Multiple Re-Renders of 24P 2-3-3-2 AVI

Norm Sanders

Norm Sanders

Script Wrangler
When rendering out 24PA to an AVI that's going to be used in another project, I know you want to go out with 2-3-3-2 vs. 2-3 to retain quality. But does anyone know how much, if any compression is happening at that point?

Reason I'm asking is that I've got several renders that need to be done in steps for color correction, additional masks, etc. If I take the same 2-3-3-2 and re-render with a new mask/cc, etc. back out to a 2-3-3-2, to bring in & finally render to a completed MPEG2, am I going to lose noticeable quality in the AVI re-renders?
 
Every time you render out, you face the potential of losing quality (depending on which codec you're using, some are much better than others on multi-generational tests). Vegas has an excellent DV codec.

Re-renders should be avoided when possible; if it's not possible, then you can try rendering to totally uncompressed -- that'll avoid any additional compression artifacts, but will take up a lot more space.

If you're only going one or two generations, the Sony DV codec should be able to handle that with minimal quality impact.
 
Thanks, Barry! But how does someone render out to uncompressed & still keep their widescreen aspect ratio? If I go with the uncompressed codec, it comes out true 720x480 ... I've tried setting a test project in both regular DV, as well as Widescreen DV, and when rendering out they look exactly the same when using the Uncompressed option.

Someone else was having a problem with this a while back when trying to render to PAL Widescreen (trying uncompressed), and I thought it was maybe just a PAL issue, but it's doing it for NTSC as well (no widescreen).
 
Jst change the pixel aspect ratio. When you go to render, choose "default/uncompressed", then go to "custom" and look at the video tab. There's an option there for pixel aspect ratio. If you set it the same as your footage (1.2) then it should work perfectly. Guessing, as I haven't tried it...
 
Tried it, doesn't work. In fact, when you're rendering out uncompressed, it'll automatically choose the right pixel aspect ratio, per whatever your project is set at ... still though, once rendered it's not widescreen.
 
Envision said:
Tried it, doesn't work. In fact, when you're rendering out uncompressed, it'll automatically choose the right pixel aspect ratio, per whatever your project is set at ... still though, once rendered it's not widescreen.
Click to expand...
Can't you set the pixel ratio when you import the rendered file into the next project? I do it all the time when shooting with anamorphic. Mind you, the AVIs are not uncompressed, but it should still work...
 
Okay, I just tried it. It does work. It will render uncompressed in exactly the same aspect ratio. Thing is, when you import the footage back into your project it will come in with a 1.000 aspect ratio, and that's why it looks wrong.

Work in a 1.2 aspect ratio project. Output uncompressed. Import the uncompressed clip back in. Go to properties and change its aspect ratio to 1.2 -- it'll look right in the timeline; it works fine.
 
Ah, I never tried that, thanks! I was just watching the footage through the Windows Media Player, which is probably not a good place to judge it, since it IS uncompressed & probably not meant for that player.
 
Barry_Green said:
Thing is, when you import the footage back into your project it will come in with a 1.000 aspect ratio, and that's why it looks wrong.
Click to expand...

And isn't that annoying.

Same thing happens when you render to a new track.

Wish they'd fix it. (Not that it's any great hoohoo to change the properties)
 
I just noticed something. Ideally when rendering my 24PA footage into an AVI that I'll be bringing into another project again, I want to render it with inserting 2-3-3-2 pulldown. However, there's no option between that OR inserting regular 2-3 pulldown, when doing it uncompressed ... just the 23.97 IVTC Film option.

Anyone know what it's going to render to (with 2-3, or keep the original 2-3-3-2)? Does it even matter if I render it into a massive uncompressed file?
 
Barry_Green said:
It wouldn't introduce any pulldown, it would render it to 23.976 individual frames per second. No pulldown at all.
Click to expand...

So that's a bad thing, right?? And how can 23.976 per second be actual individual frames? I thought 24, 30, 60, etc. can be individual (i.e. whole numbers, not fractions).

Keeping within the questions of rendering, if 24PA is rendered out to Quicktime or WMV at 24FPS, instead of the 23.976 (ITVC or whatever it's called), would that look as good, better, or worse?

Sorry for all the Q's ... at least I'm not pounding the same question to death ... just that responses are making me think of something else.

Thanks for your help, Barry, and everyone else!
 
Envision said:
So that's a bad thing, right??
Click to expand...
No, it's better because there's no unnecessary/field-blended interlace frame.

And how can 23.976 per second be actual individual frames?
Click to expand...
It is. The NTSC base rate is not 30 frames per second, it's 29.97 (or, put another way, it's not 60i fields per second, it's 59.94). So the actual frame rate for individual, discrete frames is 23.976 per second. There's no point in time where ".976" of a frame is displayed; instead it's implying that the thing that changes is not the frames, but the notion of so-many-per-second. In other words, one second is not the right time value to properly describe the frame rate.


Keeping within the questions of rendering, if 24PA is rendered out to Quicktime or WMV at 24FPS, instead of the 23.976 (ITVC or whatever it's called), would that look as good, better, or worse?
Click to expand...
The DVX runs at 23.976, not 24. So using 24 will be less accurate. The proper speed to use, whenever you have the choice, is 23.976.
 
Barry_Green said:
In other words, one second is not the right time value to properly describe the frame rate.
Click to expand...
It should be called the NTSC-second. :laugh: It lasts 1.001001 of real second.
 
Barry_Green said:
No, it's better because there's no unnecessary/field-blended interlace frame.
Click to expand...

Okay, hopefully last question! :thumbsup:

If there's no interlaced frames rendering it uncompressed at 23.976, when I bring it back in, is there anything special to do to it, or will it basically act just like the original captured footage, and Vegas will do what it needs to behind the scene? Then, I'm assuming, if I rendered it back out (the uncompressed file) into the widescreen codec, I'd choose at that point either to either insert 2-3 or 2-3-3-2 pulldown, depending on what I was going to do with it next (i.e. to tape, or to bring back into a project, etc.).

Barry, are you sure you're not a professor of some sort?
 
Well, I do a lot of "professing", but as to whether that qualifies or not...

Okay, back to the question at hand: Vegas will work with it perfectly (as long as you set the aspect ratio). It knows what to do with it, it knows how to insert pulldown if necessary. And yes, when you render out back to tape, that's when you'd re-insert pulldown.
 
You must log in or register to reply here.
Back
Top