I m still investigating to understand How recording vcam could be cleaner (having only one transform relative to world).
The solution I found was, creating a sequencer with the recorded Vcam + a new cam (I could set as I want). And then attch the new one to the recorded
