow again as you will probably need a dash anyways so +47


#11

Hey Doc Grammar, Are you still around the Juri forums? If so I got a question for you. The frame data for FB release is as follows: 11/52/33 (start/act./rec.). So if the FB hits on the first possible frame, it leaves Juri @ +4/+10. If I can make the FB hit on the 31st frame counting from the 1st frame, am I correct on thinking that I’ve just “added” 20 frames of frame advantage? :wow:


#12

Yeah, I’m still here (even though I kinda hate this game now). Like with any move that hits meaty, you will get more frame advantage for having it hit later, so if it hits on the 31st frame, you get an extra +30 frame advantage.


#13

Thats what I thought but Im just making sure. It seems to good to be true that I almost dont believe it. Juri is gonna be a fucking beast. I’ve been finding set ups that, together with the combo-ability of 2012 PW’s, shes gonna open people the F’ up. True 3 frame DP safe jumps that cannot be beat by delaying .
They must be blocked and in some cases they cant even fadc them if they attempt them. Akuma cant even teleport out.

She’ll be able to be as much as +24 after a blocked fireball so she can go into OH with no risk at all. She’ll be able to not only activate FSE, but be able to go straight into her string with no fear of being reversal’d in between.Sometimes even using the FSE activation to make the reversal whiff haha.

Eh, I better calm down lol. Its still needs to be tested in 2012.


#14

She can already use meaty fireball setups to make an overhead “safe”, though they can still (character dependant) just reversal through the fireball and hit you. I know she has a 3frame safe jump off of techable knockdowns that push them far enough away, but they arent automatic you need to walk forward which makes it too hard to do.


#15

Shhhh. Mike. Let the man get hype. He’s making me hype.


#16

Mmmm…FSE loops, the thought of that makes me quiver with excitment — shame we can’t – not yet anyway.