DSLR scanning 120 film

Omid_K · May 14, 2019

Hello all,

The power supply on my scanner has failed and while I’m awaiting a new one (2-3 months lead time) I’d like to use a DSLR setup to digitize my negatives. I’ve got a quadrapod, a Canon 5D3, an LED light table, Lomography Digitaliza (35 and 120) and a converted FD 50mm macro. So far I’ve shot my 35mm film in RAW and gotten 30 MB images. When I shoot 4 images for 6x6 or 6 images for 6x7 and stitch them in Lightroom I end up with 200 MB files.

Comparatively, my medium format scans from Indie Film Lab are around 30 MB (scanned at 4800 x 5900). That’s gotten me thinking that maybe I should be shooting in a lower resolution since the individual images are getting stitched together. Is that accurate? And if so, how low should I go? Canon offers me small, medium, and large JPEGs in addition to RAW. Can I Shoot medium JPEGs and have enough resolution to print 16”x20”? What about 11”x14”?

Then there’s the issue of JPEG vs RAW. I’d prefer RAW files but 200 MB files are too impractical for me. I shoot mostly medium format and have no desire to have such extreme file sizes even if it means compromising quality. I’m of the belief that my images will live or die based on the content, not some technical constraint. In my book, bigger is not automatically better.

Once my Imacon is back up and running I may end up sticking with this method as it seems much faster. Not to mention it takes up far less space than a scanner, G4 Mac, and monitor.

Thank you in advance for considering the question and offering any feedback.

Omid

bdial · May 14, 2019

If you divide the number of pixels you need for your print size, by the film image size, you will get the resolution needed for the scan. So, a 6x7 neg will need roughly 2100 pixels per inch to produce a 300 dpi 16x20 (assuming I did all the math right). Since you'd prefer RAW files, it's probably easiest to shoot at the camera's native resolution, stitch the 4 images together, then downsize the combined image to the final resolution. You'll be working with 200 MB to begin with, but once you save the processed file you should be back to the 30 MB range. If you're concerned with space make the combined file your original and ditch the 4 files used to create it. If the camera gives you resolution choices in RAW, then pick whatever pixel count is closest to 5900/2 for the wide dimension.
You might be able to get there using your camera's JPEG medium setting, but it depends on whether "medium" means more compression or fewer pixels in the original. If it's not documented, it would be easy enough to check by making a JPEG "medium" image, then checking the pixel dimensions in PS or whatever processing program you use. I'd guess that "small", "medium" and "large" refer to compression rather than pixel dimensions.

FWIW, 200 MB isn't extreme (IMHO), I have scans of some of my 4x5 negs that are bigger than 1 GB.

Adrian Bacon · May 14, 2019

Omid_K said:
Hello all,

The power supply on my scanner has failed and while I’m awaiting a new one (2-3 months lead time) I’d like to use a DSLR setup to digitize my negatives. I’ve got a quadrapod, a Canon 5D3, an LED light table, Lomography Digitaliza (35 and 120) and a converted FD 50mm macro. So far I’ve shot my 35mm film in RAW and gotten 30 MB images. When I shoot 4 images for 6x6 or 6 images for 6x7 and stitch them in Lightroom I end up with 200 MB files.

Comparatively, my medium format scans from Indie Film Lab are around 30 MB (scanned at 4800 x 5900). That’s gotten me thinking that maybe I should be shooting in a lower resolution since the individual images are getting stitched together. Is that accurate? And if so, how low should I go? Canon offers me small, medium, and large JPEGs in addition to RAW. Can I Shoot medium JPEGs and have enough resolution to print 16”x20”? What about 11”x14”?

Then there’s the issue of JPEG vs RAW. I’d prefer RAW files but 200 MB files are too impractical for me. I shoot mostly medium format and have no desire to have such extreme file sizes even if it means compromising quality. I’m of the belief that my images will live or die based on the content, not some technical constraint. In my book, bigger is not automatically better.

Once my Imacon is back up and running I may end up sticking with this method as it seems much faster. Not to mention it takes up far less space than a scanner, G4 Mac, and monitor.

Thank you in advance for considering the question and offering any feedback.

Omid

Within reason, more resolution is generally better. If you’re happy with the current stitched scans, then don’t change anything.

PhilBurton · May 15, 2019

Omid_K said:
Hello all,

The power supply on my scanner has failed and while I’m awaiting a new one (2-3 months lead time) I’d like to use a DSLR setup to digitize my negatives. I’ve got a quadrapod, a Canon 5D3, an LED light table, Lomography Digitaliza (35 and 120) and a converted FD 50mm macro. So far I’ve shot my 35mm film in RAW and gotten 30 MB images. When I shoot 4 images for 6x6 or 6 images for 6x7 and stitch them in Lightroom I end up with 200 MB files.

Omid

Omid,

I hope you don't mind a question. Are you shooting at 1:1 and therefore needing to stitch together? What about shooting so that the entire 120 negative is covered by one shot? I'm asking because I have some 120 negatives and slides, that other people shot for me at events like weddings.

Phil

Grim Tuesday · May 15, 2019

I strongly doubt that any converted 50mm lens is sharp to the edges at 1:1. My Nikkor 55 f2.8 micro isn't even perfectly sharp at the edges at .5:1, and my nikkor 55 f3.5 was an order of magnitude worse. I eventually gave up and got a V700, which is much faster and the quality is decent. More importantly I don't need to actively attend to it while it's working.

When I was switching over I did some tests on stitching vs V700. The order of quality is: 6 shots stitched > 4 shots stitched >> V700 == 2 shots stitched > 1 shot of a 120 square. This is with a d5500 and a Nikkor 55 f2.8. Any more than 6 shots stitched I ran into issues with corner sharpness and this weird pattern of sharp and un-sharp all over the image if you zoom in which is, after all the main point of ultra high res scanning methods.

One of my long term projects is building an XY table that could automatically advance a negative in a holder few millimeters at a time, and have only the center captured, and then have it automatically cropped and stitched. In other words, it would be a real scanner but with a DSLR as the capture device. I think it would rival a flextight and push into drum scanning quality territory, and should cost no more than $150.

GLS · May 15, 2019

I use my D810 + macro to digitise my 120 film. Unless I plan to make huge prints I find shooting multiple 1:1 shots and stitching to be unnecessary. A single captured frame of 6x6 or 6x7 filling the short dimension of the D810 sensor gives around a 20-22 MP image once cropped, which is more than enough for most uses.

Kino · May 15, 2019

Try this: https://www.technicolor.com/cinestyle

Yes, it is ostensibly for cinema production, but if you think outside the box, it can be very helpful for your application.

runswithsizzers · May 15, 2019

Many say 240 ppi is a satisfactory number for quality prints (and there are plenty of people who will tell you they made an excellent print with less than that).
11”x14” at 240 ppi is 2640x3360 pixels, and
16”x20” at 240 ppi is 3840x4800 pixels

I have not heard anyone say that there is any advantage to having more than 300 ppi when printing, so for your immediate needs, there is not much point in producing files bigger than 4800x6000 pixels (16”x20” at 300 ppi) unless you want to do some cropping.

I have been photographing my 120 b&w negatives with a 16MP Fuji camera that has a maximum image size of 4896×3264 pixels. So far, I have been taking only one shot, and after croping my images are about 3100x3100 pixels. At 240ppi that would make a 13"x13" print, but I don't do much printing.

My work flow is to take the RAW image from the camera into Photoshop and work with it as 16-bit grayscale for cropping, inversion, contrast adjustments, highlight / shadow adjustments, and sharpening. I also prefer to dust-spot in Photoshop (vs. Lightroom). After I have the image pretty much the way I want it, I convert it from a 16-bit TIFF to an 8-bit TIFF, then Import it into Lightroom for storage, keywords, captions, toning, etc.

Straight from the camera (before cropping) my 4896 × 3264 pixel RAW files are about 33-34MB; after cropping to 3100x3100, my 16-bit color images are about 30MB; converted to 16-bit grayscale the images are about 15-20MB, and as 8-bit TIFFs they are about 10MB.

BTW I use an enlarger lens on bellows. More affordable than a macro lens, and designed for a flat field. Some results <here>.

PhilBurton · May 15, 2019

GLS said:
I use my D810 + macro to digitise my 120 film. Unless I plan to make huge prints I find shooting multiple 1:1 shots and stitching to be unnecessary. A single captured frame of 6x6 or 6x7 filling the short dimension of the D810 sensor gives around a 20-22 MP image once cropped, which is more than enough for most uses.

Which lens?

Phil

GLS · May 16, 2019

PhilBurton said:
Which lens?

Phil

Tokina 100mm macro (https://www.amazon.co.uk/Tokina-100mm-Macro-AT-X-100AF/dp/B000CMNL52).

Any decent comparable macro would do the job just as well I expect. However, I will say it is important that it have autofocus so that you can focus on the film grain via a cable release without having to touch the camera.

PhilBurton · May 16, 2019

GLS said:
Tokina 100mm macro (https://www.amazon.co.uk/Tokina-100mm-Macro-AT-X-100AF/dp/B000CMNL52).

Any decent comparable macro would do the job just as well I expect. However, I will say it is important that it have autofocus so that you can focus on the film grain via a cable release without having to touch the camera.

I have an old Nikon 50 mm enlarging lens and a 55 mm f2.8 manual focus lens. I imagine that I would set up my camera on a tripod, set up the negative in a frame on a light box, and then focus just once. Is that approach flawed?

Phil

GLS · May 16, 2019

PhilBurton said:
I have an old Nikon 50 mm enlarging lens and a 55 mm f2.8 manual focus lens. I imagine that I would set up my camera on a tripod, set up the negative in a frame on a light box, and then focus just once. Is that approach flawed?

Phil

I can only speak to my experience using a macro lens. The two options you suggest may work, but I couldn't vouch for the results. Try it and see.

A tripod should also work, but it may be more difficult to get the sensor perfectly aligned with the film plane that way. I use a copy stand instead. Focusing manually is of course possible in principle, but in practice requires much trial and error, inconsistency and frustration. I have found that touching the camera/lens in any way whilst setting focus is enough to move the camera a sufficient amount that when you then release the camera again the film is no longer in critical focus; hence this approach requires trying to anticipate the amount of focus offset to use whilst focusing (not fun at all, believe me). Instead I use a cable release and live view in 100% to focus on the film grain. After setting focus in this way I then decouple the focus mechanism from the shutter release in the DSLR's menu, which means it won't try to re-acquire focus every time you trip the shutter.

I have previously written a detailed run-down of my method here: https://www.photrio.com/forum/threads/neg-scans-grain-with-fp4.157955/page-3#post-2137561

PhilBurton · May 16, 2019

GLS said:
I can only speak to my experience using a macro lens. The two options you suggest may work, but I couldn't vouch for the results. Try it and see.

A tripod should also work, but it may be more difficult to get the sensor perfectly aligned with the film plane that way. I use a copy stand instead. Focusing manually is of course possible in principle, but in practice requires much trial and error, inconsistency and frustration. I have found that touching the camera/lens in any way whilst setting focus is enough to move the camera a sufficient amount that when you then release the camera again the film is no longer in critical focus; hence this approach requires trying to anticipate the amount of focus offset to use whilst focusing (not fun at all, believe me). Instead I use a cable release and live view in 100% to focus on the film grain. After setting focus in this way I then decouple the focus mechanism from the shutter release in the DSLR's menu, which means it won't try to re-acquire focus every time you trip the shutter.

I have previously written a detailed run-down of my method here: https://www.photrio.com/forum/threads/neg-scans-grain-with-fp4.157955/page-3#post-2137561

Thanks for the link. Great thread. I've subscribed. And now I'm thinking about that copy stand approach.

runswithsizzers · May 16, 2019

At first I set the tripod on the floor and the light box on a table. Worked, but just barely because getting the light table under the camera was difficult.

My tripod allows me to invert the central column, and I found the arrangement below to work much better, with everything on a table top.

One trick for parallel alignment is to use a bullseye bubble level (you can see mine just to the left of the rocket blower). If the tabletop is not level, shim the light table to make it level. Then, put the bubble level on the back of the camera and level that. Just takes minutes to make everything parallel.

As for focus, I use manual. My Fuji mirrorless camera has a focus aid that lets me set the LCD/EVF to monochrome, and the in-focus areas can be picked out in red. Very easy to read, especially if I zoom in to 100%.

The setup below is using a 100mm enlarger lens, and you can see the working distance required to fill the short dimension of my APS-C sensor with a 120 negative. Obviously, for 135 film - or if you want to merge multiple shots taken at high magnification - then the working distance would be closer. Likewise, the working distance would be closer if you use a shorter focal length lens.

There is a link in my post #8 if you want to see the results I get with this setup.

jmoche · May 21, 2019

Hi Omid...

How are you moving the neg between each exposure? I've been working on doing the same thing to "scan" 6x9 negs, but I haven't come up with a consistent method of moving the neg without buying things that I don't already have. Thanks!

John Wiegerink · May 22, 2019

runswithsizzers said:
At first I set the tripod on the floor and the light box on a table. Worked, but just barely because getting the light table under the camera was difficult.

My tripod allows me to invert the central column, and I found the arrangement below to work much better, with everything on a table top.

One trick for parallel alignment is to use a bullseye bubble level (you can see mine just to the left of the rocket blower). If the tabletop is not level, shim the light table to make it level. Then, put the bubble level on the back of the camera and level that. Just takes minutes to make everything parallel.

As for focus, I use manual. My Fuji mirrorless camera has a focus aid that lets me set the LCD/EVF to monochrome, and the in-focus areas can be picked out in red. Very easy to read, especially if I zoom in to 100%.

The setup below is using a 100mm enlarger lens, and you can see the working distance required to fill the short dimension of my APS-C sensor with a 120 negative. Obviously, for 135 film - or if you want to merge multiple shots taken at high magnification - then the working distance would be closer. Likewise, the working distance would be closer if you use a shorter focal length lens.

There is a link in my post #8 if you want to see the results I get with this setup.

I have this same setup, but with copy stand instead of tripod. Admittingly I have never used it and it's only in reserve just in case. I have a very good, very capable Nikon LS8000 and use it. I do know these Nikon scanners don't last forever and repairs/parts won't be there when mine pukes. Nice to have a backup.

PhilBurton · May 22, 2019

I've been thinking about this whole thread for a bit. With my Nikon 5000 scanner, I can get a 4000 dpi scan. That's roughly 6000 x 4000 pixels for a 35 mm neg/slide, and with max-max settings, the resulting TIFF would be about 110 MB. Now if I scale that up to 2 1/4" square, I would want a TIFF of about 370 MB. But with my Nikon D3, a 14-bit uncompressed RAW file is only 25 to 30 MB, meaning roughly 800 dpi. Maybe with a larger source negative, I don't really need to scan a 4000 dpi. However, I remember the gorgeous 16 x 20 prints my friend made with a Hasselblad from a Pan-X negative. Smooth beyond anything possible with 35 mm. So maybe I would want to still do a 4000 dpi scan.

I thought about all this when I considered the possibility of buying a Bowens Illlumitran to use with my Nikon D3.

Comments?

MattKing · May 22, 2019

PhilBurton said:
I've been thinking about this whole thread for a bit. With my Nikon 5000 scanner, I can get a 4000 dpi scan. That's roughly 6000 x 4000 pixels for a 35 mm neg/slide, and with max-max settings, the resulting TIFF would be about 110 MB. Now if I scale that up to 2 1/4" square, I would want a TIFF of about 370 MB. But with my Nikon D3, a 14-bit uncompressed RAW file is only 25 to 30 MB, meaning roughly 800 dpi. Maybe with a larger source negative, I don't really need to scan a 4000 dpi. However, I remember the gorgeous 16 x 20 prints my friend made with a Hasselblad from a Pan-X negative. Smooth beyond anything possible with 35 mm. So maybe I would want to still do a 4000 dpi scan.

I thought about all this when I considered the possibility of buying a Bowens Illlumitran to use with my Nikon D3.

Comments?

My head hurts.....
The file size is essentially irrelevant to the question of scan quality, save and except when you consider issues respecting file handling and file storage.
And "dpi" (dots per inch) is relevant to printing, not scanning, although people seem to use it sometimes in replacement for the more appropriate "ppi" (pixels per inch).
The following includes some simplifications, but is generally accurate.
If you scan a 2 1/4" x 2 1/4" (actually 2.2" x 2.2") negative at 4000 ppi, you end up with a file that is approximately 8800 x 8800 pixels. That file is appropriate for printing a 29.3" x 29.3" print, using a 300 dpi printer. That is a pretty big print.
Admittedly, in order to get a 8800 x 8800 file from your D3, you need to stitch together at least 8 images - not something you would want to do frequently, as it would take a lot of time and work.
The native resolution of your D3 yields a 2832 x 2832 square image - sufficient for a nice 9.5" x 9.5" print.

PhilBurton · May 22, 2019

MattKing said:
My head hurts.....
The file size is essentially irrelevant to the question of scan quality, save and except when you consider issues respecting file handling and file storage.
And "dpi" (dots per inch) is relevant to printing, not scanning, although people seem to use it sometimes in replacement for the more appropriate "ppi" (pixels per inch).
The following includes some simplifications, but is generally accurate.
If you scan a 2 1/4" x 2 1/4" (actually 2.2" x 2.2") negative at 4000 ppi, you end up with a file that is approximately 8800 x 8800 pixels. That file is appropriate for printing a 29.3" x 29.3" print, using a 300 dpi printer. That is a pretty big print.
Admittedly, in order to get a 8800 x 8800 file from your D3, you need to stitch together at least 8 images - not something you would want to do frequently, as it would take a lot of time and work.
The native resolution of your D3 yields a 2832 x 2832 square image - sufficient for a nice 9.5" x 9.5" print.

For me the issue is not making a print at full size. It's the ability to pull out image detail by cropping. What's the point of the larger negative if I can't get more detail than I can with a 35 mm negative (or slide)?

Adrian Bacon · May 25, 2019

GLS said:
I use my D810 + macro to digitise my 120 film. Unless I plan to make huge prints I find shooting multiple 1:1 shots and stitching to be unnecessary. A single captured frame of 6x6 or 6x7 filling the short dimension of the D810 sensor gives around a 20-22 MP image once cropped, which is more than enough for most uses.

Similar here. I've said it before, and I'll say it again: In general more resolution is a good thing, however, there is definitely a point of diminishing returns when it comes to stitching. For the vast majority of uses, a single capture of the frame at 20+ MP and a reasonably decent lens provides pretty amazing detail. I'm always amazed at the lengths some people go to try to get ever more detail out of a frame of film. If that's what you have to have, then by all means, go right ahead, however, be prepared to spend a lot more time futzing around with it than you probably realize, and run into all kinds of quality problems that you either didn't think about or didn't think would be an issue until they are.

For the amount of time and effort, it's not for me. A whole frame in a single capture is way faster and good enough for the vast majority of uses. In fact, if I need that quality, it's just faster and simpler to shoot a larger format and scan it with a flatbed. 4x5 on a modern epson V series scanner is a *lot* of resolution. 8x10 is just ridiculous.

etn · May 25, 2019

GLS said:
Tokina 100mm macro (https://www.amazon.co.uk/Tokina-100mm-Macro-AT-X-100AF/dp/B000CMNL52).

Any decent comparable macro would do the job just as well I expect. However, I will say it is important that it have autofocus so that you can focus on the film grain via a cable release without having to touch the camera.

I use autofocus too (Nikon 60mm AF-D), although I think a manual focus would work too. I converted an old enlarger into a copy stand: the film-camera distance can be adjusted much more easily than with a tripod. The camera is also much more stable and less subject to wobble.

I never had corner sharpness issues, but 1) the 60mm Nikon has a flat field. This is ideal when the object to be photographed is flat as is the case here. And 2) I close down the aperture. Resulting exposure times are around 1 second but it is not a problem.

I found that bracketing the exposure brings some advantages: you can chose the best exposed shot, or even combine multiple shots using HDR techniques.

My 2 cents,
Etienne

Billy Axeman · May 25, 2019

I still like the setup from @runswithsizzers (post #14). It is an elegant solution, with a minimum of material, cost and time to set it up. It is also very flexible to choose all the parts just what you have in stock or your budget.

The only thing it still needs is decent film-holder with quick feed-through, because with the cardboard frames it is much prone to dust and it would work faster.

I bet a tripod is inherently much stabler than most copy stands, the latter colums tend to vibrate and are often tied to a wall to dampen it. If you take a short 50mm (macro) lens to reduce the dimensions of the tripod it further increases stability, and to prevent kicking it off the table too easy.

Huss · May 25, 2019

I use a copy stand, Nikon Z7 w/ 60mm 2.8G macro, light pad and lomo digitliza film holder.
With the 45mp sensor, it gives me all I want. The copy stand ($150 on ebay) is much quicker and easier to use than cobbling up the tripod routine (which I tried first)

No need for a cable release - when I used a D750 and then D850, I would use the mirror up function along with the delayed release.
On the mirrorless Z7, just the delayed release.

GLS · May 25, 2019

Huss said:
No need for a cable release - when I used a D750 and then D850, I would use the mirror up function along with the delayed release.
On the mirrorless Z7, just the delayed release.

The main point of using the cable release is so that the autofocus can be engaged without having to touch the camera. At these distances, touching the camera in any way to focus moves the camera enough such that when it is released again the focus will no longer be dead on. This is at least true with my Kaiser RS2XA copy stand (which is hardly a cheap model).

Adrian Bacon · May 25, 2019

GLS said:
The main point of using the cable release is so that the autofocus can be engaged without having to touch the camera. At these distances, touching the camera in any way to focus moves the camera enough such that when it is released again the focus will no longer be dead on. This is at least true with my Kaiser RS2XA copy stand (which is hardly a cheap model).

I use a copy stand and a 100mm cine macro lens which has a declicked aperture ring on the barrel next to the focus ring. There’s no autofocus. I open the aperture up all the way and manually focus it, then close the aperture down to its sharpest setting, which for that lens is f/11. I use an enlarger film carrier to hold the film and a strobe to light the film.

I also use a cable release, but this is mainly so I can trigger it with my foot, which lets me have both hands free. I don’t really have focus issues as at f/11 for 120 film, the DOF is roughly a 1/4 inch with that lens/camera combo. I also don’t really have movement issues as I run the shutter at 1/250, set the ISO to the best DR/least noise, then adjust the strobe to get the best exposure. The strobe time is way up in the 1/1000 second range or faster and the resulting exposure is several stops above ambient room light so I don’t have light contamination problems either. The strobe is also 5500K and the CRI is 100, which is nice.

DSLR scanning 120 film

Moderator

Privacy & Transparency

Privacy & Transparency