Dustin Cross
Well-known member
- Joined
- Mar 6, 2007
- Messages
- 1,240
- Reaction score
- 0
- Points
- 0
- Location
- Austin, Texas
- Website
- www.sandust.com
Aloha,
I started talking about this is a troubleshooting thread, but I thought I should make a separate thread under 4K Workflow so more people would see it and hopefully benefit.
http://www.reduser.net/forum/showthread.php?t=13503
----------------------------------------------------------------
I had a new project that I needed to batch convert a bunch of Red footage to Prorez for the editor. This time I needed audio, so I did a little research and found these little command line tools for Quicktime.
http://www.3am.pair.com/QTCoffee.html
QT Coffee is donation-ware for personal and academic use and $10 for commercial. It allows you to modify Quicktime readable files in many different ways.
So I modified my little script from above to use Redline to create video files and use QTcoffee's muxmovie to combine the video track from the Redline created file with the audio and timecode tracks from a proxy file.
Here is the modified script:
for FILE in $(ls */*001.R3D | awk -F "_001.R3D" '{print $1}'); do redline -i `echo $FILE"_001.R3D"` -w 11 --resizeX 1920 --resizeY 1080 --fit 2 --QTcodec 0 --useMeta -G 1 -D 100 --NR 100 -x -.4 -T 1.2 -C .5 -o $FILE"_VIDEO_ONLY.mov" ; echo "combining audio and video for $FILE" ; /usr/local/bin/muxmovie $FILE"_VIDEO_ONLY.mov" -track "Video Track" $FILE"_P.mov" -track "Sound Track" -track "Timecode Track 1" -track "Timecode Track 2" -self-contained -o $FILE"_AV.mov" ; done
I have made a couple different changes.
First, I noticed that some of the files were taking WAY too long to render. I realized that I was originally telling Redline to convert *.R3D. This ment that on really long takes, files that had been broken up to multiple 2GB files, Redline was rendering the entire movie for each .R3d segment. So I changed it to only look at *001.R3D.
Next I added a little awk script that removes the end of the filename (_001.R3D) so I can tell the script to work with the R3D file or the _P.mov file or whatever.
Then there is the Redline part. You can get all the arguments from Redline at the command line, or you can use Red Alert and copy to clipboard to get the settings you want to use. These are setting that seem to give me an acceptable overall image for everything I have shoot. I can tweek thesse parameters, but this is really just to get me something acceptable I can edit with, fast.
Lastly I use muxmovie from QTcoffee to combine the video from the movie Redline just made with the audio and timecode from the _P proxy. I could have used any of the proxys, but I just used _P.
This gives me Video and Audio in my Redline rendered file, ready to drop in FCP and start working.
So far this is working out great for me. Hopefully this info will be useful for others.
Please let me know if you have any questions.
Mahalo,
Dusty
I started talking about this is a troubleshooting thread, but I thought I should make a separate thread under 4K Workflow so more people would see it and hopefully benefit.
http://www.reduser.net/forum/showthread.php?t=13503
----------------------------------------------------------------
I had a new project that I needed to batch convert a bunch of Red footage to Prorez for the editor. This time I needed audio, so I did a little research and found these little command line tools for Quicktime.
http://www.3am.pair.com/QTCoffee.html
QT Coffee is donation-ware for personal and academic use and $10 for commercial. It allows you to modify Quicktime readable files in many different ways.
So I modified my little script from above to use Redline to create video files and use QTcoffee's muxmovie to combine the video track from the Redline created file with the audio and timecode tracks from a proxy file.
Here is the modified script:
for FILE in $(ls */*001.R3D | awk -F "_001.R3D" '{print $1}'); do redline -i `echo $FILE"_001.R3D"` -w 11 --resizeX 1920 --resizeY 1080 --fit 2 --QTcodec 0 --useMeta -G 1 -D 100 --NR 100 -x -.4 -T 1.2 -C .5 -o $FILE"_VIDEO_ONLY.mov" ; echo "combining audio and video for $FILE" ; /usr/local/bin/muxmovie $FILE"_VIDEO_ONLY.mov" -track "Video Track" $FILE"_P.mov" -track "Sound Track" -track "Timecode Track 1" -track "Timecode Track 2" -self-contained -o $FILE"_AV.mov" ; done
I have made a couple different changes.
First, I noticed that some of the files were taking WAY too long to render. I realized that I was originally telling Redline to convert *.R3D. This ment that on really long takes, files that had been broken up to multiple 2GB files, Redline was rendering the entire movie for each .R3d segment. So I changed it to only look at *001.R3D.
Next I added a little awk script that removes the end of the filename (_001.R3D) so I can tell the script to work with the R3D file or the _P.mov file or whatever.
Then there is the Redline part. You can get all the arguments from Redline at the command line, or you can use Red Alert and copy to clipboard to get the settings you want to use. These are setting that seem to give me an acceptable overall image for everything I have shoot. I can tweek thesse parameters, but this is really just to get me something acceptable I can edit with, fast.
Lastly I use muxmovie from QTcoffee to combine the video from the movie Redline just made with the audio and timecode from the _P proxy. I could have used any of the proxys, but I just used _P.
This gives me Video and Audio in my Redline rendered file, ready to drop in FCP and start working.
So far this is working out great for me. Hopefully this info will be useful for others.
Please let me know if you have any questions.
Mahalo,
Dusty