I’d prefer a macro lens for this, because they are indeed supposed to have less field curvature, and (as fixed focus lenses) are easier to correct for abberations, so can give a better overall image quality.
Then again, depending on intended use of the final image, you could only fill say half the image with the slide to copy, and crop in post. That removes the edges and corners, most likely to suffer from lens imperfections. You don’t need 50Mp for display on even a 4k screen…
A (zoom) lens with a macro setting, or macro in the name, is not a macro lens, btw…
But, I see all the focus on the lens side of the copying. How do you ensure a homogenous illumination of the slide to copy? Any irregularities will be visible in the copy…
Yes. I also have an old film era slide copying tube that came with something or other. I don’t know if that could be used or not. They are peanuts on eBay but maybe there’s a reason for that…
This is what I have but there are cheaper copycat items on eBay now:
No that Canon lens is not a macro, but you can see it’s been mounted on an extension tube (between the lens and the camera body). Extension tubes decrease focus distance and light transmission along with it.
The long snout on the slide attachment is likely all part of the same component that might attach to the bayonent mount used for lens hoods or screw into the filter thread.
While I would agree that not every lens with a macro setting is necessarily a macro lens (dependent on the exact definition of “macro lens”), I wonder what exactly your definition of a macro lens is. Wikipedia defines in Macro photography - Wikipedia a macro lens as a lens with at least a 1:1 reproduction ratio, but is not too strict about this term. Is there something that wikipedia misses? And in particular, why can’t a zoom lens be a macro lens at the same time?
The reproduction ratio is one big reason. The other is that “real” macro lenses are corrected for the close focusing distances (but they still are good at larger focusing distances). As the “macro” designation helps sell lenses, it tends to be a bit over-used…
So just giving a lens the possibility to focus at a short distance isn’t enough to make it a macro lens. The wikipedia article you cite does maintain that the lenses need to reach at least 1:1 reproduction. There are a few lenses that exceed that (e.g. Canon has one), but 1:1 is close enough for most users, if only because the working distance gets very short. At 1:1 the distance between object and sensor is ~ 4x the focal lenght(*) of the lens, the front lens of the objective is at less than half that distance. This page lists a few of the more spectacular macro lenses.
(*: to be pedantis, that is focal lenght when focused at infinity)
All that said, it doesn’t mean that a zoom with a macro setting isn’t interesting. I think it’s actually a good option for certain subjects, where you can’t get close enough otherwise. Larger insects like butterflies and dragonflies come to mind: they are either in places where you cannot get close enough for a “real” macro lens, or too skittish. A 300 mm “macro” helps in such cases…
Thanks for the clarifications, @rvietor. So a strict definition would be to reach at least 1:1 ratio and to be corrected (I assume mainly distortion, plus maybe flat field and some more aberrations) for this setting in particular. Is there any technical reason why this cannot be accomplished for a zoom lens, or is it just that there was none seen in the wild so far? I really want to understand this and you were very clear about it, therefore I am asking again.
May I additionally ask for a derivation or source? I really try to learn this, because there’s too much try-and-error in my macro photography attempts, but there are so many sources out there where you’re eventually finding out that they do this or that simplification (without telling) and are therefore not really helpful (focus at infinity assumption would be such a topic).
Fairly elementary physics, see e.g. wikipedia. The assumption here is that the objective behaves as a single lens.
One practical consequence is that focussing can become difficult when you have the camera on a tripod. If you plan to do a lot of macro from a tripod, a focusing rail is a very useful addition. (more so when you want to do focus stacking)
I do it not too often, therefore I always struggle because I don’t know the limits of my equipment and which of it to choose. As I do not own a macro lens, I always have to fiddle with extension tubes. At least, I do own a focusing rail.
However, what I think about more often and what is the reason for following this thread is “DSLR scanning”, which would also require a close to 1:1 repro ratio for 35 mm film. But what I am thinking about is reverse engineering what a Noritsu or Fuji lab “scanner” are doing which requires much more than a good lens, I am especially thinking about independent r, g and b illumination, IR illumination for scratch removal and so on. Maybe someday I find the time to write down what I already found out, as this is a really interesting topic …
This is … very sophisticated.
This video - the presenter uses different lenses one being from a projector.
The interesting part for me however is the film sample. He evaluates the sharpness of the lens before shooting the negatives.
I have no idea where I can take such a sample from. But a workaround can be a soft plastic transparent ruler or protractor. (I am open for suggestions what else can work)
Macro lenses are good but quite expensive. I am thinking with experimenting with a regular zoom lens that has “macro capabilities” (very small capabilities)
My project is on hold however for close to a year as even if I am to start scanning - my disk is going to be full so… will have to wait until I can buy a bigger one.
I will share my build of a scanning stand as soon as i receive my first roll of 120 back from the lab. Yesterday i sent my first roll of 120 to the lab. At the moment i think of a tripod, my d750 wit the 105 micro and e led light from Aldi.
At first my setup will be not very sophisticated as i am not sure how many rolls of film i will scan in a year.
I am especially curious how uniform the light will be.
Keep in mind that you need the 1:1 repro ratio for 35 mm film if you use a full-frame camera.
With a crop sensor, the ratio you need will depend on your sensor, but will be 1:1.5 or less…
That could stil give an advantage to a macro lens (after all, they are supposed to be designed for close-up work).
I know, but I think this can be achieved without the need for a scanner with the price tag of a car. Furthermore, the resolution of the fuji is a little limited (noritsu is a bit better in that regard I think). And the benefits seem huge, especially for color film. There might be no need to remove the orange mask after scanning, and the separation of color channels can be much better than with a traditional scan. Furthermore, the speed is much higher than with a traditional scan, but it would still give you the IR channel. However, I need to build a custom LED array for that purpose, which is beyond my available time at the moment.
My full-frame camera is my best option currently, as I have the best glass for it. Non-full-frame I only own cheap glass or really old cameras.
Furthermore, if everything else of my setup is working, I will probably be willing to spend the money for a canon 100 mm macro lens (repro ratio 1.4:1). But for tests, the extension tubes have to be sufficient.
Still there are many open questions: How to get a really good diffusion of the light, how strong does the IR need to be to get through all the IR filters, which IR wavelength is best to not have too much blur but still good separation between image and scratches, etc., etc.
If somebody has some insights, I would be happy to hear.
While I expected the macro lens to be better, I don’t think the video is very helpful regarding the film scanning requirements:
I would choose a longer focal length for scanning which I would expect to have less of an issue with softness at the edges as the field is much more planar when observed at a distance.
In film scanning we typically have time and/or light, so stopping down to f/8 would probably mitigate much of the issue.
As these are assumptions, please correct me if I am wrong, but I think a comparison of 100 mm lenses at f/8 with the newspaper shot would better fit our purpose.
You are right that a different focal length and F stop would be a better approach for film scanning. I see your point.
The reason why I recommended the video is because it illustrates the differences between the 2 tools. A set of extension tubes is about $30 here (and this is about the minimum). Compared to few hundred dollars at least for a macro lens. But even $30 would be too much if the item does not achieve what is needed for. I maybe wrong but currently I am not convinced that it would be a money well spent (the extension tubes one).
Below are a few items that come to consideration. They may not be applicable for everybody but this is where my thoughts come to.
I tried a non macro lens (without extensions) - Tamron 70-180. For this lens - the biggest magnification happens at 70 mm (so I am not benefiting of the bigger zoom). But the other interesting item is that the slide can cover around 60-70 % of the sensor (just an estimate).
If I am to scan a slide - I want it to fill the frame as much as possible (otherwise there is a lot of wasted space).
If I am able to achieve that - then the effect of the focus loss would be much more pronounced (comparing to having the slide occupying a small part of the center).
stopping down the lens to F8 would benefit for sure (but I would still be starting with something less than acceptable). IMO - if I start with something “good” and then move to F8 to improve - that would be a better approach.
for this point - it maybe just me - to my estimate I have somewhere between 400 and 1000 slides to scan. I wouldn’t want to repeat the process - few scans for a test are okay but not repeating the process. So it has to be as robust as it can be so at least I don’t have to re scan. If the sensor is not well used then I am going to have a lot of extra space that replicates when it comes to storage and archiving.
Let me know if I a missing something or my conclusions can be better.
I wouldn’t use a zoom lens for film scanning (then again, I have a macro lens, so that may make me biased…). They are harder to correct to the same level over the whole range, and the extremes tend to be a bit less well corrected than the centre of the range (something to verify perhaps in your setup).
Depending on where you live, lens rental could be a viable option if you only need such a lens occasionally. Try to hire a focusing rail with it, it will make your life a lot easier.
I use a 105mm Macro lens. A colleague at work printed a Holder for me to keep the negatives perfectly flat. The light that i use is a cheap one from Aldi. It turns out to work very well. Mind you, i do only occasionally scan a film. My main work is done digital. For the 6x6 negatives i take two images and stitch them together with hugin. As the light is not perfectly flat i have to move the light with the holder on the table, otherwise the images are not sharp and i have to constantly adjust the focus. I tried to scan the images at 1:1 magnification. But it meant to take 6 images for one negative. Which was to much faff for now. maybe i try it later. I did some initial editing in Darktable, stitched them with hugin together and made some final adjustments in Darktable. Now i am finally happy with my workflow. It took some time, getting used to edit Negatives. Here are two images from my first roll:
