Hasta La Motion!

Ok, this is just my curiousity. If someone ever really work with tons of nodes…. i mean really using a lot of nodes, I want to know how my workflow will affect the performance. So i’d like to focus on rotoscoping. I ussually masking the clip using 2 nodes : attach a clip to a merge node , set mode to ‘in’ then pipe the a rotopaint node into B-input of the merge node (this is the first method). That way i can get the same result compare to using just single rotopaint node, apply directly to a clip then set the proper parameter (this is the second method). For me the first one offers greater flexibility and easy to read compare to the other which have benefit of using just single node (less proccessing??).

I was shake user, so there’s why i ‘m familiar with the first method ( where the second method is not an option in shake). How this affect performance in a hugh project since i’m using 2 nodes, compare to other method with single node. Should i change the workflow? Or maybe technically or internally they’re just exactly the same in REAL cpu proccessing? Any real experience on this ?

…… btw i have a very simple question in head, just another thing…. why Nuke won’t give me warning when rendering to existing files? any good reason?

thanx.