I have footage from 22 GoPro cameras, mp4, which I have imported natively into AE (11.02, OS 10.8.3 Mac 2.66mhz Quad Tower 16g of RAM) into 1 composition. I plan to achieve the effect desired and then output a production codec, 422 prores for inclusion in my Premier project timeline. Is this a valid workflow?
Next the compositional question:
Most of the 22 cameras (in an array) recorded two or more files of the 16 mintes of run-time, (there are 6 separate takes in that time period). I pre-composed those footage files (after aligning a slate and for the seamless play of each layer) to neaten up and make more compact my AE timeline.
The effect I'm trying to achieve is a Time-Slice/Bullet Time where I can "cut" from one camera to the next all along the array, thereby freezing the action, but having the "camera" move around the subject - and then have the last layer/camera keep playing. At the precise frame I want to freeze I put the opacity of each layer to 0 successively at the desired "freeze point" however in my ram preview only the first footage layer plays. Without the opacity transformation on each layer I can achieve a rough version of what I'm trying to do by Solo'ing each individual layer and I cannot figure out why making each layer 0 opacity successively isn't working.
Also once I completed all the Opacity transformations, the bottom layer should be visible 1 frame after the layer above it goes to O opacity, but it is in fact BLACK! What the heck?
Do I need to slip each layer 1 frame forward in time so only 1 frame of each layer plays? Again "turning on and off" each layer with Opacity? I've found the answer to this question, at least for now. Each successive clip gets trimmed to the desired frame, then slipped one frame in front of the layer above (save the top layer, which is trimmed to the frame). This way the same frame is played over and over from each camera along the array.
Have I made a mistake by using the Pre-comps mixed with the 1-take footage? In other words, my main comp now includes Pre-comps of the 2 files produced by most of the cameras, along with the original mp4 footage of those cameras which only recorded one file for the action.
And further; once I have the frames visible in the correct order would I Pre-compose again and apply a sfx like pixel motion to interpolate between the difference in angle/frame due to the difference in each camera's position? Please suggest an alternative workflow or compositional layout!
I hope I've been clear, I'm here and willing to clarify if I've omitted any details.
I still would appreciate answers to the remaining issues if anyone has input for me. The question now is: to time remap, or timewarp (which one?) before Pixel Motion - and is Pixel Motion best for frame interpolation to smooth it all out?