Am I understanding this correctly....I should export to individual images - there for I will have 25 images per 1 second (for PAL). Then I merge them in an external program such as VirtualDub. Is this correct? Just to reafirm what I want to achieve. I want to overlay a model that is walking over a video scene. In the video, there is a doorway, and I want to make it appear the model walks in the doorway. I thought I could uses a mask of some sort that masks out the models when it gets to the doorway. I hope this makes sense. Cheers.