Nikon D800 as a "Scanner"

I've tried a similar scanning method with a sony nex 5 and their 30mm macro lens. For B&W it works really well, but for color negatives a simple white balance to get rid of the orange mask is not emough, because the mask is too strong. At least in the Sony RAWs I would put the white balance slider all the way to the edge and it would still not be neutral... There is simply not enough information in the other channels...

Here's an example of a B&W "scan" that I'm quite happy with:


... by TheManWhoWasn'tThere, on Flickr
 
j j said:
Why shoot the frame? You can take a white balance reading from blank film.
Click to expand...

You would be best shooting a frame of a gray card, properly exposed, and basing the whit balance adjustment for base tone on that. It's a better representation of the dye mix for neutral-white than film base color.
 
d__b said:
I've tried a similar scanning method with a sony nex 5 and their 30mm macro lens. For B&W it works really well, but for color negatives a simple white balance to get rid of the orange mask is not emough, because the mask is too strong. At least in the Sony RAWs I would put the white balance slider all the way to the edge and it would still not be neutral... There is simply not enough information in the other channels...
Click to expand...

As someone else suggested, filtering the illumination source to push more cyan-blue into the digital capture improves the ability to correct within the software tools' adjustment range. A custom camera calibration based on a negative of an Xrite Color Checker would also give you a lot more adjustment range.
 
I've had my D70 on the PB-6 for several years now to digitize old film. The two problems that I have yet to overcome are the unavoidable cropping of the film frame, and the terrible results with color negs.
How have you dealt with the cropping issue and the white balance in the color negs?

One of my old Tri-x frames from 40 years ago (http://www.rangefinderforum.com/photopost/showphoto.php?photo=157174).
 
Last edited:
d__b said:
... but for color negatives a simple white balance to get rid of the orange mask is not emough, because the mask is too strong.
Click to expand...

That's why I am experimenting with a very blue light for shooting the negative with the digital camera. Exposing the C41 negative with white light, there just isn't enough blue coming through the mask.

You were using a color balance to adjust. I used curves where I can make even more extreme changes. Still no go from a 16bit RAW file.

And, I like your B&W image.
 
I'm interested in this method, too. I scan 35mm and 6x7 color negative with V700 and silverfast. While the result with 6x7 is fine, I hate the speed and the IQ of 35mm. After reading this thread so far, while i love to try this method, I am not still sure if this method works well for my negatives, since the inversion process seems not to be easy. The color that I can get from Silverfast is better than Epson software, and I always think that there must be some secrets in the Silverfast process...
 
One problem that I have not solved well is that of uniform frame illumination. The milk glass diffuser on the PB-6 slide copier still requires a uniform cold light source and still some frames produce various artifacts. So far the best light is a flat monitor. Larger formats could make the problem of uniform illumination a bit more difficult.

I've given up trying to compensate for the base color and just use this setup for my old B&W film and color slides... Maybe it's my old D70 that needs to be replaced?

...and thanks for the kind comment Colonel. Moni's 43 years old now.
 
Tony Pio said:
One problem that I have not solved well is that of uniform frame illumination. The milk glass diffuser on the PB-6 slide copier still requires a uniform cold light source and still some frames produce various artifacts. So far the best light is a flat monitor. Larger formats could make the problem of uniform illumination a bit more difficult.

I've given up trying to compensate for the base color and just use this setup for my old B&W film and color slides... Maybe it's my old D70 that needs to be replaced?
Click to expand...

I have used the PB-6 and Nikon slide copier quite a bit with good results by pointing an electronic flash at the milk glass and with a lamp and reflector.

I'm encouraged by my experiment in working with color negatives. By using a light source with enough blue to counteract the orange mask, and using the "curves" tool rather than white balance, I'm getting reasonable color easily. How about a dichroic head from a color enlarger? These should be cheap.

With this, I don't see why a D70 wouldn't work, up to it's sensor limits.
 
ColSebastianMoran said:
Let us know what filtration on the color head gives you a reasonable balance with this setup.
Click to expand...

I am rebuilding the Chroma-Pro. What a Hoot.

I have now converted it from Nikon F to Minolta SRT. The color head still works, goes to 130CC. Magenta, Yellow, and Cyan. 100W Tungsten bulb as you would find in a slide projector or color head.

Heading off to the hardware store, to get a machine bolt, to attach the heavy base column to the Dichroic "box". Using a 50mm F3.5 Macro Rokkor MC on the bellows section, when I get it setup with an SRT body, then I will add a Nex-3.

Looks like it might do 6x6 to me, as well.
 
Tony Pio said:
I've had my D70 on the PB-6 for several years now to digitize old film. The two problems that I have yet to overcome are the unavoidable cropping of the film frame, and the terrible results with color negs.
How have you dealt with the cropping issue and the white balance in the color negs?
Click to expand...

I was playing around with the custom white balance on my dslr (e510) while 'scanning' and found the best color inversions came when I had it set between 2000-2400 degrees kelvin. The flourescent tubes under the neg were probably of the 4000-5000k range.

The results of a quick 2 shot stitch looked decent. I'd say I got most of the detail that the "24 mp" lab scan obtained. The dslr scan showed less color-grain-blotch, but also less contrast, and duller skin tones. White balances of 2000 K in-camera gave a warmer look, 2400 cooler.

DSLR scan @2000 K custom white balance, Olympus Trip 35, Kodak Gold 200
8109921516_929ea8c14f_c.jpg


Downsized '24 mp' Lab Scan:
8109895893_00380a04ae_c.jpg
 
Godfrey said:
You would be best shooting a frame of a gray card, properly exposed, and basing the whit balance adjustment for base tone on that. It's a better representation of the dye mix for neutral-white than film base color.
Click to expand...

Really? Wouldn't that add another variable (the light conditions under which you take the reference image)?
 
Noll, that looks pretty good.

I worry with WB in the 2000-2400 range, that good color will be hard to obtain. Your DSLR scan looks pretty good, but the building in the upper right looks a little ill.

I wonder if a higher K bulb would produce a better result. Say a daylight balanced bulb?

I had 110 C and 60 M in the filtration in my test and the histograms were reasonably balanced at 4000 K.

Any color-savvy engineers reading this?
 
Just thinking... We could make a fortune if we created and sold the right light source for digitizing color negatives with a simple slide copier!
 
Godfrey said:
You would be best shooting a frame of a gray card, properly exposed, and basing the whit balance adjustment for base tone on that. It's a better representation of the dye mix for neutral-white than film base color.
Click to expand...

Godfrey, this sounds right to me. The C41 film-base shows the mask color, but it's signaling absolute black in the positive image.

Probably best to get white balance from gray, not from black or white. I'm going to shoot a Macbeth Color Checker in my next roll of C-41.
 
Why not just pop it with a SB-80, SB-800 or 900 strobe on a light stand?

Power is adjustable in essentially 1/3 stops down to 1/128 full power. Gels with a wide variety of color temperatures are available.
 
You must log in or register to reply here.
Back
Top Bottom