I'm rendering clips in Sony Vegas using the mainconcept mpeg plugin. The default frame rate setting for NTSC DVD is set to 29.97, but I'vre read that 23.976 is the "true" NTSC DVD compliant frame rate. I'm a bit confused as to which one to render to. Can someone shed some light on this for me?
+ Reply to Thread
Results 1 to 12 of 12
Originally Posted by mltwitz
Live NTSC is 29.97 fps (59.94 fields per second) native and should be encoded that way for DVD.
Film (telecined) is native 24 fps run in the projector at 23.97 fps and field repeated to 29.97. If what you have is film source, you can inverse telecine (remove the repeated fields) back to 23.976, set a 23.976 progressive project and then encode as progressive 23.976 fps. Or you can leave it as telecined 29.97 fps and encode it interlaced.
Standard definition NTSC TV runs at one rate only: 59.94 fields per second. You never see a frame.
If you encode at 29.97 frames per second, each frame is split into two fields and sent one at a time to the TV -- at 59.94 fields per second.
If you encode at 23.976 frames per second, frames are split into fields and some are sent to the TV once, some twice (the process is called 3:2 pulldown). The final result is 59.94 fields per second.
As edDV pointed out, if your source was originally film you might be able to inverse telecine back to the original 23.976 progressive film frames. This has the benefits of lowering the bitrate requirement (fewer frames per second requires less bitrate) and fewer interlace artifacts. But IVTC can be difficult to do properly.
Sorry, but now I'm even more confused, I can't keep up with what you're saying, pl. bear with me
THe sources in my Vegas timeline I'm rendering are usually mixtures of direct DVD rips (vob renamed to mpeg, 29.97) and divx/xvid files usually 29.97, sometimes PAL (25).
So, does that mean I should render to 29.97 then? What is the difference in terms of the end product? Is there any visible difference betwen the two once I watch the finished DVDs on TV (HDTV)?
So, does that mean I should render to 29.97 then?
Not necessarily. You render at the "base" or "original" framerate (if progressive), and encode for the same framerate. When all done and you have an MPEG-2 file at that same framerate, you then apply pulldown (DGPulldown for xx.xxx (original framerate)->29.97fps, so that it outputs 29.97fps interlaced (or 59.94 fields per second) when played in your DVD player. If the framerate is already 29.97fps, then no pulldown is necessary.
In some cases you might want to slow down what was originally 25fps to 23.976fps. But that will also mean slowing the audio. Easier would be to keep it at 25fps.
For 29.97 source it is easiest to just encode 29.97 even if the source is film. This will play as 480i from the DVD and display on any TV or computer.
29.97 encoding preserves the telecine field sequence and passes it to the HDTV or progressive DVD player for further processing in hardware. The HDTV or DVD player then has the option to inverse telecine in hardware or apply some form of deinterlace to create a progressive image. ALL HDTV sets and progressive DVD players have some means to do the progressive conversion. Not all can do the inverse telecine so they cheat by doing a blend deinterlace or something similar.
25 fps source is a different issue. It will probably need both frame size and framerate conversion.
Could you please elaborate more on this?
For example I want to know if "23.976 is the true NTSC DVD compliant frame rate", then when I rip such DVD to an xvid AVI file and get 23.976 FPS (that has visible skips in video) is it a ripping problem, or it's because of the quality of the file? In animations this is even more visible, (I guess) because of the duplicate frames that were used to ease creation of it.
Will DGPulldown* (pull up maybe?) or frame rate change with Virtualdub** suffice or I should rip the DVD from the source with 29.97 FPS again?
Is there a software player that can do this on the fly, so the video runs smoothly on the PC?
* seems that can not handle XVID streams.
** warns that it may de-synchronize audio-video.
Originally Posted by ho1
here's a quick rule of thumb: always either use your most powerful equipment to do the converting, and or scaling, or else do the conversion, and or scaling during the coding process, that way the end device does not have to do it. The quality you get is only going to be as good as your least quality component.
Last edited by sanlyn; 24th Mar 2014 at 12:53.