Esoteric Frame Data

If you have the time could you test this: Gohs 46k sabakis mid and high kicks. what are the active sabaki frames on that move?

 

Ducking Pai standing
Pressing down
P1
on F10

P2
held from F22 --> hit
held from F21 --> ducked


pressed on F11-21 --> duck/DM
pressed on F22 --> hit
pressed on F21 --> hit
pressed on 11-F20 --> DM

Releasing down
P1
on F10

P2
on F0-22 (release on F23)--> not hit
on F0-21 (release on F22) --> hit

Compare with normal guarding
P1 Pai on F0
P2 Pai on F11 --> guarded
P2 Pai on F10 (release F11) --> hit

 

Goh sabaki
Goh's move completes on F20 ( input)

Akira input on F5-18 will be sabaki'd.

(grey 14-1, between and , where Akira also presses on F20 as a marker. )

 

VF is certainly funny in the way that it interprets commands.

For example (all inputs for a single frame each, unless otherwise specified). / will designate the end of one frame.

/ gives standing punch
/ gives sidestep
/ gives side step, as does ////// etc.
/ gives ducking jab
/ gives +
/ gives +

/ gives
// gives
// gives
/// gives


/ gives
/// gives

Only reason I went into depth in this is that I couldn't get a yoho to work with what I thought should be the command. If anyone is programming combos, it's worth being aware of. Not sure if it has an impact on buffering, probably not.

 

I also suggest to do it with bot evades and

 

Akira vs Akira

on F0
on F14 is earliest possible (nothing happens with input on F13).
Same applies to on F0.

Note however, this is buffered, so it is not executing from F14. 2P Akira inputting on F25 will double hit with 1P Akira F0 F14.
Same applies to on F0.

F0/F1//F3 gives no obvious evade, does give crouch dash. F4 will execute immediately (i.e. clash with 2P Akira F4.
F0/F2//F4 gives obvious evade cancelled into crouch dash. F5 will be beaten by 2P F5. 2P F12 will double hit with 1P F5.
Same applies to F0.

F0/F2//F4/F5 gives evade into . 2P done on F12 will double hit (as per above note, 2P input is F12/F13.
Same for F0 and F0.

 

Putting the CD and dash techniques against each other as you suggested Myke yields a double hit. I.e. the two techniques appear equally advantageous.

1P
F0//F2//F4/F5/F6

2P
F0//F2//F4/F5/F6

--> double hit

 

1. P2 El Blaze does + on F0 --> P1 presses on F17 --> blocked.

2. P1 does F0/. If P2 El Blaze does F8+ it is not blockable during the evade normally (tested by holding guard, and also it gives +9 on hit instead of +6.
If P2 ElB does F9+ it is blockable normally (holding guard and also -6 on hit. Checking with PStick, this means pressing on F26 for P1.

The same applies to pressing 1PF0.

Note, Blaze's + has at least 2 active frames (7 according to the guidebook). So with spacing, it is possible to evade a F8+ and block, however this is revealed by the fact that the frame data is +4 on guard, not +5. All testing I've tried to do at point blank range to eliminate this problem.

If P1 ElBlaze does F10 (F10 as I need to have 2P Blaze's + occur earlier) then he can block P2F4+ upwards. Interestingly, this is -1 on guard (normally +5, i.e. you are blocking the final active frame.).

2PF3+ will hit a F10//F12//F14/F15onward.


Using

Elblaze 2PF7+ can be blocked by 1PF10//F12//F14/F15onwards. A 2PF6+ is not blockable.

 

Testing against low full circular

2P ElBlze does F4+/F5. This can be blocked by 1PF10//F12//F14/F15onwards. Gives +18 on block.
2PF3 +/ cannot be blocked in time.

2P ElBlaze does F1+/F2. This can be blocked by 1PF10//F12//F14/F15onwards. +18.
2PF0 +/F1 cannot be blocked in time.

 

Yes 2P is just throwing out naked. The reason I looked into this is it explains how much time is spent in the ECD motion before an attack can come out, i.e. 12F.

 

Good job noodals but i have to admit most of the time it takes times to understand your notations .

If i resume your results i understand this :

/ means next action

// means the next action was input neutral aka aka nothing

You could also write this // instead of //

------

When you input something with player 2, you have to add the one frame of "lag". The game wait one frame to see your next input.
Input on F12 means actually start on F13

-----------

Normal evade took exactly 24 frames after you go in neutral to finish. You start acting (block/attack) 25 frames after you go neutral or 26 frames after release / (F0)

<div class="ubbcode-block"><div class="ubbcode-header">Quote:</div><div class="ubbcode-body">2P Akira inputting on F25 will double hit with 1P Akira F0 F14 .
Same applies to on F0.</div></div>

means akira start the on F26 (lag input from player 2)

<div class="ubbcode-block"><div class="ubbcode-header">Quote:</div><div class="ubbcode-body">If P2 ElB does F9 it is blockable normally (holding guard and also -6 on hit. Checking with PStick, this means pressing on F26 for P1.</div></div>

The example show that you can start blocking at F26 !


Those data applies to attack and defense

-----------

Dash evade take 12 frames to finish after you go neutral. You start acting (attack) 13 frames after you go neutral (if you do the dash fast enough of course)

<div class="ubbcode-block"><div class="ubbcode-header">Quote:</div><div class="ubbcode-body">F0/F2//F4 gives obvious evade cancelled into crouch dash. F5 will be beaten by 2P F5 . 2P F12 will double hit with 1P F5 . </div></div>

That means from player 1 actually start on F13 because of the lag from player 2 (he input on F12 but in reality it's on F13)

-----

You examples with the guard are strange.
It's look like you cannot block thought you can already attack. You cannot block on F13. This is what i understand from your example. More stranger is it's look like let you guard earlier that in some cases

is good versus high circular and bad versus low circular
is good versus low circular and bad versus high circular

That's it ?

Can you rewrite your examples in adding the frame where P1 block those attacks. It will help people figure out those cases

----------------

This confuse me a lot
<div class="ubbcode-block"><div class="ubbcode-header">Quote:</div><div class="ubbcode-body">/ gives
// gives
// gives
/// gives </div></div>

With Eileen 99% of the time when i try to do i endup doing and that piss me a looooot.
Your examples show that is hard to get. Can you do more case because i really don't understand. I am suppose to always hit before input the last ?

 

Did you do any tests where you actually input or instead of neutral at the end? Im saying this as I believe it is possible to duck the high circulars after evade crouch dash cancel instead of having to block them. I also think that at the same time you also block the low circulars.

I believe the old "you instantly duck highs if you are pressing down" is now extended to crouch dash also. My proof:

I set CPU to block and punish with P. I hit Jacky beat knuckle on block (-10) and buffer crouch dash = I will duck the punch. This wasn't the case in vanilla.

Obviously my character cannot be crouching from the first frame of CD or fuzzy would be bonkers, so I believe the "instaducking highs" is the cause.

Because of this, I believe evade CD cancel is superior to evade dash cancel.

 

El Blaze vs El Blaze

2PF4 +/
1P F10/./F12/./F14/F27 will block the +, giving -1 on guard. Any slower inputs (P1F26 etc. will not be hit, as ElBlaze will crouch dash under.)

If you hold P1F27 then he will duck under the +.

---

2PF7 +
1PF10/./F12/./F14/F24 will block the +. +5 On block.

Can't just hold to duck after this.

 

Is it possibile know how many frames are needed to cancel a successful evade into a offensive move? tnx

 

In the last 2 days I have been in the dojo doing a lot of cross tests to take apart the ECDG and the EFDG for an article on our italian blog and only after posting my results here on vfdc Cnul told me about your work Noodalls.
Not having a programmable stick I'm happy to see that my findings are confirmed by your professional work

Making it short:
-ECDG -> you can standing guard from the 17th frame onward
-EFDG -> you can standing guard from the 14th frame onward

ECDG however is faster for guarding low, being able to do it from the 15th frame onward.

About the > > and > > , as arleady proven by Noodalls, I found that you can cancel a failed evade in the forward dash and crouch dash after a minimum of 10 frames and that you can cancel the forward dash and crouch dash with an attack after a minimum of 1 frame making the two options identical.

 

noodals, could you test how many frames opponents are vulnerable to low throws after techrolling from different knockdown positions? face down head toward, face down feet toward, etc, and also the spinning knockdown like Gohs 3pp.

I'd like to know how tight the window is to catch people with low throws when they techroll. and also if they can attack out of this low throw or only have the option to tech.