The reverse up-B's that the bros. have are well-known but rarely used
inb4 star king uses it every day and how could I not know that
lol you know me well. I actually do use it. I remember when I went on my month long-hiatus and then went back and read some old posts, I saw everyone being all "lol so little gain for such risk who would ever use this" and I silently raged. Silently because I didn't want to reply to such old posts.
I think Boomfan might do them too, don't quote me on that though.
I only started doing it late in my keyboard career though, because I didn't know about it for a long time. I don't think it's THAT well-known.
no knowledge how to do them consistently
I know how. This is why it took me a long time to start doing them on keyboard. I had to figure out how to do it myself because I couldn't find any post on how to do it, and the description of
this video was confusing/misleading. Anyways, you at the start of the up-b you have to move the joystick backwards and then immediately forwards. So if you're facing right, the joystick will be up (to do the up-b in the first place) -> left -> right.
My hypothesis is that you're reversing it with the backwards input like usual, but you cancel the reversing with the forward input while you're still in the window for determining up-b direction and the result is some weird in-between motion. Maybe Sangoku can tell us how many frames the window for determining up-b direction so we know how much time we have. Pretty please ;o
It's not that easy because you have to do it pretty fast, but not that hard either once you get used to it/know the timing.
I'm not sure about this, but since the input delay was already counted in the frame perfect case, adding 4 frames of hitlag just switch the whole thing and the delayed frame is already counted. I think you're counting this frame twice: one for the no-opponent case + one for the hitlag case.
Apologies if the following is confusing.
If I understand correctly, hitlag freezes you in the state you're in when you land the hit for a certain number of frames. So the window for jab-grab is extended by hitlag if jab hits in first frame hitbox is out, because in the first frame the hitbox is out you can cancel the jab with the grab, so that state is extended by hitlag.
The reason I think it's 4 can be explained by extending your scenario. You posted this is what happens without hitlag:
Well, had he come up with proofs, I'm sure everybody would have believed him. And to be honest, I don't think anyone will care/listen what I found either. These weird things only raise the interest of the one who discovers them, such as aa back then or me right now.
I, for one, like weird things like that