Where did these lines come from?
 

I am trying to export a video and for some reason every time I export I get some really annoying lines across the bottom of the video. Have a look: http://www.rpmproductions.info/videos/lines.mov

Here is my workflow: shot on 5DII, converted to ProresHQ via MPeg Streamclip, Edited in FCP6, stylized and color graded in FCP with Magic Bullet (look suite unbloom), exported video without recompressing frames.

This particular clip was then converted to mp4 but then trimmed using QT, but the lines also exist in the uncompressed Prores export from FCP which is weird because the lines do not exist anywhere in my FCP timeline.

I think that magic bullet may be the culprit here, but I need the video to maintain this look. The only way that I could get the lines to disappear was to export via H264, but my client is on a windows machine and hasn't been able to view a single H264 file that I have sent to him yet.

I need the export mainly for preview purposes, but I am affraid that the lines will show their ugly head when I convert to beta. Any help is greatly appreciated.

That's the rolling shutter attempting to chop up your image. OK just kidding...

Why is MPEGStreamclip part of the workflow? I've seen MPSC create those same lines when making an MPEG-2 encode before. I don't think it's MB at all; take MPSC out of the equation and change your workflow.

I actually found out what is going on. The footage was still hanging onto an upper field dominance all the way through my workflow. No matter what I was exporting through I was still getting these lines. I finally just de-interlaced on export from MPSC and problem solved. No more "interlaced" lines. Still don't comletely understand the whole upper field dominance of the 30P footage of the 5DII. Even when I try to transcode my footage to PreresHQ as progressive it still hangs onto the "interlacing". If anyone had an explanation for this I would be all ears.

I use MPSC because it is ten times more efficient than transcoding via compressor every time. I always get incredible results using MPSC and don't see why it would be the program since it uses QT to encode anyway. It just gives an easy way to batch the transcodes.

I'm probably way off with this guess...

Did you realize that HD is always up field first. SD is always lower field first. Maybe you mixed it up somewhere?