My Camera Scanning Experiments

Chriscrawfordphoto

Real Men Shoot Film.
Local time
5:37 AM
Joined
Apr 1, 2007
Messages
11,390
As many of you know, my Nikon film scanner that I had used for 19 years died a few months ago. I can't afford another and would be afraid to spend the money even if I had it since they're often not repairable and all of them are getting quite old now.


So...I had to find an alternative. I experimented briefly with camera scanning but wasn't totally happy with the results. Then, in October my girlfriend bought me a Plustek 8200i. That was a bust, too. I tried and returned two of them and both had misaligned optics that caused one side of the scans to be EXTREMELY soft. You can read about that in this thread. Gave up on getting a decent Plustek. I need stuff that 'just works.' I don't have time to deal with poorly made junk.


So, before giving up on shooting film altogether, I began thinking about how to make camera scanning work better. My son helped me build a jig to hold the camera and film in place using Lego bricks. Mack is really into building stuff with Lego. I'm using the glass carrier from my dead Nikon 8000ED scanner to hold the film.




scan-setup-1.jpg





scan-setup-2.jpg





scan-setup-3.jpg





I'm using my Olympus OM-D E-M1 mark II with the Olympus 60mm f2.8 Macro lens. I'm photographing the film with the camera set at its base ISO (200) using the high-res sensor-shift mode that gives a 50mp image instead of the normal 20mp the camera gives in normal shooting.
 
Cool. Are the images better than your old Nikon scans?




Here are some results:




PB0800277-2.jpg





PB080016.jpg



These were both 35mm negatives shot on Tmax 400, developed in D-76 1+1. The first was shot with a Voigtlander 28mm f3.5 Color-Skopar and the second with a Leica 50mm f2.4 Summarit-M.






Here are some 100% crops from the second image:


PB080016-center.jpg



Center




PB080016-left.jpg



Left side




PB080016-right.jpg



Right side




As far as resolution, I think the results are slightly better than the Nikon scanner. Not a lot, but slightly. Tonality was a different story. It was A LOT easier for me to get great image tonality from the camera scan than it was from the Nikon scanner!



I have had only one issue with this. For some reason, the Lego stand is not giving perfect alignment between the camera and the film. The sensor must be perfectly parallel to the film to keep the whole image sharp across the frame. I'm finding that the camera has to be turned to the left very slightly to get even sharpness. I'm still trying to figure out why that is. As long as I do that, the results are flawless.
 
That looks fine to me. I found when I was doing this years ago that light piping or light from the side was a problem. Maybe Mack can build you something (maybe even Legos) that would act as a 'bellows.'

I brought my rig out the other day just see if I could get it to work with a new digital camera. This was the only negative I shot but it is surly better than my scanner.

This the scanner full frame and then a scanner blow up:

Olympus 35RC Trix HC-110h by John Carter, on Flickr

and this is the scanner blow up:

Olympus 35RC Trix HC-110h by John Carter, on Flickr

This is the digital camera 'scanned' negative blowup.

Olympus 35RC Trix HC-110h by John Carter, on Flickr

I think I could have done better as I was having a little trouble with the focus, I tried 'peaking' but I don't know if I was doing it correctly.
 
Cool solution Chris!
Which surface of the camera body are you using to locate off the Lego's?
I'm betting the back of the bottom plate. To my thinking, that would be a surface the factory could really care less about as far as being parallel to the sensor.
Try the front surface of the bottom plate if you're not using it and see what happens. I'm betting it is more aligned to the sensor since the mount is on the front of the camera too, and it needs to be aligned precisely to get focus correct.
 
That looks fine to me. I found when I was doing this years ago that light piping or light fromr the side was a problem. Maybe Mack can build you something (maybe even Legos) that would act as a 'bellows.'

I brought my rig out the other day just see if I could get it to work with a new digital camera. This was the only negative I shot but it is surly better than my scanner.

This the scanner full frame and then a scanner blow up:

Olympus 35RC Trix HC-110h by John Carter, on Flickr

and this is the scanner blow up:

Olympus 35RC Trix HC-110h by John Carter, on Flickr

This is the digital camera 'scanned' negative blowup.

Olympus 35RC Trix HC-110h by John Carter, on Flickr

I think I could have done better as I was having a little trouble with the focus, I tried 'peaking' but I don't know if I was doing it correctly.




The camera scan looks a lot better than the one from your scanner! Even if the focus was off with the camera, you still got sharper results. I just used the autofocus on my camera, since I was using a lens made for it and the AF has been consistently perfect with the Olympus gear; it focuses right on the film grain.


I do the actual 'scans' at night with all the lights in the house off. I haven't had any issues with flare, but if I keep the lights on, I get glare on the negative carrier's glass.
 
I agree with the addition of a bellows, even with all the lights off in the house.
Look at what the lens is pointed at!
You want all the light entering the lens to come thru the film only, IMO.
 
The camera scan looks a lot better than the one from your scanner! Even if the focus was off with the camera, you still got sharper results. I just used the autofocus on my camera, since I was using a lens made for it and the AF has been consistently perfect with the Olympus gear; it focuses right on the film grain.


I do the actual 'scans' at night with all the lights in the house off. I haven't had any issues with flare, but if I keep the lights on, I get glare on the negative carrier's glass.

The nice thing about these setups you can use a closed down aperture because everything is connected therefore longer shutter speeds are not a problem
 
I agree with the addition of a bellows, even with all the lights off in the house.
Look at what the lens is pointed at!
You want all the light entering the lens to come thru the film only, IMO.




The end of the lens is very close to the film so i don't think its seeing the rest of the light from the light box. I could cover the rest of the light box with black paper, or buy the lens hood Olympus sells for the macro lens (which I have avoided because the damned thing costs $45 for a plastic hood). I have seen NO flare though.
 
Thanks for reporting on your results. Was this all 35mm scans?
Using my considerably lesser spec EM10 this is how I would get my Pen F half frame negs into my computer. Have to use a very old 55mm f2.8 Vivitar macro, it will focus down to 1:1 without needing an extension tube.
For myself I’d just cobble up a rig from hobby plywood,
Really like plywood.
 
It's not about 'flare' Chris, it's all about controlling the light that is hitting the sensor and coming in thru the lens.
 
For alignment: put a mirror at the film location. Stop down lens to get a small aperture. Turn on the grid in the camera screen. Use room lights, not light table. Center the aperture on the grid center.


You'll need to change exposure to get the aperture to show. And focus will need to be changed to get the aperture to be sharp.


At 50mp, I guess that's the Olympus JPG output? If you are doing RAW, be aware that not every RAW converter handles multiple files for high resolution. I am using a Pentax K3ii and Affinity photo. Then read about the high resolution issue. Tried another RAW program (RawTherapee) and sharpness and detail jumped.



Your images look pretty sharp, but figured I'd mention it.
 
The nice thing about these setups you can use a closed down aperture because everything is connect therefore longer shutter speeds are not a problem




That's an issue with micro 4/3; diffraction kills image sharpness if you stop down much. Even f8 shows a slight drop in sharpness and f11 shows a severe loss of resolution. f16 is unusably soft. I do these at f5.6.


When I'm shooting normal photos with the OM-D E-M1 mark II, I usually use the 12-40mm f2.8 Olympus Pro lens. I rarely stop down beyond f5.6; with such short focal lengths that actually gives a lot of depth of field. At the longer end of the zoom range, I'll use f8 sometimes if I need the extra depth of field, but at f11, the lens becomes too soft because of diffraction.


That shocked me when I first got into m4/3. The day I got my first m4/3 camera, a Pen-F, I shot some pics in my neighborhood and did one at f16 because it was one that needed a lot of depth of field, and on a fullframe camera f1.6 would have been needed. The image had NO fine detail resolution. It looked like it had been shot with a pinhole camera! It was BAD. My heart sank, I thought I had wasted a lot of money on junk. Then I realized that the test shots done at wide apertures were tack sharp...it had to be diffraction. I was using a 25mm lens, so stopping down wasn't needed for a lot of DOF. Once I figured that out, I started getting incredible images.
 
For alignment: put a mirror at the film location. Stop down lens to get a small aperture. Turn on the grid in the camera screen. Use room lights, not light table. Center the aperture on the grid center.


You'll need to change exposure to get the aperture to show. And focus will need to be changed to get the aperture to be sharp.


At 50mp, I guess that's the Olympus JPG output? If you are doing RAW, be aware that not every RAW converter handles multiple files for high resolution. I am using a Pentax K3ii and Affinity photo. Then read about the high resolution issue. Tried another RAW program (RawTherapee) and sharpness and detail jumped.



Your images look pretty sharp, but figured I'd mention it.




Thanks for the tip on using a mirror, I'll try that!


I used Lightroom for my RAW conversions. It does a great job with the Olympus high-res mode.
 
That is what flare is. You're talking about light that isnt part of the image, right? That's what causes flare.


You're correct, that is what causes flare. But, any light entering the lens has to be taken into account by the camera sensor to get the proper exposure. Any 'extra' light that isn't coming in thru the film is not going to help your desired exposure of the film. It may not cause any visible flare, but the sensor and camera are taking it into account when rendering the image.
 
I'm basing my advice on the fact that my Nikon bellows slide copier attachment has a bellows that extends from the film holder to the front of the lens.
Just trying to help.
 




That's a great setup. Olympus made a slide copier like that for their film SLR system. Unfortunately, the Olympus one, like your Nikon, is made to be used with a macro bellows between the camera and lens. The slide copier attaches to the macro bellows, so you can't use the copier without it. I'm using Olympus's m4/3 macro lens, so you can't use the macro bellows.


Nikon makes a slide and neg copier now that is a metal tube that screws on the end of certain Nikon lenses and has a slot on the other end to slide in a film strip or slide holder. Much simpler than using a bellows, though for people like you and I who are using other cameras and lenses, you have to use adapters to fit it to the lens. I'd buy one, as I think it would solve my alignment issues and be easier to use than my Lego setup but they've been backrordered for months.
 
A quick and inexpensive solution for extraneous light could simply be toss a bit of black velvet over the film holder and the camera. Ala LF view cameras.
 
Back
Top