Creative Communities of the World Forums

The peer to peer support community for media production professionals.

Activity Forums VEGAS Pro Rendering doubts

  • Rendering doubts

    Posted by Fabio Mereghetti on September 18, 2010 at 2:57 pm

    I have this timeline with DV clips mixed together with HDV clips.
    I have to render to DVD: what about the interlace settings?
    If I am correct, DV is bottom first and HDV is top first and I understand that, in theory, the best quality I can deliver is leave it interlaced. So, which settings should I put in the rendering and in the project (no graphics added, just cross-fades).

    And I wonder if the necessity for a resize (1440×1080 par 1.333 to 720×576 par 1.4568, for the HDV clips), before the rendering, will introduce the need to deinterlace, before converting to MPG2.

    By the way: who will be “The Risazer”? Vegas? The target codec?

    I read that the resizing algorithm of NLEs and codecs as Mainconcept sucks. Is it true? For CCE and Procoder 3 too?
    Shall I have to use Avisynth or Virtualdub with “splitfields” + “resize” + “weave” to get the best possible results?
    It’s all this just an urban legend?

    Thanks.

    John Rofrano replied 15 years, 10 months ago 2 Members · 7 Replies
  • 7 Replies
  • John Rofrano

    September 18, 2010 at 3:23 pm

    If you are delivering on DVD. Set your project properties to DV and don’t change any of the settings. Then edit your DV and HDV source and finally render to MPEG2 using one of the DVD Architect templates. Let Vegas worry about field order and interlacing. It will do the right thing.

    ~jr

    http://www.johnrofrano.com
    http://www.vasst.com

  • Fabio Mereghetti

    September 18, 2010 at 4:02 pm

    Just because I like to understand, let me ask you something.

    1) If I don’t generate any media in a DV + HDV timeline and I would like to render to a deinterlaced output, which of the following settings will be part of the process and will influence the output results?

    Thanks again.

  • John Rofrano

    September 18, 2010 at 4:51 pm

    Just because I like to understand, let me ask you something.

    Sorry I wasn’t more explicit. When I said “don’t change the settings”… i really meant “don’t change the settings!“. You are changing the settings which is not going to work.

    Let me answer your question and then I’ll tell you why what you are doing won’t work.

    1) If I don’t generate any media in a DV + HDV timeline and I would like to render to a deinterlaced output, which of the following settings will be part of the process and will influence the output results?

    The only settings you need to be concerned with are the two Field order settings (project and render). Do not touch the Deinterlace method (you already have it set correctly).

    Why this won’t work:

    DVD is a very strict specification which allows:

    29.970 fps interlaced video (commonly referred to as 60i)
    25.000 fps interlaced video (commonly referred to as 50i)
    23.976 fps progressive video (commonly referred to as 24p)

    You are trying to make 25 fps progressive DVD (i.e., 25p) which is not supported. If you want your video to be progressive, you need to create an NTSC 24p DVD. If you want a PAL DVD then your only choice is to render interlaced. If you don’t, DVD Architect will happily take the progressive video you created and re-render it to 50i to be in compliance with the DVD spec so you might as well do it right the first time.

    ~jr

    http://www.johnrofrano.com
    http://www.vasst.com

  • Fabio Mereghetti

    September 18, 2010 at 8:51 pm

    I didn’t know of the different allowances between PAL and NTSC. I thought I could encode a progressive PAL DVD.

    I’m not a fan of progressive encoding. In fact I’m a fan of MPQ (Maximum Possible Quality).

    In my search to understand which path to follow, I made some tests, rendering little bits of the project I will have to put on DVD.

    Today I rendered a few seconds of an HDV event deshaked with Mercalli 1.

    As I wanted to see the differences, I first rendered to Mainconcept MPG2 with everything at default (interlaced) and then with everything progressive (project and render).
    Watching the interlaced one with MPC HC I noticed that in many points (in time) of the HDV clip I could see a sort of a general blur for just a few frames.
    The progressive render don’t show that.
    So I went and captured the same frame from the two clips and you can clearly see the interlaced one is quite blurred.
    My explanation is… yours! I mean, in another post I asked you if a deshaker would introduce blurriness and you answered that it won’t but, after the process we can see as steady some frames that before were not steady = blurred.
    Is it correct?

    But I don’t understand why the progressive render doesn’t show the same effect.

    Have a look at the three clips, if you like.

    1100_interlaced1.mpg.zip

    1101_deintatrender1.mpg.zip

    1102_deintatmethodandatrender1.mpg.zip

    And here are the captures from the three clips:
    (I don’t know why MPC HC captured them shrinked while played them wide)

    This is interlaced at project and render

    This is interlaced at project and deinterlaced at render (Mainconcept MPG2)

    This is deinterlaced at project settings and at render.

  • John Rofrano

    September 19, 2010 at 5:23 pm

    It depends on how you deinterlaced. If you threw half of the fields out that would eliminate the blurring but it would also reduce the vertical resolution by half.

    Does the blur bother you when you watch it? It doesn’t matter what a single frame looks like. These are not still images, they are video. All that matters is what the video looks like. Motion blur is natural. removing it is unnatural.

    ~jr

    http://www.johnrofrano.com
    http://www.vasst.com

  • Fabio Mereghetti

    September 19, 2010 at 8:35 pm

    Thanks.

    Yes, it bothers me.
    If you have time have a look at the first clip yourself and then at the third.
    Now I’ll go and watch them on a TY though, because I only watched them on my PC display, till now.

    Thanks again.

    P.S.

    Any opinion on the “resize” in/out the render engine?
    I’m making some tests to see if the straight way (Vegas->Mainconcept) is far worst than frameserving to Procoder, CCE or HC through a smart resizing Avs script or not.

  • John Rofrano

    September 19, 2010 at 11:02 pm

    Now I’ll go and watch them on a TY though, because I only watched them on my PC display, till now.

    I assume you mean TV and if you are watching interlaced video on a PC… yea, it’s always gonna bother you.

    The reason I said to feed the original interlaced video into Mercalli was so that it could do a good job with the motion detection because deinterlacing would loose this information. If you like the look of deinterlacing first then use that. There is no right or wrong. it’s just a matter of what looks best to you.

    BTW, most stabilizers would require you to deinterlace the footage first because they can’t handle interlaced footage. Mercalli is one of the few that can but it’s not uncommon to have to deinterlace before stabilizing.

    ~jr

    http://www.johnrofrano.com
    http://www.vasst.com

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