Creative Communities of the World Forums

The peer to peer support community for media production professionals.

Activity Forums Apple Final Cut Pro Legacy DVCProHD to SD DVD

  • DVCProHD to SD DVD

    Posted by Dov Yellin on March 1, 2008 at 8:09 pm

    Hey All,

    I’ve been having a strange issue with Compressor trying to convert a DVCProHD program to standard Mpeg2 for DVD. I’m using a MacPro w/Leopard, FCP 6.0.2, newest version of Compressor.

    I have three 54 minute programs to put onto SD DVDs. There is an AJA Kona 3 card in the machine, but I do not have a stand alone DVD recorder. For 2 of 3 shows, I was able to use Compressor to go straight from a stand-alone video file to Mpeg2 (at the Best 60 minute setting, 6-9 bit rate range, progressive, everything at highest quality) in roughly 7-8 hours. Not exactly quick, but it worked overnight and the resulting video looked pretty good.

    But the 3rd one, for a reason I can’t fathom, consistently claims it will take 30-40 hours, and when left to run its course, did actually take over 40 hours. Why is this? All the settings and file types were identical to the previous 2 attempts that did work. I chalked it up to the new Compressor being “not quite ready for primetime’, but is there a different way to get this done?

    If I had a stand-alone DVD recorder (unlikely I will get one just for one DVD) I could hook up to the AJA Kona 3, what is the workflow for laying it off in real-time?

    Thanks!

    Dov Yellin replied 18 years, 2 months ago 3 Members · 7 Replies
  • 7 Replies
  • Uli Plank

    March 2, 2008 at 7:53 am

    It all depends an Compressor’s “Frame Controls”. If they are activated, rendering times explode. For some reason they were probably active in one case and not in the others. They are only used in converting formats, not in compression. My suggestion is: separate downsclaing and compression.

    I’d recommend MPEG Streamclip (free) for downscaling. It does a very good job in a fraction of the time, as long as you don’t try to downconvert from 50/60p to 50/60I (it’ll loose 50% of temporal resolution there). Save in a good format like Photo-JPEG@75% or DV50.

    The feed the SD version into Compressor, make sure it has not activated “Frame Controls” and let it do it’s compression job.

    Even if you are using two steps here, overall speed will be better. One caveat: this approach can only be used as long as the number of frames per second doesn’t change.

    Regards,

    Uli

    Director of the Institute of Media Research (IMF) at Braunschweig University of Arts

  • Dov Yellin

    March 2, 2008 at 5:02 pm

    Thanks for the ideas Uli. Let me respond:

    The frame controls were ON and consistent with all 3 attempts. Without them set to their highest qualities, the video comes out much worse. Still, with those settings all ON, the first 2 shows came out via Compressor in 7-8 hours, the 3rd had trouble. I should mention this also happened when compressing trailers for these shows–2 of 3 compressed in much shorter times than the 3rd, and no, that 3rd was not part of the same full program that is taking a long time now.

    I am starting with a stand-alone DVCProHD 720p 59.94 file, and ending in a MPeg2 59.94 file. No FPS changes. I do have to resize the aspect ratio when I spit out of FCP, to get rid of the HD anamorphic–so I’m exporting as 1280×720 rather than 960×720. I’ve done tests, and this step seems necessary to ensure the final DVD aspect ratio is consistently correct.

    We do have MPeg Streamclip–I will try your suggestion for downscaling from HD to SD with that rather than Compressor, separating the 2 steps. Likely will save a lot of time. Are Photo JPeg and DV50 the best codecs these days? I am more concerned with saving time than saving drive space.

  • Uli Plank

    March 2, 2008 at 5:11 pm

    You can always use ProRes or Uncompressed if yo are concerned about quality. In my experience it’s hard to see a difference over the codecs I mentioned.

    The problem is: DVCProHD in 720 is either 24p with 3:2 pulldown or progressive. If it’s interlaced by pulldown, go ahead with MPEG Streamclip. If it’s progressive, I’d rather use After Effects for this, since it can compose 60 full frames into 30 frames with interlace (much faster than Compressor too). Or get FieldsKit with it’s re-interlacer for FCP.

    Unfortunately, FCP is not yet able (yet?) to preserve temporal resolution when going from full-frames to interlace.

    Regards,

    Uli

    Director of the Institute of Media Research (IMF) at Braunschweig University of Arts

  • Dov Yellin

    March 2, 2008 at 5:26 pm

    Thanks Uli:

    I’m not going from progressive to interlaced, I’m staying progressive. The original FPS is 59.94 and it is staying at 59.94.

    ProRes and Uncompressed are good suggestions, will likely require much less compression time from HD.

    Also, I’ve had success de-interlacing HDV footage using MPeg Streamclip: if I remember right, I selected frame blending in addition to the de-interlace settings. It renders a much smoother motion in the picture.

    Unfortunately After Effects is not available for this job, though Motion is. But again, I’m staying in progressive, not going to interlaced.

  • Uli Plank

    March 3, 2008 at 7:57 am

    Yes, you are going interlaced – or you’ll loose lots of temporal resolution! 60 (or 59,94) fps does not exist in SD.

    Regards,

    Uli

    Director of the Institute of Media Research (IMF) at Braunschweig University of Arts

  • Lon Keller

    March 4, 2008 at 1:42 am

    You’ll always get the best quality downconversions using hardware over software. You have a great piece of hardware with the Kona 3, so use it. Set your Kona3 to output a downconverted signal (you’ll have to decide whether it’s anamorphic or letterboxed) and record on your highest quality SD video deck. Re-digitize your SD video, then encode for DVD.

    Good luck,

    Lon

    Lon Keller
    Chief Editor / Producer
    Sundog Studios

  • Dov Yellin

    March 4, 2008 at 3:13 am

    Thanks for the suggestion Lon–unfortunately I don’t have the option of hardware conversions, as the facility I’m working in is bare-bones. All decks we use are rentals. The only option I have is a firewire-d HDV camera, which kind of defeats the purpose.

    It’s going to have to be a software solution. I already have an HD QT file, 1280×720, non-anamorphic. Compressor’s been terrible at compressing this to an SD m2v, so a 2-step process has been suggested.

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