Originally posted 2013-07-03 02:32:06.
So what is there to sing in praise of film?
Surely it is a nasty, dirty, smelly procedure best consigned to the bucket of history? Surely digital is cheaper, easier, faster, more modern? And worst of all, film is analogue—well that’s just not right.
Well, between 2005 and 2008 I used digital exclusively, and observed a few things about it. First, it’s not as easy to use as some people are guilty of making out—the latitude of a digital sensor is nothing like as wide as colour negative film and in fact is closer to transparency, as any newcomer to digital who has just seen his highlights totally blown out will testify.
Secondly, speaking as a professional photographer, going digital didn’t save me money, it cost me money—a LOT. Many thousands of pounds a year in fact.
How so, I hear you cry? Once you’ve bought your storage medium, digital is FREE. How can it have cost me money?
Well, that is because I, like most pros who work to commission, regarded film and processing not as a cost centre but as a profit centre. We charged our clients for the film we used and we charged a healthy mark-up too. In other words, selling film was a money-making part of the business.
The same was true of prints except much more so; why do you think I had a full colour laboratory in my studio complete with Kreonite Pro-Mate processor? You think that was because I like the smell of esters? Er, no, even at trade price I was making a very damn healthy profit per finished print thanks very much, which is why I used to come out of the lab grinning like a cat. Singing sometimes. You wanna know how much profit there is in a couple of hundred 8×6 prints sold at £8 a whack? Do that every week and yeah, you miss it. You even get to like the smell of esters.
And just don’t get me started about Polaroids—a licence to print money if ever there was one.
What all that meant was that I, and others like me, could charge less for my time because I could make plenty on the film and prints. It also gave me control; I knew a PR client, for example, would be sending my pictures out to be published, and I would get nothing for the reproduction; so I charged a good price for the prints, which made it bearable. When my clients started saying, ‘Just send us the files on disk,’ I had to rethink my entire pricing structure. It still seems strange that someone who needs one digital copy each of three pictures might have to pay as much as someone who needs twenty copies each of ten, but that’s the way it is now. You pay for my time and the nature of the usage, not for quantity.
Obviously, in order to maintain the standard of living that we as photographers had got used to, we have devised new ways of pricing; but I hope I have finally laid to rest the blatant lie that digital is more cost-effective for pros. It is for amateurs, but not for us.
So is Digital Easier than Film?
So having shot the ‘Digital is cheaper’ horse smack between the eyes, let’s draw a bead on the ‘Easier’ nag.
Nope. Using colour neg film in a modern SLR camera is far, far easier than using digital. Colour neg film latitude is enormous. Yeah, I know, as long as the sun’s over your shoulder, digi works fine. And if you’re an experienced pro you’re used to taking the little matter of exposure seriously and probably even have a hand-held meter (a what?) in your bag for those times when even the best in-camera meter could get fooled. Yeah I know you can look at the screen on the back, but that’s a bit like bracketing—it just proves you’re not really in control. And the time you get that once-only opportunity, you can guarantee will be the time the exposure screws up.
Nothing like that killer digi cut-off blowing your highlights right out to teach you old Ansel had a thing or two right.
Digital is faster than film
Digital turnaround is faster. Oh boy is it faster. And is that ever a pain in the neck sometimes. See now, your graphic designers (GD’s) and other curses of the hardworking pro, many of whom seem to resent the very existence of professional photographers anyway, think all you have to do is whip out the card and give ’em the pix.
Bullshit. Even the best digitally captured images benefit from a tweak or two in Photoshop, but that takes time. Oh yes indeedy. The people that commission photography knew they had to wait a while for their prints or trannies and they didn’t take into account that it wasn’t just printing we did in the lab, it was image manipulation—cropping, dodging, shading—or for that matter just removing the dud trannies before they got to the client. So they waited. Nowadays, even if there’s no real rush and they could easily wait, they want the pictures……NOW.
And then, consider your photog who has taken maybe 250 snaps at a wedding. Each has to be checked, Levelled, possibly Unsharp-masked and a range of other tweaks, by which time his eyes are crossed and he has a pounding headache. Yes, you can automate a lot of this work, but not all. Is it any surprise, I ask myself, that I know pro wedding snappers who have actually gone back to film?
But digital is sharper than film, isn’t it?
Digital is sharper, then. C’mon. We’ve seen the tests. Digi’s got resolution nailed.
Yes and no. Compared with 35mm, 6 megapixel cameras can resolve more than 200 iso colour negative film. 12 Mp cameras can resolve more than 100 iso colour negative film. And new digital 35mm replacements far more. This isn’t actually saying very much but it does mean that for a very wide range of the purposes of very demanding pro photographers, digital is plenty sharp enough. And as the film speed goes up, digital really shines—a 1600 iso colour 35mm will look like golf balls in soup compared to a 1600 iso digital capture, and at 3200 35mm colour film is an embarrassment. I know. Been there, done that. So digi is great for editorial, PR, sport and wildlife photographers (just so long as they watch those highlights.) However, larger film formats leave digital behind, as we shall see.
Furthermore, the advocates of digital are guilty of the pretence that the resolution of the sensor is the only critical factor in quality; it is not. However, sticking with the issue of resolution, they also omit to mention that factors other than the maximum potential resolution of the sensor, or film, are often the limit to the maximum achievable resolution of a given image.
There are many such factors but let me begin with the simplest and the most frequently overlooked—movement. Having been a senior photographer for many years and run my own photographic business for over two decades I can assure you all that this is the single most common technical fault I have seen both in my own work and in that of others. Movement can either be of the subject or of the camera, when it is called camera shake. Although movement can be drastically reduced by good technique, it is impossible to totally eliminate it, and it will always be a limiting factor on ultimate resolution.
The other frequently overlooked issue is lens resolution related to enlargement factor. We have, thanks to industry PR, become fascinated by ultimate lens resolution and there is no doubt that modern computer-designed lenses are very sharp indeed. However the principal benefits are seen in zoom lenses, which now are as good as many prime lenses, but no equivalent leap forward has taken place for prime lenses themselves.
For any given sensor or film size we need to know the enlargement factor for the ultimate print before we can assess the sharpness of the print, which is what actually matters. This is a crucial test because any error of technique or of optical performance will be multiplied, and made more apparent, pro rata, as the captured image is enlarged. In 35mm the reference was usually taken, though somewhat perversely because of the different aspect ratios, as being a 10×8 inch print.
Now, a 35mm negative will need to be enlarged by a factor of eight to make a print of this size, so let’s use that as a reference. The Nikon digital sensor is 23.7×15.6mm and will require an enlargement factor of about 12 times for the same print. However a 6×4.5 rollfilm neg will require an enlargement factor of between 5 and 6, and a 6×7 rollfilm neg about 3.5x. So for an image captured on digital to achieve print resolution as high as a 6×7 can, the digital camera lens has to be able to resolve between three and four times as much detail as the film camera lens. (Note: not the sensor; the lens.) The improvements in lens design do not even come close to compensating for this, and furthermore medium format cameras are usually used with extremely high-quality prime lenses, which as we have seen represent the standard to which zoom lens designers aspire; so by the simple rules of optics, the claim that 35mm equivalent digital cameras can compete with medium format film is proved to be false—just another piece of marketing bull.
In fact, the effect of the increased enlargement factor will still be apparent when we compare a half-frame digital sensor like the Nikon with full frame 35mm, even though a modern zoom lens is more likely to be used on the 35mm camera. The issue of lens sharpness is most critical in regard to wide-angle lenses, as to achieve full coverage of the negative or sensor their image circles have to be enlarged as much as possible. Since lens design involves a compromise between the size of image circle and centre sharpness, this is a headache for half-frame, which needs wider lenses, and is one of the reasons why there are now so many “digital-only” wide lenses available, which have much smaller image circles. (Another is the incident angle of light falling on the sensor, but that is not relevant here.) Granted, modern aspherical, moulded lens elements have been particularly effective in improving the quality of wide-angle lenses without reduction of image circle, but they have principally been used as a means to make short zooms competitive in quality with primes.
Sensor vs film resolution
This leads us to several inevitable conclusions; one that sensor resolution is less important than we have been led to believe, because in the first place, the maximum resolution can only be achieved with near perfect technique, and will require the use of a very heavy tripod to eliminate as much movement as possible, something most photographers seem reluctant to do, along with a cable release and a system for locking up the camera mirror—a facility found on very few digital slrs. Of course the need to use such equipment and technique is the antithesis of the “quick, point and shoot” philosophy behind digital, at least in the 35mm equivalent form.
Another is that sensor and film size is hugely important in itself; it doesn’t matter if a half-frame sensor could be made to yield as many pixels of information as medium format film, since when the enlargements are made the inevitable errors of movement and optics will be several times more obvious in the images from the digital sensor. Clearly full-frame digital sensors can be compared directly to 35mm in this regard, but film still has several times more potential ultimate resolution than any full-frame digital sensor currently available. This applies even more to medium format, where, frankly, the solutions available now are simply outrageously expensive curios and (no doubt) status symbols for those daft enough to buy them.
Although amateurs usually equate ‘film’ with ‘35mm’ there are plenty of perfectionist pros who have long held the view that 35mm was borderline at best. We would only use it when we could get away with it; as soon as the demands went up, we got out bigger cameras, that took bigger film. The above shows that the reasons for this are still as true today: size matters.
Now let’s deal with ultimate resolution, even though as we have seen, this is available only to a very few photographers.
A few stats here first; the old Nikon film scanners, which were for years an industry standard, scanned at 2700 pixels per inch (ppi.) This would yield at maximum a 9-megapixel image from a 35mm frame, so a 12 Mp digital camera has it well whacked. However, modern desktop film scanners scan up to 6400 ppi and a drum scanner is good for double that or better. The desktop can therefore yield 60 megapixels worth of info from a 35mm image and the drum over 120. Both clearly leave anything digital for dead—but they also leave most 35mm film the same way, because even under completely ideal conditions, which, as we have seen, are extremely hard to achieve, it just can’t record that much detail, and all you get is meticulously rendered grain. However, the old Kodachrome II 25 transparency film could yield approximately 8,000 ppi of real information, or 96 megapixels for a full frame 35mm scan; not only leaving any 35mm ‘equivalent’ digital sensor waving its legs in the air, but the stratospherically expensive medium-format equivalent ones as well; and of course a 6x6cm K2 tranny could give you 320 Mp and a 5x4inch one—wait for it—1.28 gigapixels, all extractable using our drum scanner. Digital, eat your heart out. Pity Kodak stopped making it, really.
However, the question is not really “How much detail can the film or sensor record?” but “How much clear detail can the photographer get on it?” And as we have seen, there are practical limitations to this that affect both film and digital, but which will be more obvious for digital with its smaller sensor size.
Consider the debate about quality in the pro context; the required quality is that which satisfies the client at a price he will be content to pay; and for a great deal of professional work, that used to mean 35mm. But there was always work that needed that extra poop and for times like these, we had medium and large-format cameras or we went with a specialist film like K2 or Velvia, or both. But in an era when most GD’s and other commissioners want their images on disk, the demand for such things has largely dried up. We pros follow the market; why would we expose film, process it and then scan it, when a digital camera does the job up to the standard our clients require? This is not about absolute quality; it is about acceptable, saleable quality. Photography is our business. We don’t nancy around.
And, for us, there’s another question to be answered; what are the pictures for? If they are to be reproduced in a newspaper, use digital. If they are to be reproduced in a glossy magazine, go digital. If your end, however, is to make prints, it’s not so clear-cut. And all my comments regarding the positive side of digital are only in respect of colour; for Fine-Art quality monochrome, digital is totally outclassed by film. Serious photographers know that a great monochrome image starts with a great negative and the ones who think you can just greyscale a colour image don’t know the difference anyway.
Grain and graininess
Perhaps I should take a moment to clarify the above. There are many factors which explain why, for serious monochrome work, digital is not worth considering. Of these two are most important. The first is grain, or more properly graininess. Grain is what we call the clumps that the exposed and developed metallic silver forms within the emulsion of an image, and graininess is the extent to which this is obvious. Graininess begins to be a problem in 35mm colour film at speeds of 200 iso and above. Remember that there are three or four light-sensitive layers in a colour emulsion, each exhibiting graininess; and graininess in colour film is not something I have ever found particularly appealing. What we are seeing here is not the metallic clumps of silver themselves, but the rather amorphous blobs of dye which are left when the silver is bleached out of the emulsion during processing. One of the big early scoring points of digital capture, as we have seen, was the removal of the excessive graininess associated with colour negative film, especially at higher Exposure Indices. This made it the darling of Press photographers even when the cameras could only just manage a couple of megapixels of resolution.
In monochrome, however, graininess is caused by the actual clumps of silver; in a well-developed, good quality emulsion the effect of these is not at all unattractive. These silver clumps form clear, distinct shapes, not the ugly fuzzy blobs we find in colour films; in fact, rather like the dotted effect of a mezzotint or an aquatint, graininess in a monochrome image is integral and can be very pleasing. And of course, when we don’t want the graininess, we have the simple expedients of picking up a bigger camera or using a slower film.
The second issue is to do with the characteristic curve of film. Any film emulsion, being analogue, exhibits a non-linear response to light. If this is plotted on a graph it typically forms a sinuous s-shape with a linear section in the middle and flattened or compressed values at the extremes. This equates very closely to how human eyes, being analogue, actually see; and as well as this, rather small differences in the characteristic curves of different emulsions provide very different cumulative effects. Ilford HP5 Plus, for example, renders white skin tones in an exquisitely appealing manner, while Kodak T-Max 100 is definitely my favourite for mechanical or engineering subjects. Furthermore, as we reach the extremities of over or under-exposure, the flattening curve compresses the values so that we can record many more levels of light intensity, rather than blowing off the scale into solid white or black. Film, in other words, when crashing into the buffers of its performance limits, does so gracefully, even beautifully.
Now, of course we could footer about on our computers with the curves of a digital capture and we can use Photoshop to add noise that will form a facsimile of film graininess, to achieve a resemblance to a film image, but consider what we would be doing: using digital to imitate what a perfectly well-understood, reliable technology does naturally. It doesn’t come much more ersatz, and to be frank, downright naff, than that. I mean we can use programs that will make photographs look something like paintings or drawings or etchings, but does anyone take that seriously? It’s right down there with the horrible posters of big-eyed girls and soppy dogs or those digital watches some people wore in the 80s—terminally crap. Trying to make a digital capture look like real film is just as tasteless and reveals an absolute lack of artistic values.
The photography business is not run to service the needs of the very few professional photographers in the world, but the legions of amateur happy-snappers; it is the regular sale of equipment to enthusiasts who ‘upgrade’ every year that keeps the manufacturers in profit, not the infrequent sales to pros who work their cameras till they are beyond servicing. The camera trade’s interest in we few professionals is largely as a marketing device; the sight of myriad Press photogs gazing down their Canon or Nikon Long Toms at sports events sells a huge amount of lesser, and more profitable, equipment to amateurs all over the world. Digital is a wondrous fillip for the camera trade, at once bringing the amateur his holiday snaps straight to his computer desktop and at the same time giving it something new to flog him every year. And within most photographers there is a love of equipment for its own sake, no more so than amongst amateurs, which brings a gleam of joy to the trade’s eyes.
This is the real reason why Zenza Bronica no longer make cameras. Bronica had for decades established itself as the ‘next step up’ from 35mm for amateurs who either wanted to get serious or to impress their chums. The cameras were medium format, well made, practical, handled well and had great optics yet a price tag much less than a Hasselblad or a Rollei. Bronicas were also popular and capable pro cameras, but it was not from sales to us that they made their profits; so when the amateurs went digital, we lost Bronica.
But enough of that; why am I writing this,
when, surely, the battle is over and film has lost, consigned as a relic to the museums of history in a digital age?
Well, for those amateurs unburdened by artistic consciousness or professional standards, the decision is simple; after all, very few ever had their colour negs made into anything more than 6×4 prints at a high-street shop. Some may have had larger prints made and a relatively small number of serious ones had discovered the benefits of transparency film or dabbled in the arcane art of monochrome printing; but for most, the best place to view their images is on a computer, and digital is the answer to their prayers. Indeed, for almost all amateur and most professional photographic purposes, digital has proved its worth beyond dispute.
For professionals whose work is in colour the decision is a no-brainer; 35mm colour negative film has been replaced by the 35mm equivalent digital camera. There is no point in fannying about here: this is a done deal. Colour negative film was always the weakest link in film photography, especially when it had to be scanned, which itself introduces a new layer of changes to the image. However, for those occasions when only the best will do, a colour negative from a big camera is still better.
For colour work where transparency film was demanded, which was already a rapidly declining market before the advent of digital, you picks your choose. Ultimately, the best transparency emulsions are still better than digital; practically, digital has carried the day.
Where film remains king is in monochrome. Digital simply cannot match the subtlety of the silver image, and no matter how good a digital print is, no matter what digital trickery or manipulation is made to the image, it doesn’t come close to the quality that a fine monochrome print has simply because it is what it is. And while this is not the place to discuss in depth the business of monochrome printing, let me assure you that if the above is true for the silver gelatin process which most people are aware of, and it is, then it is even more true for Platinotypes, Palladiotypes, Kallitypes, Salt Prints, Gum Prints and a host of other, older processes.
So perhaps you will not be surprised to learn that last month I opened up my darkroom again. Most of my work in there will be in monochrome and my DeVere enlargers have been cleaned and serviced and put into harness once again.
I’m not giving up digital; far from it. I will continue to take as much profitable digital work as I can get. But now I will also be actively seeking work where I can use film, and the skills I have developed with it, particularly in monochrome. And I have rediscovered the pleasure and indeed the importance of photography, which, for me, is quintessentially a printmaking art.
For the first time in years
I have gone to bed happily smelling the sulphite on my fingertips and remembered the pleasure of seeing prints emerge in the developer like some almost forgotten black magic; I have revelled again in the velvet comfort of the true obscurity found in a photographer’s darkroom and had my mind soothed by the warm lustrous glow of an Ilford 902 safelight.
Again I have found myself wrestling with the controls of my Deardorff 10×8 under a flapping dark-cloth. I have looked in wonder at the exquisite rendition of skin tones that HP5 can achieve and I have rediscovered how pleasant a Bronica ETRS is to handle. Once again my trusty MPP MkVIII has trudged a deserted beach with me and I have burst out in a smile at the long-lost pleasure of seeing an image shining through the pentaprism of a Pentax 67.
And when I look at my prints I finally realise—The simple truth is that although digital is fine for the professional work I do, in the final analysis it is just not as good as film. It is certainly sharp enough to replace 35mm for all editorial and most advertising work, and it is much faster and more convenient. Yet I am at my most demanding in the work I do for myself and digital quite simply isn’t good enough for me.
And this is something that can’t be corrected, can never be changed by technical development, is not another hurdle, like resolution, that will eventually be jumped by digital; no, for in the end the reason film is better, in the absolute terms that the artist, rather then the professional in me demands, has nothing to do with resolution or ease of use or speed; it has nothing to do with pixels per inch or lines per millimetre—film is better than digital simply because it is analogue.
The world I live in—the houses, the trees, the people—are all analogue. My mind is analogue, not digital, and so is my artist’s eye. I am analogue. In the end, when the cards are counted, digital is a caricature of the world, albeit one good enough for most purposes perhaps—but analogue is real, film is a true reflection of the world, and no amount of computerised smoke and mirrors can change that.
Put it another way: film has soul—and you either get it or you don’t.
This article was originally written in 2008 and some of the details may be outdated. However the general conclusions remain valid. Rod Fleming, 2013.
Copyright © Rod Fleming 2008