scanning negs

[Glen Irvine]

Dear Folks,

This is a bit of a follow up to an earlier question. My goal is to scan
negatives ( using an Epson photo stylus 890-no plug intended ), take these into photoshop for some manipulation, and then import these to Final Cut
Pro for inclusion on a video project. Any advise on what dpi I should scan
the negs at, the size I should scan at, or for that matter any little "tricks"
that might improve the final product. As always, thanks in advance for
your help. I would imagine many of you have done this before. All
the best, Glen in Whistler

[Ken Tanaka]

Hello Glen,

The answer depends on what you plan to do with the stills in FCP. For example, if you just plan to display them for a few seconds in full-frame then you should just scan them at 720x480 at 300 dpi. Alternately, if you plan to pan or zoom the stills (with the motion control) for visual interest then bring them in larger to allow yourself room to move.

p.s. Are you in Whistler B.C.?

[Glen Irvine]

Ken,

Thanks Ken. I'm playing around right now. When saving the scanned
negs is there any prefered file type ( jpg vs. tiff, etc.. ), and do you have any
tricks to help sharpen the image presuming that the orginals are quite sharp to start with. And yes, I am in Whistler and it's starting to snow again.
What a year. Glen

[Ken Tanaka]

Re: format, I generally use uncompressed TIF under the theory that FCP will have to compress the image anyway. So I want to give it the truest version I can.

Re: sharpening, use Photoshop or a similar tool to get the image as good as you can get it before importing it. Then, once it's in FCP, just treat it as regular footage and apply any sharpening or other manipulations as needed.

Re: size, just remember that it's better to reduce the size (for a zoom-out) than to try to increase the size (for a zoom it). That is, bring it in at the largest size you think you'll need.

[Glen Irvine]

Ken,

Got some stuff that's acceptable this afternoon and am learning a
bunch in the process. But here is the thing that still puzzles me. I scanned
some images at 300 dpi , took them into photoshop, played around with them, and then resized them to 720 by 480 with a resolution of 72 dpi
as suggested in Lisa Brennel's Final Cut Pro Visual Quick Start guide.
I guess my rather vague point is, if I'm going to eventually end up
at 72 dpi any way, why don't I just scan at this resolution and then
mess with the scale in Final Cut when I want to apply motion.
Does what I'm saying make any sense or have I missed an essential
step in my reasoning ( altogether possible ). Anyway, thanks again,
you're a wealth of information. Glen

[Ken Tanaka]

A very logical question, Glen.

Basically, it comes down to the amount of information you want to work with, primarily in the Photoshop step. That is, you want to hand-off the best possible image to FCP.

Consider this. Assume for a moment that you're starting from a film image (ex: a slide or negative) which may have an effective resolution of 3,000 dpi. Scanning it at 300 dpi means you're already only working with 10% of the original image's intrinsic information. To Photoshop, which earned its stripes working with print-quality images at around 1200 dpi, a 300 dpi image doesn't even make the work interesting. It's very, very good at handling any manipulations you might throw at such an image...including changing the size and/or resolution of the final image. In fact it's far better at that task, and offers you more control over it, than any normal scanner. So I would scan the image at whatever my scanner's highest -native- (i.e. non-interpolated) resolution might be (often 150 dpi-300 dpi), manipulate the image's color, etc. in Photoshop at that resolution and then let Photoshop down-sample the image's resolution for consumption by FCP.

In the end, of course, it's all about how the results look to your eyes. If you have some lab time try a variety of resolutions and techniques with your source material to see what looks best to -you- using -your- equipment and software.

Have fun!

[Ian Austen]

May I just add that if your final out put is to tape rather than web, you should first create your image to 720 by 540 then resize to 720 * 486 because of the difference between sq and non sq images. Your computer screen uses sq and the video mon uses non sq. Also if you capture from video to create a still for printing reverse the process.

[Glen Irvine]

Ian,

Thanks for that. For some reason I had already figured out this piece of the puzzle. Still it was a useful reminder. Thanks for the response.
Glen