Preface: This article seems to have caused a few raised eyebrows, and in some cases heated responses that amount to the idea that the author of this article is “wrong”. The purpose of this article – or at least why I chose to approve it for publication – was to encourage the discussion. There is no one answer to the problem of digitising film. This article simply aims to highlight some of the perceived issues from both a technical (theory-based) perspective, and from a practical (subjective) perspective. If you wish to discuss the pros and cons from your own perspective in the comments, then please feel free, but please refrain from just stating the author to be “wrong” as this adds nothing to the conversation. Please also note, we are only talking about photography here, if this makes you angry, then it might be worth taking a step back. Finally, if you wish to read an article that makes the opposing argument – ie. in favour of camera scanning – then please read this article. Thanks, Hamish… now, over to Marco:
Featured image movie still from Modern Times, Charlie Chaplin (Platinum-toned remix of United Artists, Public domain,<https://creativecommons.org/licenses/by/2.0>, via Wikimedia Commons.)
While there is much interest in scanning with a digital camera, there are other ways of scanning negatives: using a lab or home-scanning with a dedicated scanner. Scanning with a digital camera sounds appealing: “I’ll save money and get better results.” But… There’s always a but. There is a downside – cost in space, $ € ¥ £, time. Lots of time. It’s a bit like the home-brew PC culture of yore. Before it was ubiquitous. Bespoke computers, just like suits. But DIY.
The setup will take up space, more space than the standard standalone scanner. Much more than a lab scan. And, unless it’s hidden away, it may not be that attractive. And even if it is hidden, it may be time-intensive to setup.
Cost $ € ¥ £
First the equipment: mirrorless digital camera, macro lens (not extension rings), film holder, light source, copy stand/tripod… While there are many options to consider, some brands offer helpful accessories: pixl-latr. Negative Supply. Valoi. There are others with bundled packages: copy stand, light source, film holders. But they don’t include the camera or lens, both of which are expensive and problematic.
My take: the camera will need at least 45 mega-pixels for a 35mm standard frame and even more for medium format. If you’re only scanning b&w, then use a monochrome camera. When the monochrome has the same resolution as a colour digital camera, the monochrome will be far more accurate with a higher resolution (line pairs per mm). The object is to try to make the sensor as high-res as possible to eliminate artefacts. Some might disagree, but this is what the maths says to me for me to get the sort of quality of images I want. There are many choices, but of course higher resolution cameras tend to be newer and more expensive.
Some maintain that a camera for scanning can do double duty – scan negs and make images in field. Seems like a plus, but then there’s the ordeal/cost of setting up the digitising rig each time you want to use it to ensure that everything is in “focus” and rock-steady. Leaving the camera in place also has a downside – spending a fair amount of money for a camera that just sits there doing nothing for most of the time. Even using the lens in the field would be problematic – subjecting it to damage etc. Now we’re stuck evaluating the trade-off between money and time. Complexities abound.
As for the components (film holder, light source and copy stand), they are usually assembled separately, even from one manufacturer. You’ll need to choose a good copy stand – one that is stable and has a sufficiently long throw to capture the entire frame of the negative with the camera and lens setup. All of this to ensure good results. The issue is, the good ones are expensive and have a bigger footprint.
Even then, lens choice is problematic. You don’t just need a good macro lens vetted by the experts, you need a good copy of that macro lens. This can make even finding a lens an issue, even after you’ve done all that research to find that one with a good-enough flat field etc.
Then there’s the software, and the continuing cost – the upgrade cycle of digital. And the subscription model for software: low upfront cost at the expense of monthly payments.
The scanning software I use was a one-off purchase with lifetime support, augmented with open source (free) software packages to process the RAW files. No ongoing subscriptions to worry about.
While cameras produce high-quality RAW format, there is a downside. Not all RAW is created equal. Each camera manufacturer has its own proprietary version. As a result, interpreting (reading) RAW requires specialised tools. Tools can either be imbedded in the image processing software or as add-ons (aka plugins).
As for converting RAW into a standardised format, Adobe Camera RAW Photoshop, Photoshop Elements, After Effects, and Bridge all support RAW. Plugins work in some but not all programs. The Negative Lab Pro plug-in only plays well with Lightroom and is not free.
Adobe DNG Converter can convert some RAW files into DNG, just not RAW files produced by all camera models.
Then there’s the legacy issue. Read any VHS/Betamax tapes, floppies…lately? Will that particular variant of RAW be readable in twenty years, or even five? To be safe, convert RAW files to DNG (Digital Negative Format) for safe-keeping. Twice the disk space required. Just saying.
Some cameras can produce DNG files. For those cameras, consider making images in DNG, to maximise the amount of information:
- Use image editing programs that can read the DNG directly or that can invoke plug-ins/add-ons to convert DNG to TIFF
- Use external programs to convert DNG into TIFF or other formats specific to the image editing program. Gimp can process files produced by Darktable in TIFF or XCF [Gimp] format, preserving the maximum amount of information in the source.
Choose lossless file formats to preserve as much information as possible, rather than JPG, for example.
Time is another “hidden” cost. There’ll be time spent:
- Choosing the set up and assembling the physical components.
- Setting-up. Putting the physical (copy-stand etc) in place, unless it’s always in place and one camera is always there, adding yet another expense. In contrast, flatbed scanners are small and unobtrusive.
- Scanning the negatives. There is no downtime. Scanning is labour-intensive. Each frame must be positioned and advanced by hand. And there are no breaks. It’s tedious, unlike dedicated scanners. And no, you cannot “Set it and forget it”. Time spent scanning a set of negs with a scanner is more like attending to the developing process during semi-stand developing.
- Post processing. It will require much more time than expected to fix deficiencies in the way digital cameras render a scene.
- Extra time inverting unless you opt for spending more $ € ¥ to have an image processing program like Lightroom or a plugin for Lightroom like Negative Lab Pro do the conversion. Or use an alternative like Rawtherapee or Darkroom, which both have problems. And even then, they may not be able to deal with the artefacts that are the result of scanning with a digital camera.
- Extra time eliminating dust (His Dark Materials). Film attracts dust like a magnet, requiring more time during post-processing to stamp out and can remove most of the dust from colour negs through the IR channel. And there’s the benefit that a scanner can invert while scanning. As a result, no waiting and scanning only the negatives deemed “keepers”.
- Extra time fixing small artefacts. Camera resolution is over-stated. Each sensor element (sensel) responds to only one RGB channel while the output pixel contains all three. How? Interpolation using proprietary algorithms. Different sensors may have different “views” of the scene, just like the human eye. Perhaps scanning b&w negs with a monochrome digital camera might work.
- Time wasted, scanning negs that aren’t worth it. Although loupe and light table can show you if the neg is sharp, it cannot tell you it is worth doing. Contact sheets were useful in past. The tradeoff is akin to the proverbial “pray and spray”. Not every neg on a roll is worth scanning, unless the images were made thoughtfully, with care. It’s a tradeoff between enough choice and the “tyranny of choice”.
- Scanning larger negatives. This will come at either a higher cost or lower resolution. Try scanning an XPan image 65mm x 24mm. Or a Chroma Cube image 72mm x 24mm. Or a medium format neh up to 6 x 12. A 6 x 6 colour neg will need 100mb at 1800dpi. A 4 x 5 …. And, no, stitching won’t work. It would require too much precision. Besides, when moved, 35mm film may bend slightly differently
There are always tradeoffs. Remember “the perfect is the enemy of the good”
- Consider choosing software and deciding whether to buy or opt for the subscription model – low upfront costs but more over time.
- Consider bulk loading, for example. It may save money but at the expense of possibly scratched negs, fewer film stocks to play with and convenience. It’s worth doing a thought experiment. Unintended consequences.
- Take the tradeoff between home-scanning and lab-scanning. Do I want to spend the time to go to the lab/pick it up and wait? And the expense of coming and going – time and money. Just some of many factors to consider. It’s a judgement call, after all.
For scanning, I opted for purchasing a scanner several years ago, buying scanning software and using open source programs for manipulating (even inverting) the images.
Your mileage may vary.
And no I do not use a digital camera for scanning. I’ve never had one. Only an iPhone. So it’s an opinion. Sort of… But having worked so long in analogue photography, computer science and computer graphics, I know the underlying principles. Instead, I use a scanner with Silverfast AI Studio to convert negatives to positives and to output in RAW for open source programs to convert and process.
I consider that with film it is worth taking the time to compose carefully and thoughtfully, to see and not merely look. As for cameras, I just prefer analogue to digital, older cameras too… A camera is merely one element in an arsenal of tools. For me, using a camera is a way to slow down, to contemplate, to be in the moment and to focus on quality not quantity. Film is expensive. So is time. That’s my approach – only one of many.
Contribute to 35mmc for an Ad-free Experience
There are two ways to experience 35mmc without the adverts:
Paid Subscription - £2.99 per month and you'll never see an advert again! (Free 3-day trial).
Content contributor - become a part of the world’s biggest film and alternative photography community blog. All our Contributors have an ad-free experience for life.
Sign up here.