-
XDCam HD Program Downconvert to SD for broadcast?
I have done my homework and I’m still confused. What’s the latest thinking on properly down converting a program shot with Sony XDCam 1920×1080 35mb/s for broadcast in 720 (4×3) with no letterbox and making it look good?
I shoot PMW-350 and EX1R cameras and edit on FCP with either ProRes 422 or HDCam EX 1920×1080 35mb/s VBR codecs. For HD, I just output a Quicktime and layoff to HDCam tape. Easy.
For downconverting to SD (for broadcast), I’ve tried a few experiments with StreamClip but have not come upon the perfect solution. Doing research has just made me more confused. People have actually output video to memory cards and used the camera hardware to generate the downconvert. Really? I have not tried that solution for BitVice yet but I’m wiling to do anything to make this workflow manageable.
Can someone walk me through what they do using either StreamClip or BitVice starting with a QT source file? Is there some other solution I should be testing?
Any guidance from you pros is much appreciated!
Cheers!
Steven
PMW-350
EX1-R
Nikon D3
(wlstevens.com) -
6 Replies
-
Thanks Craig. I saw those articles in doing my homework. They were helpful in identifying the source of the problem with XDCam but I still need a decent workflow suggestion.
StreamClip is producing files that look stretched and distorted so I’m not even at the stage where I’m trying to refine the look. I’m doing something wrong and just wondered how other people are managing.
PMW-350
EX1-R
Nikon D3
(wlstevens.com) -
The fast(er) and simple way:
1. Edit your HD project in an HD timeline.
2. Apply the flicker filter set to MAX on everything
3. Render and export a self-contained HD movie
4. Create an SD non-anamorphic sequence in FCP.
5. Drop your HD movie into the SD sequence.
6. FCP will automatically letter box and pillar box the movie.
7. Drag a corner of the frame in the canvas until the top and bottom of the frame is filled.
8. you’ll lose the sides of the frame but now it’s in the correct 4:3 aspect.
9. Render and export.
10. Go for tea/coffee.Post Script: the slow way is to create the SD non-anamorphic sequence and drop each clip into it and repeating the step in number 7. Then apply the flicker filter set to MAX, render and export. This takes up less storage space because it’s only one render stage but it’s very tedious.
– Don
Don Greening
A Vancouver Video Production Company
Reeltime Videoworks
http://www.reeltimevideoworks.com -
Don: Your solution seems waaaay too easy and sane. So no multiple transcoding steps using five different software processors running on 2 networked machines? No exporting to hardware and then re-importing and rendering thrice? No reciting ancient Mayan sacrificial incantations over a boiling cauldron of shark fins?
Given how you so skillfully solved my other nagging problem regarding the XD Transfer software a few weeks ago, I’d say you were bucking for a consultancy gig or something. Next time we are needing some backup in that department, you wil be the first call!
Many thanks Don.
Steven
PMW-350
EX1-R
Nikon D3
(wlstevens.com) -
Before you start reciting more Myan poetry just go through the entire process from start to finish with a 5 min. test piece. See if the result meets with your satisfaction. Also the jury is still out on whether it’s more beneficial to apply the flicker filter before the down
convert to SD or after it’s in the SD timeline. I haven’t done any tests myself but the result may be different.– Don
Don Greening
A Vancouver Video Production Company
Reeltime Videoworks
http://www.reeltimevideoworks.com -
We have tried this method. There is ghosting of images which we can’t get rid of, particularly when there is movement.
Is there a difinitive hardware solution to this? Is the AJA Kona card being used widely still? Any better options?
Thanks!
PMW-350
EX1-R
Nikon D3
(wlstevens.com)
Reply to this Discussion! Login or Sign Up