"A4" format flatbed scanner big enough to proof whole roll of 35mm in strips of 4

Like David Anderson (http://www.photo.net/bboard/q-and-a-fetch-msg?

msg_id=001aO1) I have a large amount of 35mm film that I wish to scan

for proofing and screen viewing purposes. 1200dpi would be fine for

this, giving me a roughly 1600x1200 proof for each neg which is more

than adequate. (I have a Nikon LS-30 film scanner, which I am happy

to use when I want printable quality.) The film is cut in strips of 4

frames, which can be laid out in 5 rows, 2 strips per row.

 

I am in the process of transferring all the negs into Kenro

clearprint binder pages, in the hope that these can simply be placed

on the glass of the flatbed and scanned as-is. Unfortunately, the

pages are about half a neg strip too wide for my old A4 scanner

(which lacks a transparency backlight anyway, so it's up for

replacement). I would prefer to avoid an A3 scanner, not only for

reasons of cost but also space. Can anyone suggest an A4+ scanner

that is wide enough to take 5 five rows of neg strips in a standard

filing page?

 

Also, if I go for an A3 unit like the Epson 1640XL/Pro, will that

take the negs flat on the glass in the filing pages, or is it

necessary to load them into some kind of carrier?

 

Many thanks in advance,

I can't answer your question directly, but I've been using an Epson 2450 in this fashion for some time, so perhaps some comments might be of interest.

<p>

The transparency area on the 2450 is (just) big enough for 3 strips of 6. I load 6 strips of six to a page, so that ends up being 2 scans per roll.

<p>

I'm not sure 1200 dpi is realistic. Using my geometry (6 of 6) at 1200 dpi and 24-bit color, the image size for 1200 dpi would be over 300 MB.

<p>

I use 600 dpi for proofing. The scan sizes are 37 MB for each scan of 3 strips. I combine them in PS into a single image, and then save as a quality 5 jpg. This yields a file that's 5460x4800, and disk size is about 2.2-4 MB, depending on how well a particular sheet compresses. Of course, once you pull it into memory, it expands back up to its original size as a bitmap. :)

<p>

At 600 dpi, an actual size (100%) view of a single frame is about 850 x 575, and subjectively this ends up being 2/3-3/4 of the full PS client area with the monitor set at 1280x1024. Which works out to be convenient. Any higher in resolution, and I can't see an entire frame on the screen at 100% without scrolling around. Resizing artifacts make viewing at other than 100% less than ideal, ime.

<p>

Other issues are Moiré patterns and light bleed. For examples, see this folder:

<p>

<a href='http://www.photo.net/photodb/folder?folder_id=228579'>Epson 2450 Proof Sheets<a>

<p>

I put this together to accompany a little how-to article I proposed to photo.net (which I haven't heard back on). It illustrates both what a proof sheet looks like and also the cardboard mask I made to help with the process.

<p>

For a specific example of the effects light bleed causes, see this image:

<a href='http://www.photo.net/photo/937819'>Example of "orange bleed"</a>

<p>

As you can see, with neg film I get an orange bleed opposite the sprocket holes, and also at the strip ends, caused by light entering in at the sprocket holes or film edges and diffusing through the film base. With slide film it's not as noticeable, but I definitely still get light areas opposite the sprocket holes and edges.

<p>

Also, since I do the scans with the strips mounted in a page, I get some Moiré patterns. If you look at the large version of that single frame, notice the blotchiness in the clear blue sky are in the middle. I picked a particularly bad example for this demo, but still, I believe you are likely to see this sometimes in any light area of pure, even color. None of this makes the scans unsuitable for my proofing purposes, but they are definitely annoying artifacts. It's certainly not the equal of a real proof sheet from a proofing frame on photo paper....but then it's not $8.00 either, and it's easily fileable on my hard drive. :)

<p>

I also sometimes will copy a single frame off the proof sheet and fool with the colors, or try desaturation, just to see if the image "works" that way. If it does, I pull the actual neg and do a "real" scan at full resolution and go from there. Hard to desaturate a real proof sheet. :)

<p>

Upshot, I can't imagine doing scanning with the amount of film I shoot if I had to scan all the images separately just to see what they look like (I mostly shoot negs, and don't get prints). The proof sheets are far from perfect, but they do what I need, and flipping through them using a slide viewing program is kinda fun. :)

<p>

I still need to buy a real film scanner (Real Soon Now), but I'm glad I bought a flatbed first.

<p>

Good luck with your work!

Hi Gannet,

 

Thanks for that useful info. I hadn't appreciated that the light-scattering issue was the reason for using a mask or holder, although looking at your example scan I don't think it will be too much of an issue. Regarding the practicality of scanning the whole sheet at 1200dpi, I was hoping that a gig of RAM would allow for that but maybe the smart move is to follow your approach of 600dpi and two scans per sheet. It's still way less effort than using the film scanner.

 

Thanks,

Heidelberg make A4 flatbed scanners with full-size transparency hoods.

I believe the Heidelberg scanners are available at a much lower price as UMAX Powerlook models. I had a Powerlook III, which I wish I had kept, which did a great job making contact sheets.

Thanks again for the helpful suggestions.

 

I'm slowly resigning myself to the fact that I'm going to have to use an A4 scanner and do two scans per sheet in most cases. Some Microtek models (e.g, X12USL) can scan legal, giving an extra 50mm or so of length on the page which would help, except they don't do a full-size backlight. The UMAX Powerlook units have a full-size backlight, but the price far more than I'm willing to spend just for proofing. I had a look at the Heidelberg website but I could only see big pro units - Jeff, do you know which Heidelberg models I should be looking at?

 

Maybe I should rethink completely and try and find another film scanner that is faster than the Nikon LS-30 I currently have. Quality wouldn't be an issue, I would still keep the Nikon (which will in fact be upgraded to an LS-4000 in the next few months). This approach would also be a lot more space-efficient than using a flatbed.