Spanik
Well-known
I love this setup and I'm very jealous - but it required quite a bit of construction work, and some specialized bits-and-pieces, so it's effectively disqualified from the thread.
But if you want you could do a kickstarter with a ready-made model which we could all buy :angel:
Not really that much specialised bits. Enlargers can be had for a bit of spare change if you don't just get them if you take the time to go get them. Whatever the result of this thread, you will always have to do something yourself.
mani
Well-known
The idea is to come up with a generic setup, so that anyone reading the thread in the future can put together the optimum selection of parts by buying them off eBay, Amazon and maybe even their local flea market, and then attaching them together in a way that doesn't mess up the original intention in some unforeseen way.
I think many of the contributions to the thread so far have provided a lot of those answers: I like the thought of re-using Durst enlarger stands together with a camera-holding adapter, for instance, because not only does this involve no metalworking, it also recycles currently unwanted film-photographic equipment in the service of a new wave of film interest.
NEW EDIT: anyone following the thread from the start will see that one contributor has removed all of their posts and also some of my answers. I'm re-editing so that the entire thread makes more sense.
Jason Sprenger
Well-known
Given the constraints, a new title for the thread to something other than "definitive" seems to be in order.
Might I suggest "Constructing a low-cost DSLR film scanning setup".
In this way, folks looking to exceed flat-bed scanner results with copy stands and dedicated macro lenses won't confuse the setup definitive for your purpose with the ones definitive for their purpose.
Might I suggest "Constructing a low-cost DSLR film scanning setup".
In this way, folks looking to exceed flat-bed scanner results with copy stands and dedicated macro lenses won't confuse the setup definitive for your purpose with the ones definitive for their purpose.
mani
Well-known
NEW EDIT: anyone following the thread from the start will see that one contributor has removed all of their posts and also some of my answers. I'm re-editing so that the entire thread makes more sense.
Cost isn't supposed to be a factor.
quejai
Established
I think this thread would be great if people could post 100% crops of their scans in the center and in the corners, so we can more accurately know what each persons' setup is capable of.
philipus
ʎɐpɹəʇɥƃı&
What would be the best way to scan larger format film? Taking the example of 120 film, would it be to simply move the frame and shoot, say, four quadrants? Would the rig need to accommodate somehow precisely how these frames are shot to facilitate the stitching process in post?
I'm very much considering a digitizing setup for the future but would like it to deal very well with larger film formats too.
br
Philip
I'm very much considering a digitizing setup for the future but would like it to deal very well with larger film formats too.
br
Philip
mdarnton
Well-known
Personally, I'm trying to get scans that will provide inkjet-printed results either replicate, at the minimum, or improve on what I used to be able to do with silver. That's the only process I compare with. It that regard I've been completely satisfied with camera scanning. For me it has nothing to do with ultimate resolution and wall-size prints that I never will need to make, and everything to do with tonality and adequate sharpness for the job I intend to do.
Isn't that how it's supposed to be???
Isn't that how it's supposed to be???
calebarchie
Established
I am mainly speaking for the faux ideology that gets thrown around.
After fiddling with a lot of scanner options I got a drum scanner because of the control it offers, rather than some kind of subjective 'look', super-resolution or anything.
Things inherent to the technology, the fact it is a continuous tone process, colour resolution/bit depth, adjustable aperture and exposure to suit different grain structures among other things.
These things seem to be disregarded with 'high end' DSLR scanning solutions - just the bit depth and bayer array sampling are limiting enough.
PS call me a freak but I enjoy scanning, as with the whole picture making process from thinking, tripping the shutter to printing.
Nothing is supposed to be anything
After fiddling with a lot of scanner options I got a drum scanner because of the control it offers, rather than some kind of subjective 'look', super-resolution or anything.
Things inherent to the technology, the fact it is a continuous tone process, colour resolution/bit depth, adjustable aperture and exposure to suit different grain structures among other things.
These things seem to be disregarded with 'high end' DSLR scanning solutions - just the bit depth and bayer array sampling are limiting enough.
PS call me a freak but I enjoy scanning, as with the whole picture making process from thinking, tripping the shutter to printing.
Nothing is supposed to be anything
calebarchie
Established
That DSLR scanning ≥ Drum scanner (or close, comparable, etc). Depends on who you ask and where you look.
I have a scanmate 4000, I got it for 1/5 the price of a MF coolscan. Definitely not for everyone though.
Anyway, gone off-topic enough, lets resume to DSLR scanning set ups.
cz23
-
I see there are lightbox apps for the iPad. The surface does not seem very bright to me, but this solution is recommended across the web. Klaus, are you satisfied with this approach? Anyone else lighting this way?
John
Sorry to see that this thread has fizzled out, as it could be the basis for a practical scanning solution — and the scanning issue certainly is an impediment for many that are considering coming back to shoot film.
A question: what dMax would this type of negative-copying solution have? I'm interested in a dMax of 4.2, which is that of the old Imacon Precision II/III.
A question: what dMax would this type of negative-copying solution have? I'm interested in a dMax of 4.2, which is that of the old Imacon Precision II/III.
mdarnton
Well-known
When I shoot my B&W negs, they hardly fill half the histogram, so there's plenty of room for more contrast. I don't know what that means in terms of dMax.
calebarchie
Established
Histograms mean little. Dmax and density range has got to do with the bit-depth of the A/D converter for the chip. A density range of 4.2 requires a 14-bit capable camera sensor.
SayCheese
Established
Well, it works, somehow. But it's not the best way to do it, IMHO. First, you've got to turn the display to max. brightness and Auto-Luminance to "off". Then, you've got to maintain some distance between the iPad and the negative. Otherwise, you're going to see the dotmatrix of the display in the scans, even with a "retina" equiped device. Using a 90mm lens at f/8, 3-4cm are enough.
But anyway, the exposure times are way to long. I'm considering a lightpad or a speedlight for the next rebuild of my scanning setup.
K
mdarnton
Well-known
Histograms mean a lot if they indicate that only the tiniest amount of the camera's dynamic range is being used.
quejai
Established
Try to use a brighter light source, that should widen the dynamic range seen by your dslr
Here is an article on digitalizing negatives using a camera versus scanning. It's undated but seems to have been written in 2014 or later:
https://luminous-landscape.com/articleImages/CameraScanning.pdf
One of the authors (Mark Segal) wrote an interesting Epson V850 review, comparing it to well-known film scanners. Looks like there are a good number of issues that need to be dealt with in digitalizing negatives with a camera, including how to invert B&W negatives to positives, how to deal with color negative film and how to deal with vibration. Another issue that I saw in an article on the BEOON is dust landing on the negative. Looks like digitalizing with a camera has a lot of issues that need to be dealt with that I had not realized, and therefore looks more complicated than I thought.
https://luminous-landscape.com/articleImages/CameraScanning.pdf
One of the authors (Mark Segal) wrote an interesting Epson V850 review, comparing it to well-known film scanners. Looks like there are a good number of issues that need to be dealt with in digitalizing negatives with a camera, including how to invert B&W negatives to positives, how to deal with color negative film and how to deal with vibration. Another issue that I saw in an article on the BEOON is dust landing on the negative. Looks like digitalizing with a camera has a lot of issues that need to be dealt with that I had not realized, and therefore looks more complicated than I thought.
etn
Newbie
Team,
thank you for this interesting thread.
I have been reading this forum for quite a while, even registered, but hardly posted anything.
I am mostly a medium format shooter, although I have a Zeiss Ikon ZI which I use from time to time.
Here is my experience with DSLR scanning. Remember, I only express my opinions based on my
experience. Your experience and therefore your opinions may vary. Disclaimer: do not flame me
if you disagree
General considerations
DSLR scanning seems to be a valid method - have a look at this video, even pros use it:
https://www.youtube.com/watch?v=UD12e8Gb_Jw
Obviously, those guys use a rig whose price is way above what mere mortals like us can afford.
But this convinced me that DSLR scanning, even if it does not reach the quality of Hasselblad X1
or drum scanning, is still a valid method - at least from some use: social media, small-size printing, etc.
In the rare occurrences when I want to print big I have the film scanned on a Hasselblad scanner.
Hardware
The ideal hardware is, in my opinion, a repro stand allowing to easily adjust the height of the camera
to maximize the usage of the sensor.
I have been successful with an LED light box from Kaiser, similar to this one:
http://www.kaiser-fototechnik.de/en/produkte/2_1_produktanzeige.asp?nr=2450
(mine is an older model). Illumination is even. Do not use an iPad or a computer screen: you will
see the pixel structure on your scans!
I use a Nikon DSLR but I guess that anything can work here.
A flat field macro lens is a must. I use a Micro-Nikkon 60mm AF-D. Although I have owned mine for decades,
those are available used for a few hundred $/€. I use the camera in live view and autofocus on each scan
somewhere in the middle of the frame. I do an exposure bracketing from -2 to +2 EV and select
the best exposed frame in post. (Auto exposure with aperture-priority mode).
I use the lowest possible ISO (for maximum dynamic range) and the optimal aperture
of the lens e.g. f/8 or f/11. Under these conditions exposures range from 1/2s to a couple seconds
depending on the brightness of the scanned picture.
I realized that the lens is very sensitive to flare from light coming from outside the scanned frame.
I thus created a mask out of black cardboard. I also scan in a dimmed room to avoid light reflections
on the film.
Scanning glass-mounted slides is the easiest. I place them onto the light box, "black side" of the GEPE mount
facing up. This is the "anti-newton" side of the glass mount facing the screen when projecting the slides.
Scanned pics are left-right reversed but this is easily corrected in post.
Scanning unmounted film might be marginally trickier, you have to ensure film flatness.
Software / Post-processing
This is in my opinion the trickiest part of the game.
There is a special edition of Capture One which is dedicated (among others) to film scanning.
Dubbed "Capture One Cultural Heritage", it is the ideal software for this task with auto-crop,
auto-rotate, negative-to-positive conversion, etc.
It also costs $5000 so forget it. It would be great if Digital Transitions, the editor of the software,
would propose a reduced version without a few pro features for a few hundred $ maybe.
But if I were to fork out $5000, I'd rather put that money towards a Hasselblad X1.
Here is the workflow I use:
1) starting from the RAW files, I select the best exposed frame.
(in the future I plan to play with HDR. I am not there yet)
I use Capture One for this but here again, I guess any software will do.
2) For slides: crop, levels and curves adjustment, (left-right flip if necessary, see above).
Pretty straightforward. Do not forget to adjust the color temperature to the color temp of your light source!
(5300K in my case)
For BW negatives: similar process. Crop, invert using the levels tool, brightness/contrast/curves adjustment.
Quite straightforward too.
Color negatives: this is the tricky part. Color negative film has a color cast which cannot be removed
by just changing the color temperature.
The rest of the workflow description applies to color negatives only.
3) Export the best exposed frame from Capture One into a 16-bit TIFF. Ensure there is no clipping
in the histogram as this will have an influence on the color rendition.
4) I am developing a small Matlab script which does the following:
- open all TIFF files in a directory
- remove the color cast due to film die
- invert the picture
- perform histogram equalization
- save the resulting pictures as TIFF files.
The (commercial) Matlab software has great image processing functions. There are open-source alternatives (Scilab, Octave)
with similar functionality, so I guess it can be done with those too. My script is still under development but
it already brings usable results. Still a lot of work to be done here, and many ideas for improvement:
e.g. auto-crop/auto-rotate. Auto-detection of the color cast. Better automatic color correction etc.
5) A final adjustment is done in photoshop. The Auto Color feature does wonders here.
Hope this helps - let me know if you want further info!
Thanks
Etienne
thank you for this interesting thread.
I have been reading this forum for quite a while, even registered, but hardly posted anything.
I am mostly a medium format shooter, although I have a Zeiss Ikon ZI which I use from time to time.
Here is my experience with DSLR scanning. Remember, I only express my opinions based on my
experience. Your experience and therefore your opinions may vary. Disclaimer: do not flame me
if you disagree
General considerations
DSLR scanning seems to be a valid method - have a look at this video, even pros use it:
https://www.youtube.com/watch?v=UD12e8Gb_Jw
Obviously, those guys use a rig whose price is way above what mere mortals like us can afford.
But this convinced me that DSLR scanning, even if it does not reach the quality of Hasselblad X1
or drum scanning, is still a valid method - at least from some use: social media, small-size printing, etc.
In the rare occurrences when I want to print big I have the film scanned on a Hasselblad scanner.
Hardware
The ideal hardware is, in my opinion, a repro stand allowing to easily adjust the height of the camera
to maximize the usage of the sensor.
I have been successful with an LED light box from Kaiser, similar to this one:
http://www.kaiser-fototechnik.de/en/produkte/2_1_produktanzeige.asp?nr=2450
(mine is an older model). Illumination is even. Do not use an iPad or a computer screen: you will
see the pixel structure on your scans!
I use a Nikon DSLR but I guess that anything can work here.
A flat field macro lens is a must. I use a Micro-Nikkon 60mm AF-D. Although I have owned mine for decades,
those are available used for a few hundred $/€. I use the camera in live view and autofocus on each scan
somewhere in the middle of the frame. I do an exposure bracketing from -2 to +2 EV and select
the best exposed frame in post. (Auto exposure with aperture-priority mode).
I use the lowest possible ISO (for maximum dynamic range) and the optimal aperture
of the lens e.g. f/8 or f/11. Under these conditions exposures range from 1/2s to a couple seconds
depending on the brightness of the scanned picture.
I realized that the lens is very sensitive to flare from light coming from outside the scanned frame.
I thus created a mask out of black cardboard. I also scan in a dimmed room to avoid light reflections
on the film.
Scanning glass-mounted slides is the easiest. I place them onto the light box, "black side" of the GEPE mount
facing up. This is the "anti-newton" side of the glass mount facing the screen when projecting the slides.
Scanned pics are left-right reversed but this is easily corrected in post.
Scanning unmounted film might be marginally trickier, you have to ensure film flatness.
Software / Post-processing
This is in my opinion the trickiest part of the game.
There is a special edition of Capture One which is dedicated (among others) to film scanning.
Dubbed "Capture One Cultural Heritage", it is the ideal software for this task with auto-crop,
auto-rotate, negative-to-positive conversion, etc.
It also costs $5000 so forget it. It would be great if Digital Transitions, the editor of the software,
would propose a reduced version without a few pro features for a few hundred $ maybe.
But if I were to fork out $5000, I'd rather put that money towards a Hasselblad X1.
Here is the workflow I use:
1) starting from the RAW files, I select the best exposed frame.
(in the future I plan to play with HDR. I am not there yet)
I use Capture One for this but here again, I guess any software will do.
2) For slides: crop, levels and curves adjustment, (left-right flip if necessary, see above).
Pretty straightforward. Do not forget to adjust the color temperature to the color temp of your light source!
(5300K in my case)
For BW negatives: similar process. Crop, invert using the levels tool, brightness/contrast/curves adjustment.
Quite straightforward too.
Color negatives: this is the tricky part. Color negative film has a color cast which cannot be removed
by just changing the color temperature.
The rest of the workflow description applies to color negatives only.
3) Export the best exposed frame from Capture One into a 16-bit TIFF. Ensure there is no clipping
in the histogram as this will have an influence on the color rendition.
4) I am developing a small Matlab script which does the following:
- open all TIFF files in a directory
- remove the color cast due to film die
- invert the picture
- perform histogram equalization
- save the resulting pictures as TIFF files.
The (commercial) Matlab software has great image processing functions. There are open-source alternatives (Scilab, Octave)
with similar functionality, so I guess it can be done with those too. My script is still under development but
it already brings usable results. Still a lot of work to be done here, and many ideas for improvement:
e.g. auto-crop/auto-rotate. Auto-detection of the color cast. Better automatic color correction etc.
5) A final adjustment is done in photoshop. The Auto Color feature does wonders here.
Hope this helps - let me know if you want further info!
Thanks
Etienne
Just received the Leitz BEOON stand from someone in Oz. Digitalizing this was is fast, of course; and the BEOON is smaller and lighter than I thought it would be — that makes it portable enough for me to take it on my nomadic trips to three continents each year. However, I right now, I'm using a Logan/flourescent light box. I find the focusing with the BEOON not that easy — does anyone know whether it would be easier with an LED light box?
mikyor1
Established
My biggest question isn't the set up but how to remove that damn color cast. Drives me nuts getting color the way i want