I have done a movie for school shot with dv camera, meaning that my video material is interlaced. Then my friends have made cg material to that video material. I'm supposed to compress this. I have cg progressive(p) and interlaced(i) video material mixed together, mostly video material. I know that you should avoid mixing progressive with interlaced material. But that will have to do for this time. I have tried different deinterlaced methods and are wondering which would be best to use for p and i material mixed together.
Blending:
When I blend the material everything looks good, except for some scenes. That is when two fields gets seperated by fast changing scenes. Then you see two distinct seperated fields blended together. Otherwise the video material looks really nice.
Discard field then resize
First I have to get understood what this means. When you discard the fields do you take away every other field? Doesn't this mean that you lose half of the vertical resolution? Does it show the same field twice (p) instead of two diffrent (i)? Beacuse it looks like it has dropped resolution in the cg material when I resize it back to 720x576 (pal). I see jagged corners in only the progressive material, which I do not like to see. Does this mean that I have lost resolution.
When I compress the video to a 720x576 divx file it is very heavy. The framerate stays around 22 frames/s when it should be 25 (pal). Is it because of the high resolution. Can you prevent this without losing resolution. It seems like the bitrate doesn't matter.I get the same results if I use more data/s.
What do you recommend me to do?
Ps. Do not say read ermacs guide. I have, believe me. Just don't fully understand the Fields vs. Frames part.
