Digitizing medium format film with m4/3

[acnomad]

Problem: Need to digitize old Hasselblad 2-1/4 x 2-1/4 negatives.
Solution 1: Send them out to be scanned - Expensive.
Solution 1a: Buy a film scanner. Expensive and possibly lower quality results.
Solution 2: Figure out a way to shoot them with a m4/3 macro setup. (Lots of 35mm slide copying products out there, but nothing for medium format).

Here's the beginning of my attempt at solution 2:

Subscribe to see EXIF info for this image (if available)

Subscribe to see EXIF info for this image (if available)

So far, I have fabricated a masonite mask with a 6cm square opening, a PVC closet flange, the right length of 3-inch PVC for the 55mm Micro-Nikkor I will be using, and a 3- to 2-inch reducer. To follow: black paint for the inside surfaces, a negative carrier, and a 52mm to 77mm step up ring that will be epoxied to the 2-inch end of the reducer.

 

[MAubrey]

Might I recommend a film holder from an Epson Scanner? MF holders can be had for the V600 and up.

 

[ean10775]

How does the half pint of beer fit in?

 

[siftu]

How does the half pint of beer fit in?

Looks like Guinness on St Patrick's

 

[JonVdG]

Well, if history flows forward on rivers of beer, so must art. Best of luck to you, and please let us know how it works out.

 

[acnomad]

Might I recommend a film holder from an Epson Scanner? MF holders can be had for the V600 and up.

I've been looking into those, thanks to your advice. It's a great idea, but I have one concern: I think those film holders are meant to fit into the clamshell lid of the scanner and flap down onto the glass. My contraption was going to have a channel mechanism that allows the negative strip to be slid across the opening one exposure at a time. Do the Epson holders have any protruding parts that might interfere?

 

[acnomad]

How does the half pint of beer fit in?

Looks like Guinness on St Patrick's

Well, if history flows forward on rivers of beer, so must art. Best of luck to you, and please let us know how it works out.

The pint of Guinness is key to my inventive creativity, but was not really intended to be in the illustration. Happy St. Patrick's Day!

 

[usayit]

I would vote for a used epson V700. I'm happy with the results and you should be able to find a used one for under $300USD.

This was my review from back in 2006:

Epson V700 Experiences?

 

[pellicle]

Problem: Need to digitize old Hasselblad 2-1/4 x 2-1/4 negatives.
Solution 1a: Buy a film scanner. Expensive and possibly lower quality results.

actually not expensive (certainly within ballpark of a good lens without curvature of field)

also, possibly resolution. You'll get a good 2400dpi from almost all the Epson scanners which is 5400x5400

Next you'll be able to batch scan and if its one of the later scanners it has ICE so you'll get dust removal that's quite sufficient for "generally clean" negs

You'll have a stable work environment and when you've done (assuming you're not pursuing 120roll ongoingly) will be able to sell the unit later.

I've done a lot of scanning over the years and cameras are just about the most cumbersome way to go. Every now and then someone "rediscovers" CopyStand work and eventually progresses to "why scanners were invented".

Food for thought:
in my view ...: black and white neg scanning

You can also "batch" scan a few at a time. Sure scan time will be longer but I'm confident that you'll soon find that you spend less time retouching and monkeying about with the quicker grab from the camera in copy stand mode.

 

[tkbslc]

I would vote for a used epson V700. I'm happy with the results and you should be able to find a used one for under $300USD.

This was my review from back in 2006:

Epson V700 Experiences?

You can get a refurb V550 in the US for $120 direct from Epson. 6400 DPI and a film holders for 35mm and Medium format.

 

[Turbofrog]

I have looked at the results from scanners and the results from modern digital cameras and to my eyes there's no contest in terms of quality.

To achieve similar quality to what a modern digital camera and a macro lens (of which I already have two, neither of which cost more than $100) can get, you're really looking at something like a $2000 Plustek 120 or Reflecta MF-5000. To get results that are decent you can get a $650 Epson V800. If I was going to get something like a V550 I'd be doing it purely for convenience, at which point I'll pay the lab the $5 to do the scanning for me at the time of development which is much more convenient, and I can do 25 rolls for the same price as the scanner, which will take me a couple years to shoot.

Digitizing with a copy stand-style rig just gives you a lot more flexibility. Even fairly high-end scanners are limited in terms of Dmax / dynamic range. With an M4/3 camera a single RAW file will do better than a scanner in this respect, but if you've got a really good filmstock and a B&W negative with tons of dynamic range you can just do bracketed exposures for HDR. And you can also easily do panoramic stitching, giving yourself a ~50MP scan of a single 120 negative with a lot more real information than any flatbed.

 

[Turbofrog]

But don't take my word for it. Real world comparisons contained in the link below:

Why You Should Digitize Your Film Using a Camera Instead of a Scanner

I need to get off my butt and 3D-print my negative holder design this weekend and take this a bit more seriously, since I've been procrastinating on my own rig...

 

[barry13]

Some use Lego to make negative or slide holders.

Barry

 

[acnomad]

But don't take my word for it. Real world comparisons contained in the link below:

Why You Should Digitize Your Film Using a Camera Instead of a Scanner

I need to get off my butt and 3D-print my negative holder design this weekend and take this a bit more seriously, since I've been procrastinating on my own rig...

Yes, that article was the primary force in motivating me to try this stunt.

 

[pellicle]

Hi

But don't take my word for it. Real world comparisons contained in the link below:

Why You Should Digitize Your Film Using a Camera Instead of a Scanner

...

I will be interested in what you see with your own rig.

When I look at his "chalk and cheese" comparison I know from my own experience that really, something is not right. Either he's not able to scan or he's misrepresenting things. I know that you and I can not discuss this, as I should discuss this with the experimenter, but a few things come to mind.

• I wonder about his technique at scanning and if he bothered to adjust focus or if he even set his Epson correctly
• some of the commenters also seemed to take issue with his technique
• I have not used a 5DII but his results are questionable

Of course I should question his results properly by trying his experiment and properly questioning his approach. I can't do that because I don't have that gear.

However in the past on other forums EVERY time a guy has said this (camera is better than scanner) I have offered to do a compare. I have even posted negatives to people and not once (not even once) has anyone ever taken me up on it. Mostly they just fade away from that forum and never post again.

On my blog I have tested scanners many times. I never bothered to test cameras as every test I've ever done was not adequate for anything more than a put down. So I read the above with a little skepticism.

I mean I "want to believe" ... just I don't usually see an experiment which has a thorough and rigorous approach.

This is more like what I get with even 35mm on my Nikon LS4000 ... which isn't horribly different to my Epson 4870 (an earlier model to the V700). I have also the 4990 which is between the two and somehow that isn't the full 43 beans. So perhaps his scanner is bung?

This is typical of what I see with merely 35mm film and a scanner

Subscribe to see EXIF info for this image (if available)

and detail:

Subscribe to see EXIF info for this image (if available)

and

Subscribe to see EXIF info for this image (if available)

this is startlingly better than his scanner samples

in my view ...: my Full Frame camera is 35mm film

again I will offer to anyone who wants to compare their camera rig with my Epson flatbed scanner. I'll post them a negative 120 roll or 4x5 sheet if they promise to post it back even if they're too slack to get around to (or embarrased to get around to) publishing the photo results.

 

[acnomad]

Hi



I will be interested in what you see with your own rig.

When I look at his "chalk and cheese" comparison I know from my own experience that really, something is not right. Either he's not able to scan or he's misrepresenting things. I know that you and I can not discuss this, as I should discuss this with the experimenter, but a few things come to mind.

• I wonder about his technique at scanning and if he bothered to adjust focus or if he even set his Epson correctly
• some of the commenters also seemed to take issue with his technique
• I have not used a 5DII but his results are questionable

Of course I should question his results properly by trying his experiment and properly questioning his approach. I can't do that because I don't have that gear.

However in the past on other forums EVERY time a guy has said this (camera is better than scanner) I have offered to do a compare. I have even posted negatives to people and not once (not even once) has anyone ever taken me up on it. Mostly they just fade away from that forum and never post again.

On my blog I have tested scanners many times. I never bothered to test cameras as every test I've ever done was not adequate for anything more than a put down. So I read the above with a little skepticism.

I mean I "want to believe" ... just I don't usually see an experiment which has a thorough and rigorous approach.

This is more like what I get with even 35mm on my Nikon LS4000 ... which isn't horribly different to my Epson 4870 (an earlier model to the V700). I have also the 4990 which is between the two and somehow that isn't the full 43 beans. So perhaps his scanner is bung?

This is typical of what I see with merely 35mm film and a scanner

Subscribe to see EXIF info for this image (if available)

and detail:

Subscribe to see EXIF info for this image (if available)

and

Subscribe to see EXIF info for this image (if available)

this is startlingly better than his scanner samples

in my view ...: my Full Frame camera is 35mm film

again I will offer to anyone who wants to compare their camera rig with my Epson flatbed scanner. I'll post them a negative 120 roll or 4x5 sheet if they promise to post it back even if they're too slack to get around to (or embarrased to get around to) publishing the photo results.

Not sure exactly when I'll have my setup ready to use, but I will certainly follow up with my results as soon as I have any. Admittedly, I have my doubts about the camera scanning method being superior and know that could be doing this the hard way all for naught, but I suppose that's just my style - I have to see if it can be done...

 

[Turbofrog]

Those results do look quite good. What resolution is the full-size image?

I am also curious about how my results will turn out. My main motivation for digitizing is that I already have everything I need aside from a $30 lightpad that I bought, and some miscellaneous bits and bobs that I can 3D print or buy cheaply at the hardware store. If I shot a ton of film or had boxes of old negatives to digitize, I could totally see the appeal of a scanner. But even the $230 that a cheap V550 would cost me up here in Canuckistan. And with a digital camera, even on 35mm film I can stitch 4 shots together with my 1:1 macro lens...my main goal isn't to get high throughput so much as to get very high quality for the few negatives that I find that I really like. When I do get around to it, I'll have some comparisons to make between the professionally-scanned TIFFs I have from the photo lab. They aren't the "high res" option though, which is 22MP, since I couldn't stomach spending $20/roll.

 

[pellicle]

Hi

Not sure exactly when I'll have my setup ready to use...but I suppose that's just my style - I have to see if it can be done...

its all about the journey and enjoying it ... don't stress and have fun

 

[pellicle]

... well after a few months @acnomad and I have more or less completed our experiment. He's cam-scanned the same negatives I've scanned.

I chose negative for this as colour negative is probably the hardest test of things, with black and white being probably the most amenable to the cam-scan (due to density levels) and colour slides being that bit easier than colour neg (due to densities and an even range of density to colour channel).

So without further adieu here is my blog post where I examine the results of our work.
in my view ...: digital camera as film scanner

Basically the summary is that for 35mm (which wasn't the main subject, it was 120 "square" 6x6cm) the m43 camera did a fine job and competed effectively with a Nikon LS-4000 which is well known as giving a "full 43 beans" of 4000dpi.
Nikon

Subscribe to see EXIF info for this image (if available)

and EM-1

Subscribe to see EXIF info for this image (if available)

its worth noting that you must use RAW to get this close and need a modern RAW processing tool (such as later versions of ACR that better handle highlight rescue).

I won't bore you with the Epson flatbed on this level because its simply inferior for 35mm

However for 6x6 and larger (I use 120 in 6x9 and also use 4x5 sheet) the EM-1 is out of its depth ...

Subscribe to see EXIF info for this image (if available)

even against an Epson flatbed - so a Nikon LS-8000 or later will eat it for brekky.

This will become more pronounced as your format goes up.

So, if you're intending to use your cam to scan 35mm black and white, slide and to a lesser extent colour negative, then it will be effective (although may be more fiddly and won't help you with dust). However if you're interested in larger formats of film (and actually want the benefits of increased resolution of larger formats) then a flatbed will be better, more consistent and easier to work with.

 

[siftu]

... well after a few months @acnomad and I have more or less completed our experiment. He's cam-scanned the same negatives I've scanned.

I chose negative for this as colour negative is probably the hardest test of things, with black and white being probably the most amenable to the cam-scan (due to density levels) and colour slides being that bit easier than colour neg (due to densities and an even range of density to colour channel).

So without further adieu here is my blog post where I examine the results of our work.
in my view ...: digital camera as film scanner

Basically the summary is that for 35mm (which wasn't the main subject, it was 120 "square" 6x6cm) the m43 camera did a fine job and competed effectively with a Nikon LS-4000 which is well known as giving a "full 43 beans" of 4000dpi.
Nikon

Subscribe to see EXIF info for this image (if available)

and EM-1

Subscribe to see EXIF info for this image (if available)

its worth noting that you must use RAW to get this close and need a modern RAW processing tool (such as later versions of ACR that better handle highlight rescue).

I won't bore you with the Epson flatbed on this level because its simply inferior for 35mm

However for 6x6 and larger (I use 120 in 6x9 and also use 4x5 sheet) the EM-1 is out of its depth.

Subscribe to see EXIF info for this image (if available)

even against an Epson flatbed - so a Nikon LS-8000 or later will eat it for brekky.

I read your blog and pretty much agree with what you said. Get a scanner. Even with the em5ii in high res mode on 120 film scans I think it's just too much work per scan but you will get closer to the resolution. You get a much more usable image straight out of the scanners and can use that extra time to read stuff on the forum