Creative Communities of the World Forums

The peer to peer support community for media production professionals.

Forums Compression Techniques 16:9 DVD to Web

  • 16:9 DVD to Web

     Daniel Low updated 12 years ago 4 Members · 10 Posts
  • David Dube

    September 5, 2008 at 9:57 pm

    Hi everyone. First, thanks to all the contributors on this forum. It is a priceless source of information. I have figured a workflow from 16:9 DVD (taken from a DVD-R camera) to Web diffusion. I have to edit the content on the DVD. I just want to see if I’m doing things right.

    Rip the DVD.
    Import VOBs in MPEGStreamclip.
    Fix timecode breaks.
    Export to Motion JPEG or Apple Intermediate CODEC (would DV50 do it also?) at 854x480px and deinterlace.
    Import into FCP and Edit.
    Save reference clip.
    Compress in Compressor using H.264 320×240 or 640×480 with Letterboxing or Crop

    Does that sound right?

    Cheers,

    David

  • Ed Dooley

    September 7, 2008 at 1:58 am

    Any of those would work (and so would photo-JPEG at 75%), although I tend to export to the highest quality (uncompressed if possible), and de-interlace at compression. Why the final compressed sizes? It should be 320×180 or 480×270 or 640×360 for 16×9.
    Ed

    [David Dube] “Hi everyone. First, thanks to all the contributors on this forum. It is a priceless source of information. I have figured a workflow from 16:9 DVD (taken from a DVD-R camera) to Web diffusion. I have to edit the content on the DVD. I just want to see if I’m doing things right.

    Rip the DVD.
    Import VOBs in MPEGStreamclip.
    Fix timecode breaks.
    Export to Motion JPEG or Apple Intermediate CODEC (would DV50 do it also?) at 854x480px and deinterlace.
    Import into FCP and Edit.
    Save reference clip.
    Compress in Compressor using H.264 320×240 or 640×480 with Letterboxing or Crop

    Does that sound right?

  • Daniel Low

    September 7, 2008 at 10:45 am

    Using uncompressed is rather pointless unless the sourse is uncompressed, it wastes disc space and saturates bandwidth and can slow down the transcoding process.
    512×288 is also a good 16:9 size as it’s div 16.

    __________________________________________________________________
    Please post back saying what solved your problem. It could help others, and saying ‘thanks’ is free!

  • Ed Dooley

    September 7, 2008 at 1:06 pm

    I guess it depends on what you mean by “wasting disc space”. If you have the space it’s not wasting anything. But more importantly, using UC has more important benefits. First, if you’re starting with an already compressed source (MPEG-2 in this case) why would you want to compress it even more, before you do your web compression? In my experience compression looks best when you start with the highest quality source. You wouln’t export video from FCP to an intermediate compression before sending it to Compressor, for example. You would choose Use Current Settings, which doesn’t recompress the video. Second, if you’re adding any graphics at all to the video in FCP, using the highest quality video will give you the best looking graphics. As I said in my post, they’ll all work. If you have the space, go UC, if not use any of the others (I would choose Photo JPEG as my 2nd option).
    Ed

    [Daniel Low] “Using uncompressed is rather pointless unless the sourse is uncompressed, it wastes disc space and saturates bandwidth and can slow down the transcoding process.
    512×288 is also a good 16:9 size as it’s div 16”

  • Daniel Low

    September 7, 2008 at 6:27 pm

    If you have the disc space AND the time then it won’t hurt although you may get some inconsistencies in FCP unless it’s cuts only. (Don’t forget you’ll be doing a crude [and pointless] 4:1:1 to 4:2:2 colour space conversion which may introduce it’s own errors)

    Just for reference:
    DV25 is 3.60MB/sec – (The picture has already been trashed…)
    Apple U/C 8bit is 20MB/sec (The same trashed picture, just massively bloated)

    So you’ll need well over 5 x the disc space just for that bit of the process.

    An in my opinion, as this is going from DV to web (via FCP), you’ll not notice the difference in quality. If you’re adding graphics or DVE’s and so on in FCP, it’ll look messy whatever way you go.

    Out of interest, what uncompressed CODEC do you like to use Ed?

    __________________________________________________________________
    Please post back saying what solved your problem. It could help others, and saying ‘thanks’ is free!

  • Ed Dooley

    September 7, 2008 at 6:49 pm

    But it’s not trashing it more through added compression. Going to 4:2:2 is exactly the reason why added graphics & text will look better. I realize that everyone has their opinion, and there are lots of them. In my experience, it’s far better looking added graphics that make up my mind. And BTW, that’s why I can’t wait to start using vector graphics on top of Flash video. I haven’t played with it yet, but having H.264 video in Flash and then putting beautiful lossless vector graphics over that looks amazing. I use AJA’s 8 and 10 bit UC, and I also try to never edit anything from a DVD. 🙂
    Ed

    [Daniel Low] “If you have the disc space AND the time then it won’t hurt although you may get some inconsistencies in FCP unless it’s cuts only. (Don’t forget you’ll be doing a crude [and pointless] 4:1:1 to 4:2:2 colour space conversion which may introduce it’s own errors)

    Just for reference:
    DV25 is 3.60MB/sec – (The picture has already been trashed…)
    Apple U/C 8bit is 20MB/sec (The same trashed picture, just massively bloated)

    So you’ll need well over 5 x the disc space just for that bit of the process.

    An in my opinion, as this is going from DV to web (via FCP), you’ll not notice the difference in quality. If you’re adding graphics or DVE’s and so on in FCP, it’ll look messy whatever way you go.

    Out of interest, what uncompressed CODEC do you like to use E”

  • Kris A. Wotipka

    September 14, 2008 at 4:38 pm

    Ed,

    I just finished a Flash project that was heavy on video. It was your typical Powerpoint slide grind but we didn’t want the horrible DV quality screen transfers that are often associated with type of project. We shot the video greenscreen and converted it to .flv with cuepoints for the screen changes. Now, we had to go through some pretty extreme compression (6 hours of video on a CDROM plus all of the flash content). At the end of the project, the video looked pretty great but the graphics were excellent (recreated the slides in flash). All of the video compressed well except for one gentleman who’s head didn’t seem to want to divide by 16.

    So the client was happy and I am using this on a few other projects so the setup has paid for itself. However, I am like you and can’t wait to take advantage of vector in 264. Maybe in FCP 8 or 9.

    kw

    kris@wotipka.com
    Image maker

  • Ed Dooley

    September 14, 2008 at 4:56 pm

    Wow, 6 hours on a CD! That’s some compressed video. Why FCP8 or 9? The vector graphics overlay the H.264 video in Flash (whatever authoring tool you use). FCP isn’t a factor at all.
    Ed (who talks about it, but still hasn’t tried it)

  • Kris A. Wotipka

    September 14, 2008 at 5:01 pm

    Hi Ed,

    I was speaking of some of the more advanced features of H.264. It has been a while since I read through the spec but I seem to remember it being able to handle a couple of layers of content (ie. like multitrack audio but to the animation layers).

    The FCP 8 or 9 comment was more to the:

    “Won’t it be nice when you can have your video (as video) on one layer in H.264 and then have the graphics on another (vector) layer which would be rendered back in as vector on top of the video stream on the client’s end.” How’s that for a run on sentence.

    kw

    kris@wotipka.com
    Image maker

  • Daniel Low

    September 14, 2008 at 5:28 pm

    FYI:

    Quicktime could do this years ago and is part of the reason Quicktime was chosen as the file framework format for H.264.

    Shame it got lost in the background before anybody really took advantage of it.

    https://developer.apple.com/documentation/QuickTime/RM/LegacyAPIs/Vectors/rmVectors/chapter_1_section_1.html

    __________________________________________________________________
    Please post back saying what solved your problem. It could help others, and saying ‘thanks’ is free!

Viewing 1 - 10 of 10 posts

Log in to reply.

We use anonymous cookies to give you the best experience we can.
Our Privacy policy | GDPR Policy