Well, according to the command list, it is invulnerable to lows on startup, but it seems like there may be one or two frames where that doesn’t seem exactly true.
Using the hitboxes on the training mode, the move seems like instant invulnerable to lows. The high version definitely feels and looks like it, but the low version feels deceptively off by maybe a frame or two, but within a margin too narrow to confirm visually.
I have the ability to capture 60 fps footage with an Elgato, but I don’t have any video software to playback and analyze 60 fps footage on a frame by frame basis, otherwise I would test the startup sequence myself.
However, in training mode, I can get Fulgore to sweep me constantly on recording, and when I go to use it, I can get the reversal indicator to pop up on timing, but Fulgore still beats it every time. It seems there is a frame or two by which there is no invulnerability on startup, which makes the description misleading, and possibly the move not working as intended.
Then again, those frames may be in there by design, but I don’t see the reason to it.
Thinking about it, on my road to getting RAAM to level 50, this is a move I rarely utilitzed at all. The Anti-Air, upper body emergence and the full body version were much more reliable, but lower body does seem to be beaten out too easily.
Side Note: The buff to the Kryll Shield cancel sounds awesome, I think it’ll be much more useful now. However, according to the command list, it still mentions 1 bar as the cost, when it should be have a shadow meter.