Cool Stuff That Isn't a Combo

[phoenix]

I've done some basic testing, I can't do it consistently, but I'll go along with blocking giving you an input buffer.

I'm technically on that explanation too, but I'm still trying to figure out why it doesn't work if you hold the button, but DOES work if you press and release within a frame. It seems like some kind of property of the way negative edge works is the actual key.

[TomSinister]

I've done some basic testing, I can't do it consistently, but I'll go along with blocking giving you an input buffer.

I'm technically on that explanation too, but I'm still trying to figure out why it doesn't work if you hold the button, but DOES work if you press and release within a frame. It seems like some kind of property of the way negative edge works is the actual key.

Perhaps there is a transition frame between the last frame of recovery of a normal to crouch that is not cancleable by negative edge moves. Has anyone tried to use this OS by only using negative edge on the frame after a whiffed normal's recovery?

[phoenix]

Perhaps there is a transition frame between the last frame of recovery of a normal to crouch that is not cancleable by negative edge moves. Has anyone tried to use this OS by only using negative edge on the frame after a whiffed normal's recovery?

Nope that can't be the solution. The implication of that is, that the cl.MP into MP SRK link (which is a 1 frame link) wouldn't work if you'd do it with a single button press, using the negative edge. But you can.

This macro simply works:
w60,_2,w30,dr.d.dr._d_l^2,w30

[phoenix]

Nevermind guys, yeb's figured it out:

http://youtu.be/wZ94GdfhPHM

I'm pretty sure he's right.

[TomSinister]

thanks for the link! that was a good watch

[phoenix]

I've been trying to recreate his claims with macros, but actually can't get them to work at all.

As I see it, it still seems like it is a 1 frame timing.

All the following things work, but as you might see, the timing between going to down back and pressing HK might differ, but for ever frame that you make it shorter, the time that you hold the button must become longer. Those two are linked. Holding the button doesn't display any increase of leniency as far as I can see.

w20,1,w30,d3.dr.r.d.dr.r._d_l,w12,6,w30
w20,1,w30,d3.dr.r.d.dr.r._d_l,w11,_6,w2,^6,w30
w20,1,w30,d3.dr.r.d.dr.r._d_l,w10,_6,w3,^6,w30
w20,1,w30,d3.dr.r.d.dr.r._d_l,w9,_6,w4,^6,w30
w20,1,w30,d3.dr.r.d.dr.r._d_l,w8,_6,w5,^6,w30
w20,1,w30,d3.dr.r.d.dr.r._d_l,w7,_6,w6,^6,w30
w20,1,w30,d3.dr.r.d.dr.r._d_l,w6,_6,w7,^6,w30
w20,1,w30,d3.dr.r.d.dr.r._d_l,w5,_6,w8,^6,w30
w20,1,w30,d3.dr.r.d.dr.r._d_l,w4,_6,w9,^6,w30
w20,1,w30,d3.dr.r.d.dr.r._d_l,w3,_6,w10,^6,w30
w20,1,w30,d3.dr.r.d.dr.r._d_l,w2,_6,w11,^6,w30

Interstingly, the following doesn't seem to work:
w20,1,w30,d3.dr.r.d.dr.r._d_l_6,w12,^6,w30

I think you basically input the negative edge a frame early, and that changing state for some reason increases the buffer by 1 frame. But not by multiple frames as yeb seems to imply.

[Edit]

Nevermind, I figured it out. The trick is doing the motion as late as possible. Then you have ample time in the buffer to screw around.

Compare

w20,1,w30,_d3.w10,r.^dr.d.dr.r._d_l,_6,w3,^6,w30
w20,1,w30,_d3.w10,r.^dr.d.dr.r._d_l,_6,w6,^6,w30

Both work

but also slightly earlier input works:

w20,1,w30,_d3.w8,r.^dr.d.dr.r,w2,_d_l,_6,w3,^6,w30

[TomSinister]

Nevermind, I figured it out. The trick is doing the motion as late as possible. Then you have ample time in the buffer to screw around.

yeah, I thnk Yeb kind of glossed over this fact very briefly. I wonder this OS seems to benifit hitbox players more than stick players, as they can input motions much much faster.