Video Frame Size

GlenO wrote on 7/16/2002, 11:54 AM
I have captured old 3/4" video into Vegas and the frame seems to be out of alignment on all the videos, with black strips along either side and the top and bottom. Is there a way to shrink or edit the viewing area size before rendering an mpeg? Could it have something to do with the "safe zones" in the preview overlays? Any input would help. Thanks
GlenO

Comments

SonyDennis wrote on 7/16/2002, 12:35 PM
DV captures 720 pixels wide, which includes 16 pixels or so outside of the NTSC active area. Some older DV camcorders do this too.

If your final output is a TV set, don't worry about it, because it will fall outside of the viewable area.

If your output is streaming or anywhere else that you'll see the full frame, use the event Pan/Crop feature to zoom in on the clips a little, or the Track Motion feature to do the same (kind of in reverse, you'll be zooming the track output outside of the project space.

///d@
wcoxe1 wrote on 7/18/2002, 12:18 PM
Speaking of Video Frame Size:

DV is 720 x 480,

but someone just send me a "safe area" transparant video frame so I could see it on my external monitor (SF, please let safe areas and grids show on external monitors.)

This clip is, I think, 665 by 480, yet is fills the WHOLE frame. How?
Chienworks wrote on 7/18/2002, 12:34 PM
The aspect ratio of the image file that was sent to you is 1.0, but the aspect ratio of DV is 0.9091. The image is stretched out to fill the 720-wide frame, which is then shrunk back to 4:3 for display.

Sometimes it's better not to ask why ;)
wcoxe1 wrote on 7/18/2002, 11:32 PM
Boy, ain't that the truth. But thanks, I think.

Actually, your answer cleared up a LOT of questions I had jingling around upstairs.

Thanks.
Chienworks wrote on 7/19/2002, 12:24 PM
Actually this brings to mind another thought that has been hiding in the deep dark recesses for a long time. I've noticed that still image sequences take a LONG time to render. One would think that if there is a section of the timeline that contains a still image with no effects, crossfades, transitions, etc, that it could render this as fast as doing a straight DV copy; create the first frame, then just write it out repeatedly. But this doesn't happen. It seems to take longer than many other rendering tasks. I wonder if the still image is resampled to a DV frame for every frame? Hmmm. Would certainly explain a lot. I'd like to think this isn't so, but then again, if it is so, that leaves a huge performance increase that SonicFoundry could implement ... ;)