I am working with Apple Compressor to scale up many SD files to play on a 720p HDTV and I am confused about which aspect ratio to use in the pixels.
Much of the source material was shot on MiniDV cameras in the 720 x 480 pixel aspect ratio. The video cameras call this "4:3" but mathematically, it doesn't work out to that, it actually works out to (3:2).
BUT! I understand that the pixels are NARROW horizontally. .9, so that 720 actually = 640.
So, my question is, when scaling up SD DV footage for HD (which has square pixels), should I set Compressor to make the new video 4:3 or leave it at 3:2?
To further complicate matters, when used as a DVI computer monitor, our HDTV's have a pixel count of:
1360 x 768
So the formula I was tinkering with is such: 480 gets blown up to 768.
768 / 480 = 1.6
So I should multiply 640 x 1.6 to get a 4:3 ratio.
640 * 1.6 = 1024
But I am Streaming the video files with QTSS, so they all have to match in size, so I need to encode vertical black bars on the sides of the originally SD video.
Compressor has a control for this in the Geometry Tab, it calls it "Output Image Inset (Padding).
So my dilemma is, what size should I make the padding pixels?!
Here's what I'm thinking.
If I do "strict 4:3", I'll end up with 1360 - 1024 = 336
Divide that remaining 366 by 2 to put even bars on each side = 168.
If I go with "strict 3:2" the numbers will be: 1360 - 1152 = 208
208/2 = 104
So, 168 or 104. Well, I'm sure you'll say, "Just use the one that looks right."
That's the problem!!! They both look decent!!! Aiiiigghh!
Okay, well. If anyone has any thoughts, I'd love to hear them.
Thanks Daniel. The main part of your answer, "set it to 640 x 480, Square Pixels" is helpful. But I forgot to mention some stuff.
The television is used as a Monitor for a Mac Mini, placed behind the screen. It is getting a computer monitor signal over DVI. It isn't getting a television signal, so I can't set the aspect ratio on the television when it's in this mode.
The other issue is that SOME of the videos are widescreen 16:9 SD, or are HD to begin with. So that's why I have to encode the black bars WITH the SD video clips. Because they might show up in a stream with 16:9 material. That's why I'm having a hard time figuring out exactly what the proper pixel count should be on the black bars.
Would you say that you are certain that 720 x 480 source should go to 640 x 480 square?