Scanning with a digital camera

[Rick Olson]

Agreed ... as I did this experiment last night, I found it pretty easy to get good "camera scans" from my 5 x 7 negs. When I moved on to the 120 and 35mm, I had to get much closer in macro mode, which increased barrel distortion greatly.

Rick

[Brian Ellis]

Depends on what your purpose for posting to the web is. If the idea is to send snapshots to friends and family I guess this is o.k. If you're planning to put it on a web site or anywhere else where technical quality is at all important . . . well, to be honest I don't think these are going to impress anyone.

[walter23]

Agreed ... as I did this experiment last night, I found it pretty easy to get good "camera scans" from my 5 x 7 negs. When I moved on to the 120 and 35mm, I had to get much closer in macro mode, which increased barrel distortion greatly.

Rick

If you have a macro lens, it will work great for this. I've done 120 negs with my canon 100mm macro.

[Padu Merloti]

Ok, a couple more tests.

This time I used a better lightbox (but did a crappy job on dust control) and put a glass on top of the film to make it flat.



For web sharing, I found it decent enough.

Here's the setup I've used for digital duplication:



Dupes were taken using a 10MP dslr, with reasonable glass (tamron 28-75 2.8 XR Di LD Aspherical IF).

For grounds of comparison, I sent the chrome out for my lab (Chrome Digital in San Diego) to scan it with a scitex flatbed. The first thing I could notice was the lack of latitude on the dSLR... on my dupe, if I underexpose a bit more to capture the house number (visible on the scitex) I'd start to loose details on the shadows. Not very important in this picture, but it was a good illustration on why to have a good dedicated scanner for any serious stuff.

Here's also a 100% crop comparison of both digital files:

[Gordon Moat]

Thanks Padu. Definitely interesting, and might be good enough for web sharing of images. That lightbox almost looks like an old flatbed scanner with the lid removed. I am a little surprised at the noise (grain/texture) in the D-SLR shot; almost like that plastic around the 4x5 was left in place, or something on the glass . . . strange.

The colour and tonal quality of the Creo Scitex scan really jumps out beyond the D-SLR. I was expecting the resolution to be a bigger difference, and not the colour quality. Very interesting experiment. It would not surprise me if a D-SLR, AN glass, really good two bulb lightbox, and good lens would challenge some flatbed consumer scanners on image quality. Results like these almost belong in the Comparitive Scanner Review.

Ciao!

Gordon Moat
A G Studio

[JW Dewdney]

It should work fine for what you're doing. I use a copy stand (bencher) with backlight source for similar things. I suspect the problem might be with low-quality (low-CRI) tubes in your box. Do you know what they are?

[Padu Merloti]

That lightbox almost looks like an old flatbed scanner with the lid removed.

If it looks like a duck and walks like a duck...

That's exactly it. I had an old scanner which I removed its internals and put one of these circular fluorescent lights. I got the diffusion piece from an old laptop monitor. It's composed of a few "lens sheets" and one or two diffusion sheets. I'm not sure I liked it. Maybe I'll spend 50 bucks and buy one of those slim lightpads.

I am a little surprised at the noise (grain/texture) in the D-SLR shot; almost like that plastic around the 4x5 was left in place, or something on the glass . . . strange.

I agree, and it only caught my attention when I compared it with the scan... maybe it's the notebook lens sheet that I used for the lightbox diffusion.

The colour and tonal quality of the Creo Scitex scan really jumps out beyond the D-SLR. I was expecting the resolution to be a bigger difference, and not the colour quality. Very interesting experiment. It would not surprise me if a D-SLR, AN glass, really good two bulb lightbox, and good lens would challenge some flatbed consumer scanners on image quality. Results like these almost belong in the Comparitive Scanner Review.

The color difference I believe I could've adjusted in photoshop, but what really suck on the dslr is the lack of dynamic range. I found out that resolution is not as bad as I thought. DR in the other hand may be the biggest problem, but again, this is only for web display. A file like this will never see the color of paper.

Ciao!

Padu

[Padu Merloti]

It should work fine for what you're doing. I use a copy stand (bencher) with backlight source for similar things. I suspect the problem might be with low-quality (low-CRI) tubes in your box. Do you know what they are?

They are round fluorescent lights that I got from an old x-ray viewing box. What is low-CRI?

[Gordon Moat]

Hello Padu,

The slim lightboxes are okay when you need to transport something. A better choice would be a Porta-Trace Lightbox. I got a two bulb 10" by 12" recently, which seems to make a big difference viewing transparencies. You can also get replacement bulbs for these in the future when needed.

After reading your explanation, I think the texture might be the old laptop screen. Probably a proper white Lucite sheet might have worked better, though then you would have needed a stronger bulb. Easier just to buy a lightbox. I got the Porta-Trace stainless finish, though I think maybe the oak finish would look much nicer, not that it makes any difference in how they work.

Ciao!

Gordon Moat
A G Studio