Film Diaries: Examining the costs of shooting film vs digital

_8040387 copy
Looking for an answer? I can’t give you a clear cut one, but this article might certainly help to clarify your thoughts.

There are many reasons to shoot digital. There are many reasons to shoot film, too – beyond the simple ‘I want to’. Though I find that for color work, digital is significantly better than film due to the level of control and accuracy it produces, film remains my medium of choice for monochrome work. The reason has to do with nonlinearity of tonal response, especially in the highlights – film never quite seems to clip under almost all circumstances, and this jives with the way our eyes see the world very nicely indeed. But there are typically two things that stop people from trying film: workflow, and perceived cost.

If your intended final output is digital, then it makes sense to adopt as digital a workflow as possible; if it’s optical prints, then film still makes sense. However, for a given number of keepers – you might well find that film is just as fast; I certainly do. The challenge usually lies in the developing and digitisation processes – in many parts of the world, it’s difficult to find any labs at all that handle E6 and black and white films; let alone good ones. And let’s not even talk about ones that can accommodate requests like pushing and pulling. It’s generally possible to obtain scans – albeit not very good ones – together with the development; resolution tends to be very limited. In fact, the digital files are a byproduct of the printing process – most modern mini lab machines make digital prints, not optical ones. A professional drum scan will be both time consuming and very expensive, but will yield much better output resolution.

I suspect one of the big attractions of digital for most people was not so much the immediacy as the convenience; after all, we had Polaroids years earlier, and they never really supplanted negative or slide films. The digital age brought about the ability to handle all stages of the process independently, from capture to processing and output. The rest of this article is going to examine my workflow from capture to keeper image – for both film and digital, taking into account operating costs over the course of a year. Of course, individuals can and likely will differ. I’m going to compare like with like as much as possible, and ignore the fact that larger film formats are required to compete with smaller digital ones for comparable resolution. Lenses therefore do not enter into to equation as we can use the same on both, providing we buy into a diverse system. Similarly, we will be looking at depreciated/ real-world second hand film camera prices as opposed to new, simply because this is the way most people are going to obtain them.


The lifespan of a digital device is realistically limited to 8-10 years with heavy use; I’ve known of cameras that have failed much earlier than that; however, the majority are replaced or mothballed before they become truly unserviceable. Fortunately, now that we have passed the point of sufficiency, the incremental upgrades to each generation of digital are getting fewer and fewer – there are less reasons to upgrade of switch. People will likely keep their digitals longer as a result.

A D600 or D700 body will cost in the US1.5-1.7k region; I’m ignoring the D800E for now because that camera’s resolving power and general image quality really requires us to compare it with medium format film, which would be unfair. However, the D600/700 would deliver image quality on par with or slightly better than the best of the current film emulsions. This body would depreciate by approximately $400-500 over the course of the year; more if it happens to be a year in which a new camera is released. There are no other operating costs apart from that. My keeper rate with digital is around 2-3%; let’s call it 2.5% to make the math easy. It means I have to shoot 40 frames to get one that passes muster without qualification. If I shoot 500 keepers through the cost of the year – which is moderate – the per-keeper cost is therefore 80 cents to a dollar.


Film, on the other hand, tends to give me keeper rates of 50% on a bad day. It can be as high as 80-90%. Even assuming 50%, at a per-roll cost of $5 for black and white, or $15 for color slide – that’s 28 rolls. Both include processing; DIY if it’s B&W, or lab-based if it’s color. If you buy chemical in bulk and do your development in such a way that you process in batches and fully deplete the chemical (i.e. don’t do one roll, and have to throw it out later because it’s oxidized) then it’s even cheaper. You could take that even further still by bulk loading, but not many films are available in long reels anymore. This works out at $140 for black and white film, or $420 for color slide. You’ll notice that’s cheaper. It isn’t because film is better or digital is worse; it’s just that somehow photographing with a single-use medium with a finite number of shots tends to make you think a lot harder before hitting the shutter. And this of course in turn results in better images.

We haven’t taken into account scanning, and if you get this professionally done, can add significantly to the costs – you of course have to scan the whole roll in order to get your keepers out, and that might add another $300-400 to the bottom line. Less if you are prepared to do it yourself with a flatbed, and even less if you are using my scanning rig and a DSLR. Time taken is quite another matter – flatbed scans might take up to 15min per frame.

You’ll notice I haven’t taken into account depreciation on the film body; that’s because it doesn’t really matter what you use. There is little to none in the current photographic climate. You might even find yourself making some money if you decide to resell the camera later; I feel we are at the very bottom of the pricing cycle for film cameras. Some of the more expensive, top of the line gear from recent years is now both very affordable and will still last you far longer than a digital camera. There’s one final cost: buying the DIY developing hardware (tanks, changing bag, storage bottles, thermometer, measuring jug etc.) – that’s actually pretty cheap, and a full set of kit can be had for about $120. You can of course pay more and gain the ability to do multiple rolls simultaneously.

Of course, the precise economics will depend very much on your individual usage patterns; I have of course drastically oversimplified the whole equation. If you shoot significantly more, or have exceptional shot discipline with your digital work, then it may no longer make sense to consider film as an alternative medium – artistic considerations aside, of course. And it’s difficult to ignore the frequently much higher tactile quality and haptics of the flagship film cameras. Personally, I find the whole experience of shooting film to be quite different – and most enjoyable because it provides me with some differentiation from the digitals I use for work. After all – if you’re spending this much time and money on something, it should be fun, right? MT


2014 Making Outstanding Images Workshops: Melbourne, Sydney and London – click here for more information and to book!


Visit the Teaching Store to up your photographic game – including workshop and Photoshop Workflow videos and the customized Email School of Photography; or go mobile with the Photography Compendium for iPad. You can also get your gear from B&H and Amazon. Prices are the same as normal, however a small portion of your purchase value is referred back to me. Thanks!

Don’t forget to like us on Facebook and join the reader Flickr group!


Images and content copyright Ming Thein | 2012 onwards. All rights reserved


  1. Interesting post, thanks. Reminds me of discussions over at LUF about people thinking of ditching film for an 8,000 USD Monochrom. That’s a lot of film and developer/developing costs.

    After a two-year fling with digital, I sold my 5D2 and a lens to fund a Coolscan 9000 and go back to film 100%. I’m so happy I did. The 9000 scans a 36-roll (though I get more frames usually) at 200dpi in an hour (BW is faster though). I keyword and rudimentarily postprocess the TIFFs while the scanner works so the time consumption per roll is not more than the scanning time really. When I shot digital I would usually spend that amount of time, if not more, selecting “keepers” for further post-processing from large loads of shots. I found this extremely tedious. Then again, I photograph for fun only and have the luxury of choosing my medium.

    Incidentally, developing a roll of slide film here in Holland is less than 3 EUR. And there’s Agfa Precisa to be had in Europe for the price of a roll of Portra 160.

    • There’s no way the Monochrom out resolves LF film; it’s about on par with the D800E, which is itself on par or slightly below a good MF (6×6, 6×7, not 645) negative. Large format will be quite a big notch up – and you will have movements, which are completely absent with the Leicas.

      Slide film was pretty decently priced in Holland. I bought and shot a number of rolls of Provia while I was over there. Nearly impossible to get in Kuala Lumpur, and very expensive even if you can find it…

  2. Ørjan Laxaa says:

    Can anyone recommend me a decently priced film-scanner? Looking into getting one, since scanning in a lab is so expensive.

    • The Epson V750s are supposed to be good, but I haven’t tried one. I’m looking into one too since 4×5 is too big to copy with a DSLR, and completely negates any resolution advantage of the larger format.

      • The Epson V600 is in my price-range. I’m not a pro, so I the pictures I take are mostly for web use(Flickr, Facebook etc.)

        • That should definitely have more than enough resolution – oversampling may help; scan at say 4800 and downsize to 2000 afterwards. Unless you’ve got exceptionally well-developed negatives, you’re not going to have much more than 2000dpi of real information anyway.

          • The Plustek OpticFilm 8200i seems decent enough as well, and is within my pricerange. The only downside is that it only scans 35mm negavites. But then, I mostly shoot 35 mm.

  3. The cost “per keeper photo” is the wrong way to calculate the cost. By far the most expensive element in photography is one’s time! Lets look at the total cost “per event”; landscapes, travel time, accommodation, airfares. Portrait, thinking about the type of outcome, organizing a model, makeup, hair, props etc,, If you think about the total time required to get great keepers the cost of equipment is near irreverent!

    Digital is a far quicker OTVCT loop (Observation Take shot, Evaluate, Correct problem and Take another shot) than any film other than Polaroid (do you notice that many professional film photographers use a Polaroid before taking a large format film?).

    As I consider myself still learning, I much prefer digital.

    Regards to all,
    Dave 🙂

    • Hmm…I agree and disagree: yes, time is an important factor, but you’re probably going to spend the same amount of time shooting regardless of the medium. Personally, I find the total film process time – develop, scan, convert – to be similar to or faster than digital due to the volume; and at the end I get more keepers. Since the ultimate priority is final finished images which you’re happy with, if you’re spending the same or less time – what’s not to like? As far as productivity goes, I do actually bill by time, or by quantity – if I can get more quantity out, then that’s definitely better. I certainly consider efficiency in my workflow, and if there is a faster/ cheaper way to get the same quality results, then I use it.

    • Taildraggin says:

      …OODA Loop.

  4. I started shooting film (just 35mm) when I was bequeathed a film camera last year. I started shooting it out of curiosity (as well as a sense of obligation to the previous owner), and I got really into it; I could understand all the people who enjoy it because “it slows you down”, “you don’t get a chance to redo the shot”, etc. I agreed with it all and I really liked the results.

    Then I ran into a problem; not economical, but philosophical. I’m a vegetarian, and I live with the aim of using as few animal derived products as possible, as far as such a lifestyle can be carried on.

    We all know that one of the ingredients used in making film is gelatine; I knew this even when I received and started using the film camera. It’s not the major ingredient, but it’s there. A month or so back I realised that If I look honestly at myself, I’m being a hypocrite by using film while trying in every other aspect of life to avoid using such things. It’s an “I want to”, not an “I need to” thing. (For instance if I’m prescribed medicine and it contains gelatine, I will take it. But that’s about as far as I like to go).

    I’ve tried to rationalise it by saying that the film is being used to create (hopefully) something of artistic worth and thus transforming the meaning of it. If anyone’s familiar with the philosopher Peter Wessel Zapffe, he called this “sublimation”: the transference of suffering into something of value. I’m not totally convinced by my own rationalisation though.

    Just a couple of days ago I received as a birthday present a copy of Nick Brandt’s “A shadow falls” (having discovered his work through this site). I’ve only dared look at the first ten or so pictures so far. I fear that if I look at too many too soon, I might a) not get the full effect of the book and b) give up photography in disgust at my own efforts 🙂

    Knowing that Nick Brandt shoots MF film but is also a passionate believer in the rights of animals to live, I sent him an email asking him about it – if he saw any irony in the situation (of course I was in no way trying to call him a hypocrite: his work is of great importance). He replied within a day with a very reasoned and well put answer: effectively, “I do what I can to minimise animal suffering, but I’m not perfect”. Concerning the film and the gelatine, he mentioned that he would “reluctantly” continue to use it as long as film is still available, as he doesn’t like digital.

    I don’t think this post fits in with the general tenor of the discussion, but if there’s one place to get an interesting set of comments on any photography-related topic, it’s here…so there it is. I’m open to an argument which will get me shooting film again without the concerns. At times I think I’ve convinced myself, but I’m not there yet.

    • Mark, that’s an interesting take. As a meat eater I’m sure I don’t come at things from the same perspective. But one way you could look at it is the gelatin is a byproduct of the meat industry. Cows and pigs are not being raised to make gelatin, but rather for meat. It just makes sense to find uses for all parts of the animal, so gelatin is produced. So you are not contributing to the creation and destruction of animals for your film hobby, rather you are helping to make it so the animal died less in vain. Sorry in advance if that doesn’t help at all.

    • I can see your conflict. Perhaps the most obvious question is: do all films use animal gelatine? Surely some must be using synthetic. If so, problem solved. If not, then you also have to ask yourself if absolutely everything you use/ do excludes animal products; if the answer is no, then you have bigger concerns, I think.

      Failing that, there’s always the route of coating your own LF plates 🙂

      • Larry, Ming, thanks for the comments; you both make a lot of sense. Of course, animals are not slaughtered so satisfy our demand for film, and if everyone stopped using film tomorrow the number of animals killed would likely not change at all. So in that case there is less “guilt” about using it. While I do what I can to avoid using products of animal origin, it’s essentially impossible to do so in the modern age; and on top of this it’s easy to get more and more pedantic about it until you find it difficult to live normally because you can trace everything back to the use or abuse of animals…

        As far as I’m aware, all films use gelatine; apparently there was an attempt to create a film using an alternative method, but it didn’t really take off.

        I was really just thinking out loud in that original post, but I appreciate the comments!

        • If you really feel guilty about film photography because of gelatin, please stop shooting film. A year ago you weren’t shooting film and you probably haven’t shot a lot of film rolls in the last one year. So not much damage done and it is not a big change at this juncture to forget film and shoot digital. It is not a life altering switch after all, you would still be doing photography and in all likelihood you will feel better.

        • I get your point but what you effectively do is to accept to sacrifice some pleasures (eg, shooting film) for a greater good. And you make that more sincere by imposing strict and simple rules and not allowing yourself to relativise. And you do the latter partly because you think that without simple rules, the uncertainty of ascertaining cause and effect could, at least subconsciously, make you interpret the cause and effect relationships in a way that minimises your sacrifice, which in the extreme could mean no sacrifice at all. The other part is to communicate a message by making some sacrifice visible to the outside world.

          As a utilitarian, I’d rather go about that the other way and try to maximise the greater good achieved for a given amount of sacrifice (now, measuring sacrifice is also a very difficult if not impossible thing, the easiest parts are time and money, eg, what is the maximum amount of greater good I can achieve by dedicating one hour of my day to it). Another way of thinking about this is to introduce the concept of compensation. (Something sometimes used in construction projects, need to cut down trees to build a building? Plant new trees somewhere else. This can also be asymmetric, damming a river, reduce the inflow of pollutants into other rivers. In regard to the welfare of animals, thinking of compensation might not be so straightforward but it could be to cut down time spent on something you like doing and divert that time to, eg, do volunteer work at an animal shelter).

          To summarise, we need principles but we also need flexibility, pragmatism and creativity.

          • Another interesting way of looking at it. Thanks for the input.

            The reason I started using the film camera was because it used to belong to a family member. Upon his passing, it was given to me, and at that time my rationale was simple: the guilt I might feel by using film is of less significance than the guilt I would feel if I failed to use this camera, it having been given to me and it having belonged to a person of importance in my life. Nobody pressured me to use it; I chose to.

            It’s only recently that the second notion has kicked in: that using a film camera is out of alignment with the way I generally try and live, as it is something I want to do rather than something I need to do.

            The push and pull of these two “motivations” are the determining factor in whether, at any given point in time, I feel that I should use it or not.

            To someone who doesn’t share my viewpoint, this probably seems trivial, and if I step back a bit and look at it objectively, I could probably see it that way too. Still, it has enough gravity to me to give me something to chew on, so to speak.

  5. Out of curiosity, how does something like Silver EFEX Pro (with their film profiles) compare with film for you?

    I get better results with it than I got out of Delta 400 (not to speak of Neopan 1600 or T-Max 3200). Where I like my own (photochemical) B&W prints it is mostly due to the experience of photographic paper vs a computer monitor. I especially like(d) the artisanal aspect of them, from the real photographic paper to that small irregular frame (created by the gap between the image on the negative and the mask in the enlarger, even the grain, the blown highlights and closed shadows, all make them feel more of an abstraction, an interpretation than a technically perfect image would do.

    (I should add that 95% of my analogue B&W was friends, family, and colleagues mostly in the evening indoors, meaning pushed film, rarely exposed for less than ISO 800, wide-open, handheld at the shake limit.)

    • Filters? Very mediocre and a poor substitute used by the lazy, to be honest. A given film does not have a fixed, distinct ‘look’; it depends on how you develop and scan it.

      • I don’t get what your point is. Apply one develop and scan procedure, and you get one look. Apply one filter, and you get one look. Apply another develop and scan procedure, and you get another look. Apply another filter, and you get another look. Tweak your develop and scan procedure, and you get slightly different look. Tweak your filter, and you get a slightly different look. Nobody said, pick a pre-packaged filter and be done with it. As you didn’t say, just send it to the nearest lab and let them develop and scan it.

        Your points might be that:
        (1) There are looks that cannot be achieved by any digital tweaking (eg, maximum dynamic range, maximum size without digital artefacts creeping in). Absolutely, but that doesn’t mean that tons of B&W film shots can be recreated very accurately with digital. And the higher you go up the ISO range, the delta between film dynamic range and digital sensor dynamic range goes down, particularly with ‘ISO-less’ cameras like the D800, as well as the maximum size you would have viewed your film images.
        (2) It can be hard to extinguish the ‘digital look’ completely, ie, to make a digital image indistinguishable from a film image. That might be true, though it is partly a question of practice, but ultimately what matters is what wins in a blind test (of whoever your audience is, which in the extreme would be yourself). And yes, blind tests are affected by bias (eg, I can distinguish an Aperture B&W conversion from a Silver EFEX conversion, not because one would be ‘objectively’ better but because I know where to look for tell-tale signs, even if only subconsciously) but if you can identify the bias, you can equally well reverse engineer it.

        I see zero reason why any selected develop+scan procedure could not be perfectly recreated with a purely digital workflow except for resolution, dynamic range, and bit depth limits. You know exactly that any image is just mathematical transfer function from reality and with enough resolution, any analogue/continuous/random function can be indistinguishably recreated digitally (using discretised data). And I very much doubt that recreating Neopan 1600 on FF with a D800 would run into those limits.

        Analogue and handmade has its qualities by transporting a notion of effort into the product that adds value by connotation (as I alluded to via the word artisanal) or by affecting your behaviour (but the latter certainly can be trained if one really wanted to).

        • Forget the filters. They do not work. I’ve tried them and shot a lot of film, and processed a hell of a lot of images through PS and other programs.

          It is physically impossible to recreate the look of 18 stops of dynamic range when your capture device only at most 14 or less to begin with! You cannot make up the missing data. The deep shadow and extreme highlight transitions in zones I and X are NOT the same, not on screen, not in print. Like you being able to find tell tale signs of one set of digital filters vs another, the same are there with film scans vs. digital conversions – filters or otherwise.

          • Neopan 1600 has 18 stops of dynamic range? None of my images with it ever looked as if it had nearly that much. I have lots of blown highlights, at least in my prints. And maybe digital tools applied to B&W film offers just such larger set of adjustment options/range that blows away what you could do in the analogue era (where printing meant selecting between a handful of different contrast levels and selecting the exposure time and not much more. I know there was more, the paper surface, the paper tone, actual toning, differently constructed light with different enlarger heads, burning and dodging but none really modifying the tone curve beyond a simple right and left shift and a handful of slopes to select from. Sure, development can bend the tone curve and affect grain visibly but who would use more than a handful of developers incl. their timing and temperature recipes for which he or she could pre-visualise the effect already while taking a picture.)

            But the key thing that makes things different for me is that I didn’t shoot B&W for ‘fine art’, I shot it (a) because images taken handheld in bad light looked for most subjects rather bad on colour film, in particular once you add the warm lighting (not to speak of fluorescent) but pretty good on B&W, and (b) because portraits or people in general very often look good on B&W or even better than in colour whereas for other subjects one had switch one’s mind to think in B&W to get good B&W, and (c) it was much easier and affordable to do B&W prints than colour prints and thus B&W was the only option to control the printing process. Today, (b) is still true and (a) to some degree as well. I don’t think switching back to film for those uses will deliver something better than digital or only slightly for one aspect but being worse in a lot of other aspects.

            • Neopan 1600 definitely does NOT have 18 stops. Acros 100 does – note difference in emulsion speed.

              It’s difficult to say exactly how much dynamic range there is at the very top end of the highlights – all I know is that it handles them in a far smoother and more pleasing way than digital, assuming you set your grey point to the be the same.

              But clearly I think you are convinced digital is the way to go, and it’s never my intention to convert anybody to anything. I report my observations and publish my images, nothing more. But I do know I don’t have time to waste, and I wouldn’t bother with a process that gives me inferior results.

  6. Another great article and a wealth of great comments too. In terms of the film vs digital comparison I draw an analogy from playing music since I am both a photographer and a musician. For me, shooting film is mostly about being inspired. A guitarist will reach for an old ’58 electric guitar not for its advanced features or its clean sound but because it’s an inspirational instrument to play, it feels great in the hands, it even smells great(!) and the unique sounds it makes evoke happy memories. The sounds from such a guitar can be imitated simply and almost exactly (emphasis on “almost”) using computer technology but it is just not the same as playing the real thing.

    But all that said, on the rare occasions I have been asked to shoot images professionally, the film cameras stay firmly at home in their cases. It is only with digital that I have immediate feedback that I have (most likely) got the shot and have confidence that the image will make it safely to my computer. Film is fragile, it degrades, things can go wrong in the development process, lens caps have an annoying habit of staying on, for me personally film remains an enthusiast’s preoccupation.

    • Thank you. Inspiration is a very good reason.

      As for shooting film professionally – well, I still do, if the intended output warrants it. Sometimes it does make more sense to shoot film for certain applications, but that’s pretty rare. And as you say, the risks are high…

  7. Taildraggin says:

    Tom Liles has a good response for why to shoot film; it’s additive: “digital and film”. (In some of our cases “film and digital”).

    High end digital cameras or photoshop software and race computers to run the software seem to be on a 4 year refresh cycle. Lenses seem to be going that way, too (maybe 8-10 year life?). Decent printers are $600+ ($1k+ really & most do not adequately render B&W) and the inks have annoying life problems, low capacity (output 30-40 prints) and outrageous replacement costs.

    But, very few print anymore. I did a poll of printing habits on an enthusiast/pro website and only 1/3 of the 200 respondents (‘advanced’ enthusiasts and many pros) created at least 5 *framed prints* in the prior year. The other 2/3 did none. Which means most 36mp images are posts to flickr, etc.

    Film is far cheaper for how I shoot: the majority digital, but the stuff I love, film. My sports and family shots are mostly digital. That’s the majority of my shooting. Quick and dirty, it is not practical to shoot those in film. I love my old 35mm cameras, I’ve had them my whole life, but I’m not completely satisfied digitizing 35mm (135) film. 135 was designed as a small (“miniature”) format for fast portable use and really doesn’t distinguish itself any better in the digital age. I use mine because they are fun to shoot and I’m very familiar with them. I wouldn’t use 35mm if getting started today, unless of course, you’re trying to capture the look and flaws for effect (or you just like old gadgets).

    Larger formats rule, though. My landscapes and portraits (shots that count) are made with a Hasselblad or a Graphlex 4×5 Graphic View II. Both are simple, awesome machines. I’m looking at a print of my daughters done by the 4×5 that has such depth that it never fails to capture my attention as I walk past. You’ll start thinking about 8×10 or 11×14. (A friend really likes the simplicity of 5×8, though, it’s the smallest format to create contact prints – no enlarger.)

    Intangible, but film was ‘there.’ The silver halide trapped in the film you hold directly captured reflected light from the subject. The film itself is a souvenir of the event or subject. (A dataset of filtered potentiometer readings does not quite create the same emotions.)

    I shoot mostly Tri-X and extinct Kodak E100G. Opening a box of developed 120 E6 makes your heart skip. Opening a box with a developed E6 4×5 sheet makes it stop. It’s a memorable gift every time you get a roll or sheet back from the processor.

    I’m not a C41 fan; never was. It’s easier, but prints just come out looking thin and plain. I’d rather shoot digital.

    For immediate use, I scan with a V700, which is ‘adequate’ for 120 and quite good for 4×5 (2400 as above). You can get adjustable mounts or wet mount for additional bother and incremental advantage. I just use the Epson as a quality proof/basic archive and send out the keepers for a drum scan, which is very economical for digital.

    The good B&Ws go to the massive Beseler 45 for printing on a relaxing Sunday afternoon, locked away, which is the other half of the story. Printing is magic and releases the full potential of the negatives. It’s what PS is to digital.

    • Digital printing: we’re using a $6k Epson, paper that’s about $3-4 for an A4 sheet, and 11 very large ink tanks. And it’s still %^&*!! expensive to create a print to the new standards we’re using, simply because each file has to be tweaked individually with its own bunch of settings – and guess what: that means anywhere up to 10 test prints to make sure it’s perfect. I think that’s probably too much effort for most casual hobbyists. If I wasn’t good friends with my printmaster, I don’t think I’d be doing it, either.

      I’d love an optical printing system/ enlarger and darkroom, but I have absolutely nowhere to put it in a small apartment.

  8. As a 40 year old enthusiast who is fortunate enough to do quite a bit of travel, I have bounced around this playground a few times. Largely I settled on an Analog to Digital workflow. I use a variety of film cameras going back to my dad’s M2 and do a contact sheet scan via Epson with follow up on keepers. For 135 the scans are OK for digital presentation but for sale or exhibit I either buy time at the local darkroom co-op or beg at local college. I get small, quality cameras with fantastic viewfinders and best of both world for presentation. When I want my touristy stuff I pull out my much maligned but trusty V1 for digital. Side benefit, all of this fits in a very small and inconspicuous shoulder bag.
    As an aside I’ve struggled to find the sweet spot on LF scans that have proper detail without a massive file that clobbers my editing machine.
    Also, thanks as always for your insights and commentary.

    • I think perhaps there is no ‘sweet spot’ for LF as you describe – surely the whole point is that proper detail will result in massive files…

  9. The Minox spy cameras are classics, yes, but so is the Minox 35. And for that, film is not a problem. Instead of Olympus mju, you should definitely consider the XA. I still wonder how did they cram a good, non collapsible 2.8/35 lens in it with a decent viewfinder and even split image rangefinder built in. Something that digital camera manufacturers have still not yet managed to do, not even close.

  10. Très beau traitement B&W !

  11. Iskabibble says:

    I agree with Tom Liles above: I dont care about the cost of film. Even as costs rise I simply suck it up and pay for it. The benefits I derive from shooting film are worth the costs. I spent a *fortune* on digital gear. I could shoot film for years and years and still come out ahead, as long as I step off the digital camera rat race. Thankfully, that step was taken and I no longer buy digital gear, my “ancient” 3 year old digital camera meets all my needs, as long as I stay off the rabid gear infested forums.

    • Well, at this point, the D800E is already two years old. That said, I’m finding myself reaching its limits in some of the print work I’m doing now. Most clients on the other hand remain happy with E-M1 files even…send them a D800E TIFF16 and they complain it’s too big to open.

  12. The difference between medium and 35mm formats in film should be considered as well. You can get really outstanding medium format results with a relatively small investment if you stick with Mamiya and Pentax. Some Rollei’s are affordable and the costs for Hasselblad have come down. You get very conscious of shot discipline with working with 12 to 8 frames. I actually find 36 frames in 35mm too many – makes me frivolous.

    Harry Callahan did fantastic color work starting in the ’40s. Not as well known as it should be.

    • Also true, though I wouldn’t mind a few more shots on MF to be frank. Perhaps 15-20 is the ideal number…hey, that’s about precisely how many we need for a photoessay set or assignment submission allowing for a few discards…

      Thanks for the tip about Harry Callahan.

  13. That’s funny… this evening I started writing a very similar article, and half way through I wrote “…but I still find that I get a far higher proportion of keepers from film than I do from digital…” I shoot film because I tried jamming a CF card in my XPan and it really didn’t work very well 🙂

  14. Good timing MT.

    Before I begin: Andre!, so sorry I haven’t written you yet, I’ve been getting hammered a bit by work and other stuff—the lack of Flickr uploads probably gave it away, but I’m up to my eyeballs. I’ll try to answer you as best I can sometime this weekend!

    OK, so. Film vs. Digital…

    I think MT’s actual real world position is right here with me on this, but, I don’t see any need for the “versus.” At all. Put an “&” in there. We’re photographers. And we’re rampant individualists who will do what we want: shoot film, not shoot film; shoot film, disown film, then shoot film again. It all goes.
    I don’t have a lot of time for the thinly veiled hipsterism in the “buy film not megapixels” movement, the batman-slaps-robin-and-says “shoot film!” elitism, because rhetorically it’s sophomoric and philosophically it’s myopic. Especially that tone of voice—it just makes film shooters seem like douchebags. They’re not. And most of them buy megapixels too! I do. I shoot both, I like both. Digital, nine times out of ten is better — “better” being a slippery concept that changes in time and context — but that one time out of ten for film is worth all nine of the digitals, and maybe more. Digital can be on the edge, but mostly, it is reliable and boring. Film can be infuriating, but it gives us the goods every now and then, and when it does, it’s in a league digital is simply unequipped to play in. And not just technical things, for whatever reason, film photographs are qualitatively different to digital ones. All my film photos seem like a different person took them. And the content is often much more arresting. So sentimental. So inexplicable, in a small quotidian way.
    So I follow my own caprice religiously: whatever I feel like picking up and shooting, that’s what I pick up and shoot. Somedays digital; somedays film. And when it’s one of those days that you want to shoot film, really nothing else will do. Pretending with a digital just makes the pangs worse.

    On dynamic range, a local printer in Tokyo convinced me of this, but the hierarchy for dynamic range is:

    1) Film
    2) Digital media inc. monitors/screens
    3) Prints

    I don’t know but, when it comes to well exposed C41 color negatives, I think there is in excess of 14 stops there. I have a roll of Kodak Ektar I’m trying to scan correctly at the moment, and the reason I keep going back and trying and trying again is that there is detail I can see on a lightbox and through a loupe — it’s physically there in the developed emulsion — tons of it, that neither my DSLR nor work’s snazziest top of the line Epson flatbed scanner seem to be able to retrieve for me. Not just one frame, almost every single one. I would put C41 somewhere between D610 and D800. And if you know what you’re doing [I don’t] there is a fearsome amount of information that can be recorded in there. (acuity is a different matter)
    Mono: When it comes to a good black and white film, like Acros, it seems to me there is even more range than color negatives. I can’t release them yet, but I took some photos for a friend’s fashion brand lookbook some months ago, on a D3. And on film –> just for me, I took my Bronica SQ and Nikon F2 along, both loaded with Kodak 400 Tri-x, not exactly the most tonally subtle film out there, to get some practice shots and b-roll in since we had a real live proper model and it’s not everyday a beginner like me gets that chance. I took four 6×6 Tri-X photos, serious portraits, and about a roll of “etc” type shots on the F2, without even metering, just hail-marying it; I didn’t hold any special hopes. It was just practice.
    My friend was super happy with the D3 files and has made a great lookbook from them. But back before that, shorty after the film shots came back from the lab, I showed him the four 400 Tri-X shots and he was literally stunned into silence. Me with him. We made an A2 print from the one he liked the best from the SQ, and that image has become the face of his brand for the next season. The F2 shots were enlarged to A3 size prints, framed and hung in his exhibition at the Tokyo Fashion Fair [ended today], and he commented that nearly every visitor paid them a compliment—the filmic look, I mean [to which he could tell them: “they’re film!” 🙂 ]. In both prints, the elegant graduations and nuance in the tones was simply sublime and flatly impossible to ape with any amount of post processing a digital file. None of this splendor was down to me: I was lucky with the light, but it was really all the majesty of the film emulsion. That’s game and set and match on why to shoot film. We’re photographers—the image is all. If film is what it takes, film is what it takes. The costs—you pay to play.

    But film, here and now, is ANNOYING immediately after you get developed negs. I’m thinking in particular of color negs. I’ve been trying for over four months now [at various times, not continuously] to crack the secret of scanning and color reversing color negs to the very best quality that is possible—obviously possible if you look at the premo film shots on Flickr and the history of advertising. Mine are horrid. Everytime I feel like I’m close, some exception destroys my confidence and I’m back to square one again. Yeah I could scan at a lab—and believe me, I *have* thought about that a lot, to save the stress of what color negs are putting me through. But something about just throwing my hands up and outsourcing the post creative process to a third party grates at me. I don’t do that for digital images; I want to consider myself a serious photo guy, I shouldn’t do that for film either. I don’t want to do that. Maybe this is my smug hipster elitism [but I don’t put it on a tee, let’s say that for me].

    But back to the economics—I barely even think about it. It’s like being starving hungry, with only 500 yen in your pocket, walking past a sandwich stand selling sandwiches for 500yen. You buy the sandwich! There’s not much deliberation involved. If you don’t have 500yen. Then you steal a sandwich [not recommended] or you go hungry. That’s just the way it is.

    Whatever the costs, I’ll say this: I couldn’t live without the tactile side of it. Loading a camera with film is… well, everyone who does it knows what I mean 🙂

    • ‘Better’ gets more and more subjective as you go further past the basics and into the realms of subjective art. It all depends on what your final output intention is.

      As for dynamic range, I’d say it’s slightly different:
      1. B&W negative
      2. Color negative, tied with modern digital
      3. A good screen, ties with a good print
      4. Slide film, old digital.

      Now, there’s some subtlety to this: whether you can capture all of the DR of an Acros neg with your scan depends on a) how you develop and b) how you scan. I can quite nicely get 16-18 stops of DR into an Acros neg, which is then nicely down sampled again – with continuous tones and no clipping – into the 13-14 stops of a D800E raw file. The resulting output looks very, very smooth. And something I cannot achieve if I’d photographed the scene directly with the D800E. I suppose in effect the negative is acting like a dynamic range enhancing filter of sorts.

      Color neg: puts me off mainly because I cannot find a lab that does it to my satisfaction. I’m sure I could probably get the scanning process eventually sorted, but if you start with garbage in – you get garbage out…

      The printing end of things is what has my attention at the moment.

      • Larry Gebhardt says:

        You probably know this, but a negative is usually developed to a gamma of .5 to .6 (a ratio of original brightness to film density range). This means that the 16-18 stops of original scene (in your very contrasty scene) brightness is compressed to around 8 stops of brightness range on the negative. This will still fit nicely into the D800E exposure range, even leaving you some room to expose to the right to get very clean shadows (highlights in the final). However if you had exposed the scene directly with the D800E the 16 stop range would have exceeded the capability of the sensor to capture the full range.

    • Hi Tom, no problem. I just wanted a sounding board for some ideas. Anyway, the F3 has dug its way into my heart. The Olympus OM film cameras are still in contention just because its shutter speed control seems much easier to use than the F3’s without taking one’s eye away from the viewfinder. If it’s Nikon, then it will be the F3 vs. the FM2n.

      About the film vs. digital thing, as many of you know, there’s a term “figital” which refers to a film capture which is then digitized for printing or distribution. It seems to be getting pretty popular, and takes advantage of the strengths of both media. Even if film doesn’t have a bigger dynamic range (and some of the numbers mentioned are pretty damned impressive), the way it maps the highlight tones, like it has a bigger range there than digital and gives up some tonality on the dark tones, makes it valuable for certain kinds of shooting styles.

      Personally, I don’t think high DR scenes look good — they mostly look pretty dark if done in a linear way. What’s interesting is when the photographer maps the tones so that they bring the really interesting stuff out, and I think one needs much less than 12 stops (8?) to do that.

  15. Ming,
    I think the reasons “why” keeper rates may be higher for some film photographers, over digital photographers, would make for a fascinating article all on its own. The thought process of how we plan, compose and shoot shots theoretically shouldn’t be any different in film or digital; however, it apparently it does make a big difference.

    If you only had 36 shots (like the last Kodachrome, what or who would you take, and how would you take it? My guess Is that it is partially shot discipline or perhaps also a love for reproduction of things “analog” over digital…? I would almost argue you should train photographers on film, so they learn how to visualize without an LCD screen.

    I grew up on film, perhaps as you any many others have. However, my brother (and friends) are really frustrated when I treat digital memory cards, just as if it were a precious resource. To give you an example, I may take 300 shots for an event, compared to 1000 shots taken by my friends. Although I haven’t done the calculations, I have a gut feeling my keeper rate is much higher than my friends (or little brother). From time to time, I’ve tried the spray and pray approach with limited success. I hate post processing…!

    The main thing I loved about film over digital, for quite some time, was less blinkees or blown highlights… Someone made the analogy that digital was, for a time, closer to slide film photography than regular film. Less clipping and a more gradual tonal palate is (for me) why people still have nostalgia for film.

    That being said, not being a pro, it is hard for me to give up digital convenience. Better to adapt film techniques and thought processes to digital, as much as possible.

    • I pretty much treat every roll like the last roll: shoot it only if you are really confident. I like the analog tonal feel for some subjects, digital for others; I wouldn’t think of shooting watches with film anymore – nor would I even want to. But I’d prefer to shoot landscapes with film over digital, though. Perhaps it depends if the subject feels primarily digital or organic…

      Digital is still a better way to learn the basics because of speed of feedback. But to refine your eye and technique, film is definitely the way to go.

  16. Larry Gebhardt says:

    I look at the costs differently. I do have a higher keeper rate for my film shots vs digital, but that’s mainly because I only shoot film for shots I hope to make a nice print from. With digital I also shoot a lot of things that I know will never make it past an online album. For shots like that it’s not worth the effort to scan the film or print in the darkroom, so I know they will never see the light of day. If I had a lab do the work I’d probably shoot more film for snapshots, but then the costs go up.

    If I only look at my digital work that’s intended to print and my film shots that I also intend to print the costs are very similar and low. It costs me more to print digital black and white compared to the darkroom, but the film costs more. All in all it’s a wash in terms of costs. And my medium and large format film cameras all together (not including lenses) cost less than my D800E. Lenses for the film cameras are also cheaper. So I think for an artist that focuses on black and white, and intends to print in the darkroom, film is actually the cheaper route.

    When you add in color work I think digital wins big time in terms of printing. I can print color in the darkroom, and in terms of materials it’s pretty cheap. But I’m not nearly as good, so there’s a lot of waste. And I think the inkjet prints on the new fiber based papers look a whole lot nicer than the RC papers that you get with color papers. The only real reason I shoot color film these days is I sometimes want a color shot and I don’t want to carry the weight of two camera systems. So I carry a few sheet holders with color film, or a Hasselblad back with some color film.

    • I think the D800E prints better than MF film; not LF (well, remains to be seen) – though the differential isn’t that much. As for printing: I think there’s a decent lot of wastage if you’re going to do serious calibration, but I too prefer inkjet: mainly for repeatability and consistency. And the modern barytas are just stunning – I’ve been doing test prints on Canson Infinity Platin Fiber Rag today, and the results blew me away.

      • Larry Gebhardt says:

        The Canson Infinity Platin Fiber Rag is the best of the papers I’ve tried. Simply gorgeous.

        I also think the D800E prints better than scanned medium format film, and as well as scanned large format up to 16×20. But I still prefer both medium and large format film printed on fiber based paper in the darkroom. There’s still something about that type of printing that the inkjets can’t quite match. But it’s gotten very close.

        • I’m not sure I’d go that far; I’m finding that with this new printing method I’m using, the D800E is acceptable at 10×15″, medium format stops at about 10×10″, and ~16MP DSLRs run out of steam around 8×10″. Bring on the 4×5″…now if only my film would arrive sooner!

          • D800E which was sufficient for large prints maxxed out at 10×15″? Feel almost as if I must have mis-read but the text still seems to say it after re-reading a few times. Seems to shift the point of sufficiency….

            • You haven’t misread that. It depends on what you determine as sufficient; we’re using a new printing method with much, much higher resolution – approaching the limits of the paper and not far off a contact print, in fact.

              • Wow. An exhibit / wall sized print would take a rather large file by today’s standards.

                • Replying to myself, granted that exhibit / wall sized prints might not be your intended application…that’s just where my brain went with it….

                • Well, there’s also the important question of viewing distance: you’re not going to look at a 3′ square print at three inches. We have enough resolution for that kind of viewing distance. We have enough resolution for almost any viewing distance: however, we didn’t have the print technology to let us take a digital file and make it into something that small and visually dense.

  17. Ørjan Laxaa says:

    I shoot a lot of film, but developing is very expensive here in Norway. It costs about $16 to develop each roll(35 mm) And since I don’t have a scanner that can take negavites, there’s a scanning fee as well. To get the lab to scan in high-res TIFF is ridicolously expensive.

    • Ouch. That doesn’t make sense at all – in fact, these days, getting a lab to do any part of the process in any part of the world is very expensive proposition. DIY is the way to go…though admittedly difficult for colour.

      • Iskabibble says:

        Expensive? I pay $5 per roll of film for development and scanning. I wonder if you could send your color film to Shanghai, where I get my film developed. It is very cheap, the scans are AMAZING, and I have always been happy. How expensive is postage to Shanghai from KL? (here’s one scan: )

        Fuji 400H and Reala (now going extinct, get your rolls NOW) are AMAZING films. Such gorgeous color, makes me want to shoot even more.

        • I probably could, but I don’t want to take the risk of things getting lost/ damaged in transit – seems to be happening too often lately with Malaysia post. In any case, I still prefer digital for colour because I have more control over the whole process…

      • Just got back from the lab. Developing and scanning two rolls of 120-films(B/W, one Neopan and one Ilford) cost me $97! And no decent results as well. I think the time has come to start saving up for a scanner.

    • I really encourage you to develop yourself. If you pick up a Jobo CPE or similar then you will be able to produce C41 colour very cheaply. A Tetenal C41 kit is about €25 and you can easily do 15 rolls with it. Also having to use up the developer within a certain time period encourages you to shoot more as opposed to paying a premium for processing.

      There is an enhanced sense of satisfaction in self development too.

    • try this – just for fun:
      Sorry for that language…, but it´s easy visible in the pictures and very cheap:-)

  18. I agree with your observation about film being more limited in tonal response but I personally love the “tonality” of certain color films. I always found the digital development process very cumbersome.

    As usual, great choice of subject. Would love to read about your film scanning process.

  19. Carlos Esteban says:

    Hi, Ming, I think there is a missed factor: the existence of a local film dealer/developer and so on. None next to me in Brasil (of course anyone can home develop negatives/positives – and that’s kind of magic looking a wet negative, the first sight of negative through amplifier, the final image appearing in tray etc – but… i’m pretty much lazy now since digital era).
    By the way, you said that a camera last for 8-10 years – ok, but do you know if it is true about digital backs? (Besides, i got my first one just yesterday, gave it a very unfair test – my photometer ran out of battery – and got stoned by it’s b&w converted file).


    • No decent local film developers here, either; and the Fuji Acros I like has to come from Japan. I develop my own and order film in bulk. 🙂

      In theory, digital backs should last longer since there are no moving parts…it’s just a question of availability of batteries, physical connector/ switch longevity etc.

  20. Damn, was hoping there’d be an announcement about your scanning rig at the end!

    Excellent article; I was having a similar conversation with a friend only yesterday. I’ve given up on film for now, because I find digital more useful for learning with–instant visual feedback, histograms, EXIF, and just the ability to shoot more. A week’s holiday last summer shooting the FM3a (in manual) was possibly the most fun I’ve ever had with a camera, though. (By contrast, the D800 I’ve been shooting with since is possibly the least fun I’ve ever had with a camera.)

    • The D800E grows on you. Very, very slowly. I’m beginning to enjoy mine more: the fact that it always delivers without complaint is something fairly rare these days it seems…

  21. Ming, I recall that you had mentioned your DSLR scanning rig before and you mention it now again. Have you published details of this yet? I’d be interested in something like this – I use a Leica S2 and would find it quite attractive to be able to do high resolution scans of medium format (6×6) film at a higher speed than on the usual film scanner (used to own a Nikon 9000 – great machine, sold it for more than I paid for it new, but I don’t have the time to scan the usual way, Nikon stopped updating the scan s/w long ago – and I never could get Vuescan to work properly and recognize film frames).

    • Nope, because I’m still not happy with the final product. I think a heavy reengineering is required.

      I don’t think it will work with the S2 though because of the working distance and minimum magnification.

  22. Oversimplified indeed, but very interesting nonetheless. For my usage, no matter how I calculate it, digital always wins in terms of cost and convenience. However, there’s the “solution” to shoot both.

  23. Thoughtful as always. I treat all my manual lenses (high quality ones), and film bodies (ditto) as if they are non-depreciable. Of course they will wear, but based on past performance they will outlive me, so its irrelevant. Digital bodies (and to some extent AF lenses built for them) I treat not on how long they might last (your 8 years +/-) but on how long I might continue to perceive them as useful / easy to use. The problem is twofold – firstly its GAS: I will lust after the new toy when the old one has lagged a long way behind (my D200 was “sufficient” – to borrow from you, last ear when I sold it, but the D800 was too attractive); and secondly, the software and hardware in the camera at some point are no longer supported within an up to date digital workflow – like everything else in that flow. I think 3 – 4 years is more realistic (for me). So long as I use that camera body enough, my gut feeling (no hard maths) is that its cost (life-span divided by keeper shots) is comparable to film, with maybe 10% – 25% back at the end on eBay.

    • Of course, that doesn’t take into account the fact that the cameras might develop some odd gremlin halfway through its 4-year life – usually precisely when the warranty expires – and then be uneconomical to fix but too expensive to junk. I’m finding with the increasing complexity of digital devices, this happens more and more often.

      That said, modern digital isn’t built to the same quality as the early stuff; I’m actually shooting with a 2003 Olympus E-1 at the moment and am extremely surprised by how solid the camera feels in hand, and how smooth the operation still is…the files, of course, are a different matter entirely.

      • E-1? Vintage digital camera review? Or maybe getting ready to help Olympus do an E-5 replacement … (Fanboys go crazy!)

        • Definitely the former. I gotta dig out some vintage CF cards to go with it, too – it doesn’t seem too happy with the 32GB UDMA7 cards I use in the D800E/ CFV…

          • I’m curious what’s the motivation for this. I can’t imagine this being a very interesting area to explore, but there’s probably some interesting twist I haven’t considered … BTW, I have a couple of old 4GB IBM Microdrive (spinning platter and all) CF cards if you want to get really old school!

  24. For someone who knows exposure for a scene and doesn’t bracket, the cost is less. This has always been a flaw in film vs digital cost comparison. A lot more people need bracketing or a try this, try this, try it this way to finally get it for them; thus cost goes up by 3-6 times what you stated. So the $2000-$3000 needed to replace film camera equipment seems a lot, until the # of shots hits the thousands. It’s what needed for some to “get it”.
    36 exposures X 52 rolls (1 per week) = 1872 shots. At $16 each (film & process & cd) X 52 rolls = $832 year. Do $832 X 3 (- , 0, + exposure shots) = $2496 ; now your at replacement cost. Add shots of same scene with different perspective, angles, lenses, and you’ve tripled or quadrupled that cost for film ($9984), but digital stays at that $3000 mark. (Honestly this was gone through before when auto focus came along in the late 80’s – that AF camera would get you the shot).
    Innate ability helps a lot, which you have, the rest of us learn mostly by trial and error. My keeper rate with film is about 15% (6 out of 36). With digital it’s probably 2%, but I’ve finally learned how fill light works. For $1800 (camera, lenses, flashes, white board), that’s cheap vs what film would have cost.

    • I had my fair share of trial and error with film too; that admittedly applies to digital but with lower associated costs. The *starting* costs of either are pretty daunting; I’m looking at this from me point of view of a photographer who might already a lot of the other ‘system’ gear.

      • Digital is much cheaper, there is no way around it. I think coming out of the gate with a D600/D700 and comparing it to film misses the comparison mark. It would be akin to buying an F6 which in no way is cheap for a film camera and that tech is already old. People say you can go out and buy an old used F3 for a couple hundred, but you can do the same with digital; a used D70 or D40x isn’t much. Digital is just cheaper out of the gate. It is the perfect platform to learn on. Film might be better to master on. But then back to digital to capitalize on.

        • Digital is cheaper per shot. But that doesn’t say anything about quality vs quantity. The difference is an F6 or an F3 will produce identical image quality; a D70 and a D700 will not.

  25. Great article!

    While it is fun to shoot film, develop the rolls yourself, and scan them, it does take away substantial amount of time especially if you are a 35mm street photographer and shoot a lot more rolls than a meditative medium format photographer. And if you are a beginner, development and scanning are not really straight forward – they are a learning process themselves demanding experimentation and time. The time spent on developing and scanning film can be used to read photography books, analyze your pictures in more depth, and of course on the street shooting more pictures. For the digital shooter, key thing is what you have mentioned many times – shot discipline. Some times I wish digital cameras were exactly like the old manual rangefinder cameras with the sensor replacing the film cartridge! 😀

    • It’s also possible to shoot less and think more before the exposure, but I believe that digital is probably a much better option for a prolific street photographer. That said, I shoot street with the Hasselblad…

  26. This is a timely article for me! I’m trying to shoot film with a setup more unobtrusive than with the Blad, which means a 35mm SLR, because I don’t think I’m a rangefinder person. I just borrowed a friend’s F3 (because I have to scratch the Nikon itch), and revived my old Canon T70.

    One of the main reasons I want to shoot film is to play with its highlight region: I just noticed that most of my digital pictures are low key, and sometimes, as you may know, too low key! High key in digital doesn’t look good to me because of digital’s crunchy top end, the various lens issues digital uncovers in high contrast regions, and the giveaway signature look to digital pictures that one raises in post. I haven’t yet investigated HDR as you’ve suggested. Anyway, the point is to shoot more high-key pictures on film, on both Portra and Acros.

    Some differences I’ve noticed so far:

    – Film cameras are cheap! And sometimes friends let you borrow theirs because it’s been sitting on their shelf for years.
    – The high-end stuff feels fantastic. The feel of the Hasselblad and the F3 are great. The F3 feels like a metal billet that happens to shoot photos. The cheap stuff is pretty cheesy, as usual (and the T70 is going straight back to the closet because of this). A digital camera feels like using a computer, which is OK, but …
    – Information display and exposure control on film cameras are far behind the best digital cameras (ie. E-M1). With the film cameras I have, you cannot easily set a manual exposure without taking your eye off the viewfinder. This is fine because you only have a relatively small number of frames per roll compared to digital so the enforced deliberation is good. I also feel no desire to add a motor winder to the camera.
    – Relatively long battery life, though the batteries can be pretty damned expensive: I just paid $15 for the F3’s tiny DL1/3N lithium battery. And weirdly, paying $50 for an Olympus BLN-1 somehow that feels cheaper because it’s rechargeable.

    But this is all ancillary stuff. If the images film can produce are good and what I want, then it’s definitely worth it. Now to get the hit rate up above 50% …

    Also, the highlight tones in that portrait are just lovely, and a great reason to shoot film.

    • “the highlight tones in that portrait are just lovely” Hm.. I found the highlights a bit too bright and distracting, i.e. they take away attention from the subject. But digital would have blown them away and film is so much better.

    • You’ve hit on the main reason I shoot film: tonality. And it doesn’t do any harm that the gear itself is much nicer, too – the viewfinders especially. I wonder if this has something to do with my much higher keeper rate for film…

  27. Sigmund Krøvel-Velle says:

    Hello! You mentioned your scanning rig. Can you tell us about it? And how you do it?

  28. Hi Ming.
    I’ve had a long conversation with myself about going into film – which in the end, I did not. Still look up to those doing it.
    Personally, I felt somewhat daunted by the darkroom and chemicals, although I have no doubt that would go away with time. And speaking of time, that’s what finally turned me away. As I haven’t processed film I don’t have any facts here, but I imagine that film is more time consuming than digital? At least strictly per image. Which does not necessarily apply any more if you factor in the amount of frames, film versus digital, plus the occasional blog reading in the digital workflow. One can of course also say that time spent on something like film developing is far from a bad thing, and I can’t argue with that, but at the moment it’s not for me.
    So in the end I stayed with digital. I did however do two things to raise my effectivity. Simulate film constraints by putting those old SD cards to use. 128Mb gives me 8 shots of 12Mp. I often take one of those along when I go for my evening walks. Secondly, I did a merciless clean-up in my blog RSS, of which only you and a few other “low frequency, high amplitude” writers made the cut.

    Thank you for your writings, it means a lot to have a solid source of information in this day and age!

    • A pleasure.

      I personally find film is better in terms of final keepers per time spent because of the extra upfront thinking time the whole process enforces on you. I suppose using small cards might work too – I actually don’t have much of a choice because of the size of the Hasselblad digital files 🙂

      • Ming do you find that the time between taking photos with film and development help you with culling or removing your self from the images to help you objectively chose best shots?

        I shoot film occasionally for fun but I find that by the time I decide to develop it I have forgotten half the images I took on the roll.

        • Not really; most of the time I’ll shoot and process in fairly close succession; certainly about the same ‘gap’ as with digital. So that portion shouldn’t make any difference.

  29. Patiently waiting for your DSLR scanning rig : )
    The rolls of spent film are slowly piling up. It’s gonna be like Christmas when I finally get to look at the images in detail.

  30. Wow! That’s a huge disparity between digital keepers, and film keepers. You must be a really crap digital photographer 😉
    But in all seriousness, do you think this disparity is caused purely by selectivity , or does it also reflect upon your growing appreciation of your Hasslebad? Hmm, you did remove the D800 from the equation, can one then assume all your film shots are on 35mm film?
    Either way, I consider 50-90% keeper rate exemplary, and certainly have no reason to doubt you, although sometimes I wonder if you are over critical of your digital images.
    Anyway, for readers, both are outstanding , and help “make” the site….picture being worth a thousand etc.
    /digs out old rangefinder, and tries to find a good film shop in KL…..:/

    • Yep, I suck at digital! In all seriousness, it might be because a lot of e digital work is for clients, who have different expectations (and sometimes instructions that don’t make sense, but require experimentation to achieve because that’s not always how light works in real life…

      Good luck finding film in KL. I get some direct from the Ilford distributors, and the rest from Japan.

  31. The last time I paid for a high-quality scan (of a 15 x 24 cm print, though), I paid something like $20 per image.

  32. More important to me other than the cost of film is the additional mental involvement that goes with it. By shooting manually everything is controllable within my process. From metering to film stock to developing to making an optical print. By being subject to these stages, no doubt, makes you a more complete photographer.

    This is not to say that by purchasing a D700 and a 50mm f/1.4G you can’t become a more involved photographer but laziness is a sickness in this field. The proliferation of bokeh based, flower/twig shots on Flickr and, more worryingly, the interest and appreciation of the mundane might suggest otherwise.

  33. Hi Ming,

    I completely agree shooting color digitally will weild better results that are more consistent and B&W is still, in my opinion, best left to film. BUT if your desire is to color correct your images to have a cinematic look I find cinema film the way to go, especially since you’ll get the desired color tones your looking for at a greatly reduced cost. 200′ feet of lab tested short ends will run you approximately $30 – $40 USD and will allow you to spool 40 rolls of it. The only down fall is you have to develop it yourself.

  34. Ming, realistically despite sufficiency, I doubt people will keep their cameras longer. Gear lust is a significant factor not to ignore and I think the economics of that aspect would come into play. Skill is probably another factor as well, I doubt many have the capability nor the discipline to even develop film, let alone do it properly. At least in Malaysia at least. Shot discipline on film really only comes to play for a photographer like you, I’ve seen many film junkies walk around shooting air. Literally. I doubt they have a 50% hit rate. As much as I love film and the whole experience of shooting a fully mechanical creation of beauty, I have to admit I’m just too lazy to develop film. Maybe I’ll get around to working on the many rolls of film lying around in my house someday. Someday.

    • That’s probably true. For a lot of people, film work instantly has ‘artistic’ value – regardless of actual merit. It’s probably because of the perception that a certain amount of commitment is required to do it in the first place.

      Ironically I am shooting film because I want to bring that level of concentration and conscious consideration *before* the shot to my digital work.

      • Manual focusing, I find, helps me with concentration and thinking even on digital cameras. Turning off the LCD is also helpful. I wish digital cameras had an option for a wait time of 10 seconds or more between each shot. 😀

  35. Steven Lawrence says:

    I started back to shooting film a few months ago. I only shoot 35mm and use Portra and Ektar. I shoot differently than with digital. When I use those films I am shooting with color in mind as much as I have in mind the composition. I still shoot digital but at the moment film is my favorite. I do have my film scanned when developed so it does cost more. I may get a dedicated film scanner but for now my workflow does the job for me. I use a Nikon F100 and N75.

    • I think you make an important point about color film: the tonal palette is quite fixed, so you’ll need to be conscious of which subjects work and how, at the time of shooting. The work of Saul Leiter and Steve McCurry are great examples of this.

  36. Hmmm…. Delta 400, 100 and PanF in Rodinal at 100:1 dilution, stand development.
    Bulk 30m rolls of film, spooled out as needed.
    Dunno the exact figure but it’s South of $0.1 per image all-up, minus scanning.
    Which I already do and have done for over 10 years anyway.
    At that cost, I can afford to splurge and bracket out! 😉
    But of course, it’s b&w. Colour is a bit more tricky.
    I prefer E6, but that one is getting more and more expensive.
    Might have to stay with C41 from very soon onwards.
    It’s easier to scan with the Opticfilm120, anyway.
    I’m keeping the Coolscan9000 for E6.

    • I keep seeing stand development recipes for Rodinal, but nothing for DDX – presumably the theory is sound, but I guess I will have to do some experimenting…

      • I’ve found stand development to give sharper negatives, but at the expense of evenness and other artifacts (like air-bells). I’ve ruined more sheets of film trying stand development than I’ve improved with the added sharpness or compensation. Most of my experiments were with Pyrocat HD and not Rodinal or XTol.

        • Hmm, that doesn’t sound encouraging at all. I think I’ll stick to my semi-stand method – 10-15min and agitation/ inversions every minute…

          • Sheet film is not recommended for stand development precisely because of the difficulty in managing air bubbles.
            35mm on the other hand works perfectly in stand development provided:
            1- one uses a pre-soak in water to get rid of the back coatings and their tendency to attract air bubbles.
            2- one uses a developer that works well for this. Rodinal is perfect. DDX? I have not tried it.
            3- one takes care to tap the tank sufficiently to dislodge bubbles. It’s surprising how hard that may have to be:
            I’ve tried opening a tank with a sacrificial film to learn how hard to tap and it was scary. Took a lot more whacking than I thought!
            I use DDX in normal development for Delta 400, like the resulting negatives a lot. Dunno about XTol, haven’t tried it yet although I’ve heard glowing reports with Tri-X.

            • Thanks for the tips. I’d have thought that sheet film would have issues because of gravity/ solution depletion rather than bubbles…

            • Larry Gebhardt says:

              Noons, what you say agrees with my observations. One more factor I’ve observed when I opened the tank with a few sacrificial sheets to figure out the issue is I found the air-bells were more prominent where there were finger prints on the sheets. I think the oils provide nucleation sites for the bubbles to form on. They proved almost impossible to dislodge without risking damage to the plastic combi-plan tank.

              Maybe I’ll try stand development again with roll films and metal tanks. There you can bang the tank harder, and you touch the film much less with roll film.

              • One note about tapping. I just reread my prior reply and noticed it might convey the notion one needs to give the tank an almighty whack. Apologies: that is wrong and it could cause more bubbles! 🙂
                What I was trying – unsuccessfuly – to say is that it is better to give it more small taps than just 1-2 big ones. I got the hang of it by opening the tank with the sacrificial film and with it open I simply could not whack it very hard without splashing! 4-5 firm taps with only about 1 inch travel seems to do the trick for me with 35mm rolls. But I stress: this is for my film/developer combo. I think the best is really to give it a try with a sacrificial film so one can get the hang of it for the particular combo used.
                The pre-soak was the biggest factor as I found. The anti-halation coating of many modern b&w films, while not afecting development chemicals, does indeed cause a little more problems with air bubbles and such. The pre-soak helps getting rid of it and – my theory, here! – makes the film more receptive to “wetting” by the developer solution. Anyways, like I said: YMMV.

                • I think a sacrificial roll is in order, both to test the tapping – though I suppose one could probably practice with a glass of soft drink perhaps – and the stand development time…

                  Just to confirm a popular rumor: you can mix and match ISOs in the same roll with stand development, right?

                  • That’s part of the fun, yes. Mind you: I wouldn’t go for a very wide range, it’s not a miracle maker! 🙂
                    But it can easily cope with 2 stops ISO up and down without any problems.
                    The thing I like is how it manages contrast so well, particularly with “pan” slow film.
                    The negatives are a lot easier to scan as a result.

                    • That would make sense – it accommodates the maximum tonal ability of the film itself…

                      Pan film has great midtones/ 3/4 or 1/4 tones, but not all at the same time; you have to pick. What I find especially amazing about Acros is that you don’t have to pick…

  37. Thank you for another of your interesting discussions! Speaking of scanners I just looked only briefly at Hasselblad Flextight X5 scanner. Only briefly due to the roughly $23000 (US) price tag new before accessories. Seems *that* device would help bump up the per-image cost! 🙂

    • Bump is the wrong word. Kick through the roof!

      • That’s one of my main grieves with film.
        If you don’t either own a fullframe DSLR to scan with or already have a Scanner, you have to spent atleast 600€ (for a flatbed) to 1500€ (for a halfway decent medium format scanner) to be able to scan your images. This of course does not apply to professional photographers like you, but for amateurs like me it kinda sucks.

        • The midrange Epsons are quite decent; only the top of the line V750 is EUR600. A couple of hundred gets you passable results for medium format upwards. Small formats require much higher precision to make the most of the relatively limited data.

          • That’s exactly my problem, as I shoot 35mm and 120. But I guess I will bite the bullet and rather invest a little bit more, than not being able to scan 35mm properly.

          • Ming,

            You say that the V700 results are “quite decent”. Have you ever drum scanned? made some large chromira or lightjet prints from the resulting high res quality scans? I suspect not.

            Back in the day it was all pre press high end flat beds (creo or eversmart etc) or drum scanners (tango etc.). To say an epson is “quite decent’ is “quite ridiculous” if one cares about extracting detail of out of film. Even a lowly Nikon 9000 blows the V700 to shreds. I know. I own both and only use the Epson for fast proofing before I scan for medium sized prints with the Nikon 9000. If I want to go real large with a given piece of film I send the negatives out to be drum scanned.

            Not trying to be insulting or combative but its clear that a generational difference exists in some of your posts. Young people have no idea how good film can be. Yes, digital is better now but until the D600 or Canon 5Dm2 there really was nothing that could trump a well scanned 6×6 piece of Film when it came to larger print sizes. Yes, it was and is expensive to go this route but no pro who had invested in a Bronica or Hassselblad circa 1995 (yes, these cameras were very expensive) would be caught dead printing anything as soft as the output from a consumer Epson V700.

            Just some perspective for your readers.

            • It’s decent compared to what most people expect – a similar analog is the loose pixels that are produced by 99% of photographers because they do not understand shot discipline. I agree that flatbeds do not maximise quality, far from it. But beyond 3000 – real – DPI, you’re only scanning grain anyway. I’ve used a Flextight and if you’ve read any of my other film articles, you’ll know I do all of my digitising work now with a D800E because the flatbeds are not good enough – and the D800E delivers very similar results to the Flextight, when both are used properly.

        • I don’t think you need a FF SLR to scan film. You just need to get a macro lens whose magnification, focusing distance, etc. lets all of a sensor’s pixels see the negative. If you can get all of an APS-C sensor’s 16 megapixels to see the negative, that’s just as good as an FF sensor’s 16 MP in terms of resolution. You can also stitch together higher magnification shots if you want more resolution.

          • I have been very underwhelmed with my epson V700. I’m hoping a Nikon Coolscan 5000 will yield better 135 results. As for 120, I’m hoping to sort out a macro lens set up.

            • Hmm, that’s not encouraging. Underwhelmed how?

              • Maybe Hugh is seeing what these guys saw? 2300 actual DPI, and Dmax range of considerably less than the stated 4.0:

                • 2400 bayer DPI looks good to me; 3200 is a bit beyond the grain threshold. Since scanners are all RGB, 2300 actual DPI should be pretty good. Can’t comment on Dmax though.

                  • Primarily because the critical focus distance of the V700 is hard to find and the scans always seem to be soft under close inspection. You can improve things through expreimentation but I only recall being truly happy with the results just the once and I haven’t been able to replicate that level of sharpness since. Lets imagine making an optical print, you are able to focus exactly on the grain and you know that the slightest deviation of distance will render the grain blurred. With a V700 you have no such exacting control and as a result, critical sharpness is all rather hit and miss.

                    The quality of the film holders are also poor.

                    From what I have read the Coolscan series produce much better results. Perhaps it has an auto focus system?

                    • I don’t think flatbeds are especially good for 135 frames because of small they are. At 2300 DPI, the Epson would produce the equivalent resolution of a 7 MP file, and going by Ming’s experiences, that’s leaving half of the potential resolution on the table. Some people are happy with the Betterscanning film holders and feet for flatbeds, but the whole flatbed thing strikes me as a bit too fiddly for 135.

                      MF and bigger negs are fine for flatbeds, but you need something like the Nikon Coolscan or the Minolta Dimage scanners to do a good job on 135, and they’re discontinued and very expensive on the used market. The Plustek Opticfilm 120 is available new, and said to be very good as well, but it is pricey. Perhaps DSLR scanning?

                      Rangefinderforum has a dedicated scanning forum that has a lot of discussion and example scans.

                    • ~3000 DPI is about right. Some film scanners can do 135, but they’re very, very slow.

                      DSLR scanning is the way to go for MF, but of course your DSLR must exceed the resolution of the film you’re trying to scan – or the messy process of stitching starts…

                    • Ah: it sounds like the scanner is misfocused or there’s no optical system over the CCD?

                      I’m guessing part of the problem is a lack of flatness…

                    • I did go down the betterscanning route and using ANR inserts but the issue there being that it introduces another surface that can bring dirt and to compound that problem the use of ANR disables ICE functionality so any gains in flatness come at a cost of manual dirt removal, which annoys the tits out of me.

                    • I scan/copy immediately after drying, which seems to mostly eliminate dust for me. I do very little to no dust spotting.

            • So hopefully Ming won’t mind me posting this link here … Pekka Potka’s written about using his OM-D and the Olympus 60 macro to digitize 35mm slides and negatives, including a 4-frame composite method that yields 35 Mpixels. There are notes in there on how to reverse C-41, too. His results look pretty darn good.

              The slide duplicators are pretty cheap, and Olympus in the US has a $100 off deal on the 60mm til the end of this month. It looks like you’ll need to order a negative strip carrier as well for unmounted film (see the last comment in the post for a link). For making composites, you’ll probably need to take a hacksaw to a tube to shorten it, so it’s not for everyone. I suppose one could also hack a plastic plumbing tube to the right length, and use some method to precisely calibrate the co-planarity of the two ends (multiple layers of tape?).

              It’s certainly cheaper and faster than the used Nikon Coolscans on eBay, and Pekka says he gets better results than anything except drum scans (and that includes the Nikons, which were the benchmark consumer scanners for this kind of film scanning). And you get one of the best macro lenses available to shoot things besides negatives! Don’t forget to use the affiliate links to order the lens and paraphernalia, too.

              • I wonder how he shifts them up and down; but beyond that, I’m not sure I see the point in shooting four and stitching – firstly, that’s extremely slow for a whole roll, and secondly, look at his ‘100% crop’: other than resolving grain, there’s nowhere near 35MP of information there. Put it this way: if you’d gotten something like that out of a DSLR, you’d assume you misfocused. A single frame would be enough.

                • I believe the frame of the slide holder can move up and down, but I haven’t played with one yet. Later on he says that he usually does 1 shot, which he’s printed to A3 size with no problem. 4 is for when you want to be absolutely sure you’ve scraped every bit of info off the negative. For me, I usually digitize the whole roll in one shot laid out in their archival sleeve, so I have a contact sheet, then I go back and do high quality shots of the ones that are promising.

          • Agreed. I find the limits of a good negative are somewhere around 3000dpi; beyond that there’s no more information and you’re just scanning grain.


  1. […] Download :Film diaries: examining the costs of shooting film vs digital […]

  2. […] in the way we shoot film vs digital: a lot of comments from an earlier article examining the economics of shooting both media to a similar output standard suggested that this is the same for a lot of other photographers, too. […]

%d bloggers like this: