Determining Video Pixel Size from Still Pixel Sizes

Question

Hello
I posted a few questions here in regard to establshing good Pixel Sizes in stills for use in FCE.
Tom, and Studio X were very helpful, and I picked up a few pointers.
One thing I still didn't grasp completley was how to determine (perhaps from a simple formula), what the Video Pixel Size would be from a photo's pixel size.
As Tom pointed out:
"A 2400x3000 pixel image may be perfectly adequate and not produce as many imaging problems when scaled in a video application"
And as I discovered in regard to print size:
"An 8" x 10" document that is scanned at 300 dpi has the pixel dimensions of 2,400 pixels (8" x 300 dpi) by 3,000 pixels (10" x 300 dpi).
So then all this being said, is there a way to easily determine what the Video Pixel Size will be from the given Pixel dimensions of a still image?
(Like from a re-sized Tiff in Photoshop)
My version of Photoshop doesn't seem to have that conversion option for determining NTSC Video image quality.
Thanx
Mike

Answer 1

And if I need to zoom in really tight to a small area in a particular still, I should allow myself an image size a bit larger than 1280x720. Correct?< </div>
If only it was that easy. The video screen is roughly 800 by 500 pixels. That's all there are. If you bring in a 1600x1000 JPEG (might want to try TIFF instead) and you wish to see the whole image, it will be scaled down to 25% to fit onto the small screen and all of your pixels will be thrown away or mashed together. It's terribly inefficient and the image gets mushily softened.
If you need to zoom in to a particular spot on the image, imagine that spot as filling the 800x500 pixel screen. You can use a simple ruler and measure the area you wish to be full screen (without any scaling being applied by FCP because that will destroy your image). Now measure the largest part of the image that you want to be on screen at any one time when zoomed all the way out. The ratio of those two numbers is what you will need to scan. Example (highly simplified don't try to argue about the precision) if your big image is 10x8 inches and your desired tight shot is 4x3 inches, you must provide a TIFF that is 2.5 or 2.75 times the size of the tightest area. That tight area is going to be 800x500 pixels. You will scan the 10x8 image to obtain a TIFF that is almost 2.5 times 800 pixels wide or 2000 pixels, by 2.5 times 500 pixels high or 1250 pixels. Do not bother to convert that into DPI. You are targeting an image size measured in pixels. It's got nothing to do with the size of the original image or the DPI. Repeating: Absolutely nothing.
This will give you an image into which you can zoom and suffer no image degradation. It will also give you a tremendous processing hit. FCP just doesn't like to do that kind of scaling.
Umm, doesn't look like I helped much. This issue is highly misunderstood and it's quite confusing if you come from a Photoshop background. giving up concepts like DPI is almost impossible. The only way to get it is to work through the many options Photoshop and video people offer up and try to figure out why some of them don't work for you.
bogiesan