Macro & Camera for Slides

[sengsational]

Digitizing slides with any type of "scanner" is way too slow, so I started looking into using a DSLR and macro lens. I haven't had a camera with a separate lens in 30 years (once kids came along, it was point and shoot), but my sister has an old camera and lens that I thought I might try, but I'd hate to have her ship it to me, and then have it not work. So I thought I'd tap the knowledge base here. I'm pretty sure this isn't going to work like I want, but...

And yeah, this is a very specific, nerdy question, so I'm not sure anyone here has the answer, but it's worth a shot, I guess.

The camera is this one: https://www.digicamdb.com/specs/canon_eos-350d/
The lens is this one: https://www.sigmaphoto.com/70-300mm-f4-5-6-dg-macro

The idea is that if you have the right macro lens and camera, you don't need a special purpose slide attachment (like the Nikon ES-1, which has a special lens inside, and that allows you to use lenses in the 40mm to 60mm range).

So, without any kind of additional lens, you might be able to use a DSLR and a macro lens. I filled out a calculator I found: https://www.scantips.com/copycalc.html

I'm pretty sure that I need to put in 1:4 for the lens ratio into the calculator for this Sigma lens. This would give me a result of 1404 x 934, and I'd probably have to crop out big black edges. That's pretty low resolution, for all of the work it would require to handle every slide, even if the digitizing process is "instant", by clicking the shutter. When I scanned printed photos, I settled on something around 3521 x 2376, so about 100 pixels per film mm, and the 1:4 puts me down at 39. At least that's what I've come up with when using the calculator. But I'm not real familiar with the terminology, so maybe I've got something out of whack.

So if anyone has any experience in this area and would like to add two-cents, I'd appreciate it.

 

[njhowie]

How many slides are we talking about?

Instead of investing a lot of your time, dealing with lower resolutions, etc., why not consider letting someone else do it for you?

Costco will do it and the price seems reasonable:
https://www.costcodvd.com/services-and-pricing/slide-transfer

 

[OldShooter]

In Olden Times, every major camera line included a slide copying setup. For example: https://www.google.com/search?newwi...KHQBWCBcQjJkEegQIBhAC&biw=1600&bih=1019&dpr=1

I suggest that you sleuth out a Canon setup including the appropriate film-era lens and an adapter for your EOS. That should give you a well-engineered solution. Try eBay with "canon bellows slide duplicator" for a start.

Re the 70-300 I would never use that kind of lens. A prime (single focal length) will give you better resolution, easier handling, and probably better contrast.

 

[sengsational]

Thanks for the ideas, guys.

It's not worth paying $0.33 per slide....if I was going to pay, then I'd need to cull, and I just want to scan 'em all. Also, nowhere on the site did it say what resolution you get back. I saw a guy that did 1000 in a long afternoon. I'm not even sure how many slides are in that box I just acquired... I haven't dared look, but it's at least 1000.

Yes, I've done some level of googling already. I decided I wasn't going to invest any significant money in equipment for the one-time use. I suppose I could get a 50mm 1:1 macro lens that fits my sister's camera, then sell it when I'm done, but that's not an "engineered solution", and I might have trouble with it working right. As for the bellows, no need for me to have anything flexible...I'd figure out how long I needed the tube to be, make one that I can throw out when I'm done.

A few corrections to the OP: the ES-1 doesn't have a lens in it after all. So it's basically just a tube and a slide holder. Also, then, I thought a regular 50mm (or so) would work, but it requires a macro lens, and one that's got a "good ratio" too.

On the plus side, if I went with something like the low DPI that doesn't fill the frame, it's trivial to batch crop by writing a Java program. I've already got JPG handling Java code that messes with metadata, and it would be easy to add something that crops out the big black borders.

 

[Souschef]

I know it might be a pain, but is it possible to view the slides and only select the ones you want copied.
When we cleaned out my late MIL's place, she had boxes of slides. My wife looked through them and there was nobody she recognized, so we tossed them.
Also, even at $.33 a slide, that is a little over $300. It might be worth it so you do not have to mess around.
My late father had all kinds of 16mm movies, dating back to the 1938 Worlds Fair. He was too cheap to have them duplicated (not implying you are), and tried to do it with a video camera.
Bottom line, the sprocket holes disintegrated, and they were lost forever.

 

[stepford]

On thing to watch out for is the level of distortion the lens in question produces at your desired magnification. I seem to recall that some lenses were denoted "copy lenses" for their extremely low distortion at 1:1 (or near 1:1) magnifications. The zoom lens you are considering may have low enough distortion that it will serve your purposes, but it's probably worth your while to check.

 

[sengsational]

....

It is a combination of being too cheap and knowing the project wouldn't be THAT bad, once I got set-up. If the alternative was using a scanning solution (where instead of days to complete, it would be months to complete), I'd pay for it. But if I can knock it out in a couple of days with stuff I already have or can cobble together, that's just the kind of thing that would give me some satisfaction.

 

[sengsational]

On thing to watch out for is the level of distortion the lens in question produces at your desired magnification. I seem to recall that some lenses were denoted "copy lenses" for their extremely low distortion at 1:1 (or near 1:1) magnifications. The zoom lens you are considering may have low enough distortion that it will serve your purposes, but it's probably worth your while to check.

As I indicated, I'm a novice when it comes to camera lenses, so that will be something I'd need to figure out if I were going to buy something. If I'm just going to use what I (and my sister) have, then I suppose I'll just eyeball it, and if it looks ok, just go with it.

As I indicated in the OP, when I tried to use my point and shoot camera in macro mode, it was darker on the edges. I called it "toy camera effect", but it looked like it focused well in the center as well as the edges, but was noticeably darker in the distant corners. So I guess they call that vignette effect. To me, that was unacceptable, so I discarded that as an option.

 

[Sunset]

I'm interested in this as I use to take lots of slides, and now have a couple of boxes of the round carousels full.

So I'm interested in OP coming up with another terrific solution

 

[Sunset]

......
As I indicated in the OP, when I tried to use my point and shoot camera in macro mode, it was darker on the edges....

Any chance that is simply a function of not having a bright enough diffused light source ?

 

[sengsational]

Any chance that is simply a function of not having a bright enough diffused light source ?

I was using a tablet computer with the brightness all the way up. When I looked at it with my eye, it was uniformly bright in the corners, but the photos I took had dim edges. And I went back and looked more closely at the test photo I took and it also more crisp at the center.

I think an easy test would be to take an empty slide frame, set it on a phone screen that's displaying a pure white image, move the camera (in macro mode) close enough to fill the camera frame (and still stay in focus), then take a picture of the screen pixels. It will look like you're looking through a window screen. But from there, you can easily tell the difference in focus and exposure at the center versus the corners. Reviewing the test shot I took with the slide laying directly on the tablet computer screen, it's dim and indeed out of focus on the corners.

 

[sengsational]

I think an easy test would be to take an empty slide frame, set it on a phone screen that's displaying a pure white image, move the camera (in macro mode) close enough to fill the camera frame (and still stay in focus), then take a picture of the screen pixels. It will look like you're looking through a window screen. But from there, you can easily tell the difference in focus and exposure at the center versus the corners.

After writing that, I thought I should actually do my proposed "easy test" and share it with the group.

The numbers below each sample is pixels per mm of film. My target is 100, so this camera doesn't cut it. At 63 px per mm, you can see (in bottom row) the beginnings of the exposure difference between the edges and the center (a circle). More obvious at 78, and at 145 (filling the camera frame) it's horrific.

 

[OldShooter]

Most lenses have a little fall off in the corners when at full aperture or even one stop down. Wide angle lenses especially. God only knows what a 70-300 would do. Not the lens for the job, to be sure.

You don’t want to spend a lot of money, OK. Lurk eBay for a bargain bellows copy setup and appropriate lens, then resell when you're done.

 

[Ian S]

Why not something like this: https://www.ebay.com/itm/233173708894?epid=28032167129&hash=item364a3c705e:g:TJwAAOSw2xRYj5Sf
I have an older version that was ok but this sounds considerably improved. I've also used the Wolverine 8mm film to digital device to convert 60-70 y.o. 8 mm home movies my dad had and it worked very well.

 

[sengsational]

Why not something like this:

There's a serious mis alignment of megapixels vs image quality. They've got lots of work to do in order to get "archival quality".

You don’t want to spend a lot of money, OK. Lurk eBay for a bargain bellows copy setup and appropriate lens, then resell when you're done.

We're thinking along the same lines... already tried that, but wasn't confident enough to buy anything.

 

[OldShooter]

... wasn't confident enough to buy anything.

You get something that doesn’t work or you don't like, resell it and view any round-trip cost as tuition. You're not getting married here.

 

[target2019]

Something like this may give you acceptable results. Easy enough to find on Amazon and read the reviews while looking at examples from users.

Rybozen Mobile Film and Slide Scanner, FILM to JPEG，Converts 35mm Slides & Negatives into Digital Photos with Your Smartphone Camera, LED Lighted Illuminated Viewing, Foldable

 

[sengsational]

^ should have mentioned it in the OP ... I tried my phone camera first, and it couldn't focus close enough, by a mile. I did see a couple of YouTubers use their modern apple phones with apparent success. But, yeah, I looked at what was on Amazon and found mostly slow, high quality devices, but slow is off the table. Although I didn't check the specs on every device (missed the Wolverine), but checked specs on a dozen and threw in the towel.

 

[cathy63]

Digitizing slides with any type of "scanner" is way too slow, so I started looking into using a DSLR and macro lens. I haven't had a camera with a separate lens in 30 years (once kids came along, it was point and shoot), but my sister has an old camera and lens that I thought I might try, but I'd hate to have her ship it to me, and then have it not work. So I thought I'd tap the knowledge base here. I'm pretty sure this isn't going to work like I want, but...

I am confused about how the DSLR and macro lens setup could be faster than a scanner?

I'm actually doing a slide scanning project now and I guess I'm about half-way done with about 3K slides. I also have some negatives, but I don't yet know if those are dups of the slides.

DH has Nikon cameras and lots of lenses, so I started Googling first to see if there was a way to use his existing equipment and found various pages describing the same thing you're looking to do. I quickly realized that anything we could rig up would mean loading one slide, snapping a photo, unload the slide, and repeat ad infinitum. Then he mentioned that he also had a slide scanner. I started using that, but it was no better since it only handled 3 slides at a time.

A little more googling found this scanner https://www.amazon.com/gp/product/B00O2BU8PK/ref=ppx_yo_dt_b_asin_title_o07_s00?ie=UTF8&psc=1

You just push a slide through from the right, capture it, then push the next one in on the right and the previous slide falls out on the left. This is super fast and easy. I do one carousel at a time and then copy the scanned images to my computer into a folder that is named according to the date stamped on the slide frames. I usually work on them while watching baseball on TV, and can easly get through 3 carousels in an hour.

 

[sengsational]

... can easly get through 3 carousels in an hour.

Thanks for that data point using the Scanza: better than 12 seconds, on average, to load a slide and scan it.

It appears that the Scanza is similar to several other non-brand-name scanners out there. It looks like the "guts" of many of these come out of the same factory, based on what the settings and controls screens look like. This one, like the rest, has a reasonably sized sensor, but the processor is crude, resulting in less than stellar captures.

The problem I have with these devices is that you get "just OK" level results. This one, on a quick look, produces a result that's better than the (rather flawed) Wolverine, but still not all that great. One review said it was a long way from being able to see the film grain. That indicates the scanning resolution / compression leaves some detail behind. Not that I "need" that level of detail!

My position, and yes, it's not a common or necessarily logical position, is that if I go through the trouble of loading every slide, which, as you say, is going to be a part of the process no matter what, then I don't want to leave anything on the table, so to speak. With the Wolverine, you could see compression artifacts on a 3 x 5 print! That's crap! With the Scanza (and similar) I think a 3 x 5 print would look ok? Can you see compression artifacts if you display your scanning results on a high definition TV?

I expect that almost all of the future viewing of these slides will be done on a 1920 x 1080 (HD) display, or higher (4K tv). There might be some of the better photos that get cropped, so I'd like to add a little buffer to that, if it doesn't cost too much in time or equipment.

It's time-consuming for me to ascertain how the results from these <$200 scanners compare to "the ultimate scan" (how much quality is being left behind). These slides are going straight into the trash once scanned, so no second chance.

 

[sengsational]

You get something that doesn’t work or you don't like, resell it and view any round-trip cost as tuition. You're not getting married here.

True enough. But it's not in my nature to reach a goal by wandering all over the place.

I think this one would be great, but has a lot of features I don't need:

https://www.bhphotovideo.com/c/prod...7A006_100mm_f_2_8_USM_Macro.html?sts=pi&pim=Y

Why are so many of the eBay lenses in Japan, I wonder.

ETA: I think this one would work, as well, and has fewer unused features:
https://www.bhphotovideo.com/c/product/1332311-REG/canon_2220c002_ef_s_35mm_f_2_8_macro.html

 

[cathy63]

Thanks for that data point using the Scanza: better than 12 seconds, on average, to load a slide and scan it.

It appears that the Scanza is similar to several other non-brand-name scanners out there. It looks like the "guts" of many of these come out of the same factory, based on what the settings and controls screens look like. This one, like the rest, has a reasonably sized sensor, but the processor is crude, resulting in less than stellar captures.

The problem I have with these devices is that you get "just OK" level results. This one, on a quick look, produces a result that's better than the (rather flawed) Wolverine, but still not all that great. One review said it was a long way from being able to see the film grain. That indicates the scanning resolution / compression leaves some detail behind. Not that I "need" that level of detail!

My position, and yes, it's not a common or necessarily logical position, is that if I go through the trouble of loading every slide, which, as you say, is going to be a part of the process no matter what, then I don't want to leave anything on the table, so to speak. With the Wolverine, you could see compression artifacts on a 3 x 5 print! That's crap! With the Scanza (and similar) I think a 3 x 5 print would look ok? Can you see compression artifacts if you display your scanning results on a high definition TV?

I expect that almost all of the future viewing of these slides will be done on a 1920 x 1080 (HD) display, or higher (4K tv). There might be some of the better photos that get cropped, so I'd like to add a little buffer to that, if it doesn't cost too much in time or equipment.

It's time-consuming for me to ascertain how the results from these <$200 scanners compare to "the ultimate scan" (how much quality is being left behind). These slides are going straight into the trash once scanned, so no second chance.

Yes, I think all the cheap scanners are pretty much the same. The slides I'm scanning were taken between 1956 and 1976, and the original photographer was not very good. Lots are out of focus or focused on the wrong thing. On some of the older slides, the colors have deteriorated and they now have a blue cast. So given the starting quality, compression artifacts aren't something I'm particularly concerned with. None of mine are going to be shown on a high def screen. I was just hoping to find a few of family members who weren't often photographed.

If you want to play with a few of the scanned images, let me know. I'll upload the files and PM you a link where you can download them.

 

[wyecrabber1]

Copying slides/negatives

Last year a friend asked me if I would convert 200+ slide for a retirement dinner for another friend. I purchased this scanner off of Amazon for $35.00 and it did a great job. You use your smart phone. Worked like a charm.

https://www.amazon.com/KODAK-Mobile-Film-Scanner-Built/dp/B07MTKNTPK

 

[sengsational]

...
The camera is this one: https://www.digicamdb.com/specs/canon_eos-350d/
The lens is this one: https://www.sigmaphoto.com/70-300mm-f4-5-6-dg-macro

I got the camera and lens today in the mail, from my sister and did a quick test. Result: FAIL

The test setup: I set the lens to manual focus, and to the 1:2 setting (that's the extreme), set focus to as close as possible, then moved the camera up and down on a tripod, shooting straight down onto the 'light table' that was my old tablet computer. I had the camera set to the lowest ISO and experimented with f-stops, letting shutter speed float. The things I read said f-8 to f-11 was best.

The good news was that the camera and lens did have the capability that was indicated in the calculator; I got one in-focus sample that was 82 pixels per mm (the calculator said 78).

The bad news is that there's no solid way to focus.

I focused by changing the distance to the subject, but it's super critical! And although my quick test setup didn't have it, I could have rigged something that was horizontal. Indeed, that was my plan, if the quick test worked.

The biggest focus problem was that I couldn't check focus in real time. This is a show-stopper. As I mentioned, I never had a digital camera beyond a point and shoot, so was unfamiliar with this Canon EOS DSLR. You can't use the LCD screen as a viewfinder! When I partially press the button on my Canon ELPH, the LCD shows a zoom-in that is used to make sure focus is right. But this DSLR has a blank screen while shooting. And the viewfinder is kind of blurry. There's also no way to hook up a live monitor, which would be a great way to focus. And the icing on the cake is that the shutter delay is 10 seconds or nothing (I was hoping for a 1 second option).

So to answer my own question, the camera isn't suited to this task, even though the resolution calculations work.

I'll either need to reduce my standards and go with a stand-alone "3 second" slide scanner, or give-up on the project...I'm done with trying DSLR rigs. I was pretty close to putting in a snipe on a 50mm macro lens...I'm sure glad I didn't!

 

[sengsational]

I'll either need to reduce my standards and go with a stand-alone "3 second" slide scanner, or give-up on the project...I'm done with trying DSLR rigs. I was pretty close to putting in a snipe on a 50mm macro lens...I'm sure glad I didn't!

One thing I forgot to mention that I learned today is that I've been displaying my tests using a viewer that doesn't interpolate adjacent pixels. That type of "unassisted" display isn't typical these days since we have quick processors. But the non-interpolating display really brings out the jaggies, whereas when interpolated, it just turns into a little softer focus. So given that everything nowadays is going to interpolate, my target of 100 pixels per mm can be relaxed, I think, because that was set when I was looking at non-interpolated test shots.