I’m attempting my first video. The goal is for 5 minutes length (9,000 frames at 30f/sec). Animated camera, grasses, trees, soundtrack. Also included are displaying a video on an animated plane. I’ve put groups of things on several layers due to the complexity. What I want to know is: Is there a way to view specific parts of the video so I know what needs to be changed without rendering the entire animation every time? Example, if I need to move an object to prevent it from merging through a tree, I’d like to do that in a couple of minutes instead of 2 days later. (assumed render time)
Couple of basic ways to do this. First off, I would watch the animation playback in the 3d view, looking for trouble areas. Note the frames where there might be problems, so you can focus on those frames. Yes, the 3d view playback can be slow, but you can speed it up. Search this forum section for tips on improving 3d view performance, it’s been asked many times. Basically, improving 3d view performance boils down to this: eliminate everything that you don’t need. You don’t need textures to tell if an object clips another one. You don’t need subsurf modfiers, and you don’t need to see objects that you know won’t be problems. Just focus on the objects that might clip others. You mention having things separated out onto different layers, so don’t view the layers that have stuff that won’t be a problem. I think watching a payback at 5-10 fps can help you spot problem areas like object clipping each other better than viewing it at 30 fps.
If you want to see the whole animation at 30 fps and not wait for it to render, then do an OpenGL render of the animation from the 3d view. From your camera view in the 3d view, click on the clapboard icon in the header will render the animation out as seen in the 3d view. It will be rendered according to the render settings as far as render size and file format. Easy, quick way to render an animation to a video format to be viewed by a video player.
When rendering the final video, render each frame as a single image, then combine the images and audio into a video format, using blender VSE. It’s pretty easy to do, and this way if you find a problem after final render, you can correct the problems, then only render the frames affected, instead of the whole thing.