Musings on Photography

Just last week, ordered up some more Cranes Museo Portfolio paper. It’s a paper I like a lot - nice base color, nice surface finish, and it has (for a non-glossy paper) a good color gamut on my HP Z3100. It’s on the expensive side but it’s awfully nice. What I’ve been doing is making ‘work’ prints on a cheap matte paper (it used to be Epson Enhanced Matte, now it’s InkjetArt Premium DuoBrite SUper Duper Something or Other).

I like the matte surface papers. I like the way they look, especially for monochrome prints, and I like the way they feel. And when you put monochrome prints behind glazing, you don’t get nearly as much difference in perceived dMax when compared to glossy papers are you do when you compare naked prints.

But lately I’ve noticed that a lot of the photos I’ve been making have a lot of subtle color things going on with the darkest tones, and when I go to print those images and do a soft proof, huge swaths of the print are out of gamut on the matte paper. Therein lies a tale.

The original problem with inkjet printing on matte papers was that it was impossible to get a really dark black on the matte paper. Lots of very clever people went off and put head sized dents in not very well padded walls, and did cunning chemical things to the paper coatings and the inks and the printers, and voila!, we got matte papers and matte black inksets and all of a sudden we got matte finish papers that can give us something resembling decent blacks.

But, although we all thought that matte papers with decent dMax would be the paper of our dreams (I certainly did), what I’m discovering is that this isn’t actually true. No, let me qualify that. If you’re doing black and white, or toned B&W, or for some other reason (like you only photograph in White Sands NM) all your work has a very small color gamut that never strays too far away from the luminance axis, these papers are the bee’s knees and the cat’s pajamas neatly packaged in one tidy box and tied neatly with a bow.

But for the rest of us, these papers are something of a disappointment, and it’s taken me a startlingly long time to clue in to why they disappoint me particulary. The problem seems to be this: while these papers offer a decent dMax, they don’t offer much volume to the achievable gamut in those lowest tones. If you want a very dark saturated blue, or a dark saturated green, for instance, you’re pretty much out of luck. The paper just won’t go there, sorry. So you make a print, and all of that dark stuff falls out of gamut, and you’re at the mercy of the way the profile maps these out of gamut colors back onto the achievable colors of the paper. In general, this means that all those low tones are rendered as muddy mush.

It turns out that glossy (or at least non-matte) papers just whip the bejeebers out of matte papers on this ‘volume of gamut in the low tones’ thing. Why it’s taken me so long to awaken from the trance and realize that perhaps I need to be printing these images with lots of dark tones on a paper with a bigger low tone gamut, I don’t know. Not enough tea, perhaps, or some impairment related to Seasonal Affective Disorder. But I do believe that the answer to my problems lies in giving these papers a shot.

Fortunately for me, the crop of offerings for gloss papers has now expanded away from the RC paper like offerings of just a little while ago. Most manufacturers are now offering papers that are structurally (and thus visually) very much like the famous gelatin silver air-dried glossy paper I once sneered at. That sneerage has decreased in volume as I’ve come to realize that the different gamut might have profound implications for color work.

So at the same time I ordered up the Crane Museo Portfolio, I also ordered up some Harman Glossy FB AI, and some Ilford Galerie Gold Fibre Silk, and we’ll see what the low value gamut is like and how the prints look.

Mr. UPS assures me the paper will be in my hands on Wednesday.

And, as a general question, one of the things I notice is that information about papers is scattered all over the place and it’s very hard to find places with info more helpful than “oooh, I love it”. It would be nice to have a clearinghouse that actually shows achievable gamut on various papers on different printers, etc. I am of half a mind to order up a whole slew of different papers, profile them on the z3100, and put descriptions of the surface finish, base color, and actual renderings of the gamut up so that people can gaze at them in mute wonder before plocking down $150 for a roll of paper they’ve never tried before. And I wonder what sort of contribution people would make to make that happen - would it be enough to cover the cost of the paper and some moderate compensation for the time? Would paper manufacturers offer up samples for free?

I spent this morning framing up two prints to go off to the Poncho Art Auction. It was a comedy of errors, in sort of a Keystone Kops sort of way.

The first moment of desperation came when I realized that I had no 22″x28″ plexi on hand. I resorted to my usual tactic - running around the house gazing at all the framed work the right size, and picking two victims to be unframed so that I could use the plexi glazing. Then came the scrabbling around under the studio stairs, looking for two 22×28 frames. No, we don’t want to use the ones we scavenged the plexi from, thanks, I’m trying to hang on to those.

Then came cutting new mats. For the first time in I can’t remember, I managed to waste a sheet of mat board by making a new mistake. Usually I just repeat one of my old mistakes, so this was a bit out of the ordinary. This time, I managed to ruin a sheet with the very first cut, by leaving the old center scrap under the new sheet of mat board, so that the new sheet buckled in a funny way and the cut was all wavy. Extra time wasted while I invented a new word and used it repeatedly. It’s not a nice word.

Once the mats were cut, one of the prints steadfastly refused to lay flat. I don’t know how a sheet of paper cut from the middle of a 50 foot roll and then flattened in a deroller can be anything other than planar. This print had this weird wave in one edge. A repeat visit to the deroller eliminated the wave.

I’ve given up on using the plastic corner thingies. They’re just too frustrating. Since these prints are going away and never coming back, I mounted the prints using T hinges made from tyvek tape. Much faster and simpler than those damn corners, I think.

After all that, actually shepherding the prints into the frames and buttoning everything up was simple. The only remaining hitch was that when I went to the old Windows machine to print labels for the back of the prints (the label printer is still attached to that machine) I discovered that the keyboard was dead. It’s a wireless keyboard, and eventually I figured out the batteries were dead. So I popped open the battery compartment, and was greeted by an alkaline battery that was oozing white stuff. Battery gore, I guess. After I cleaned out the compartment and put in new batteries, the keyboard worked again, and a few minutes later those labels were on the prints. Sheesh.

I always think I have this process all figured out. Somehow it always seems to evade my preparations. At least I didn’t draw blood this time.

My recent experiences with slideshows on the Mac got me thinking. The ongoing discussion about the resolution/Pixel density of the Nikon D3, along with the news about the pixel density of the newly announced Apple iPod - that got me thinking even more.

The pixel density for displays, back in the good old days, was 72 pixels/inch. Oh, if you had a really high resolution display adaptor and a really good CRT display, you could get it higher, but 72ppi was pretty much the norm. Nowadays, LCD displays seem to have a pixel density more like 100 ppi. That’s the density on the MacBook Pro I’m typing this on, for instance.

The pixel density on the recently revamped iPod Nano is, according to Apple, the highest pixel density they’ve ever shipped in a product. It’s apparently a QVGA (320×240) 2″ display, which works out to about 200 pixels per inch. Call it double the resolution of the screen I’m looking at. And the display of the newly announced Nikon D3 is 640×480 (after we sort out all the nonsense stuff about the difference between dots and pixels) in a 3″ display, which works out to 240 ppi.

Naturally, as the pixel density increases, the maximum resolution you can display rises as well. One of the main arguments against viewing photos on a computer display has always been that the resolution of the computer display was hopelessly inadequate. At 72ppi, that’s pretty much true. You look at a 72ppi screen, and the phrase ‘high resolution’ doesn’t exactly spring to mind with the speed of summer lightning. At 100 ppi, things are looking better. At 200 ppi, things are looking good; at 240 ppi, they’re really looking good. At 300 or 360 ppi, we start to encounter arguments that adding resolution won’t help much because at normal viewing distances our unaided eye can no longer see the difference.

I will here wave my hands about and make funny faces and strange noises to distract you, and while you are thus distracted ignore the fact that similar claims of “so good it can’t get better” have been made before, and then disproven (think digital audio). The point is not that some fixed pixel density is sufficient for all needs. The point is that lately, the higher resolution is becoming available, the costs are falling, and the computing power needed to drive displays with much higher resolution is everywhere. There’s a world of difference between fabricating a 3″ diagonal 240 ppi display, and fabricating a 30″ 240 ppi display. But I expect that in the end, I’ll be able to drive to the Apple store and buy a replacement for my 30″ Cinema HD that instead of being 100ppi will be more like 300ppi. Not next week, but within my actuarial lifetime as a productive photographer.

Now, viewing a photograph on a display and viewing a print are not the same. The display emits light, the print reflects it. So the display is (ignoring second order effects) more or less independent of ambient lighting, and to look really good, a print needs to be generously lit. (Here’s the formula for adjusting the lighting to the optimum intensity for displaying prints: hang the print on the wall. Increase the brightness of the lights shining on the print until the print starts to smoke. Back the lights off until the print no longer smokes. Stop.)

And, although the furor over dMax of inks, papers, and printing technology seems to have died down of late, it’s still important. A good printing technology will give you a usable dMax of, say, somewhere between 1.8 logD and 2.4 logD. That corresponds to between 6 and 8 stops between dMax and dMin. But the display I’m looking at right now has a contrast ratio of 1000:1 - 10 stops between the darkest and lightest. 10 stops. Ten stops. I’m not going to claim that a wider range is the holy grail of photo display, but I’m pretty sure that 2 to four stop difference is part of why I look at stuff on my screen and think “Holy Cats, that looks good” and then look at the print and think “Um, not so much.”

Color gamut is another story. There are colors I can print but can’t display on my LCD monitors. There are colors I can display on the monitor but can’t print - interestingly, they seem to be mostly very light colors and very dark ones, and those are often the colors I struggle with when printing. All told I’d say I’d call the color gamut issue a win for the monitor, but the fact is that they are mostly *different* and not better or worse.

And so my question is this: when displays offer similar size and resolution to the prints we make on our printers, and the displays offer better options for color, better dynamic range, and so on, what properties of the print will remain that will mean that photographers continue to make prints? Will prints become a thing of the past, or will the object properties of print (the surface finish, the weight and hand of the paper, no need for a power source) mean that despite their limitations, prints are still what we think of as the natural end point of the photographic process?

Ed Richards asks

How do you feel about prints as objects? Things to feel and touch, the effect of different papers?

I like prints.  I like big prints, and if I had infinite wall space, I’d probably make a lot more big prints than I do.  But I also like medium prints, and most of the prints I make are 10"x15" these days.  That’s big enough to be able to see what’s going on with the photo, but small enough that you can fit them into a nice portable box or portfolio case.  I like small prints, too, although I don’t seem to have the same reverence for the small print that the ‘quiet photography’ folks do.

But in comparision to viewing photos on a screen, or viewing photos projected (either slides or a projection monitor) I think prints beat the screen all hollow.  I like holding the print in my hands, I like the fact that it’s a real, physical object with its own concrete existence.  I understand completely that there are photographers for whom it doesn’t matter if the process doesn’t end in a physical object, but for me, the process doesn’t feel complete until I’ve got a real print in my hands.

In other words, my views about the reproducibility, uniqueness, and value of the print have changed with my switch to digital printing.  But the value of actual prints in my process, and the satisfaction I get from making prints is still the same.

One of the things I really like about digital printing is that I can stick a wide variety of papers into my printer, and get results that are different.  In some cases, the differences directly affect the qualities of the print (dMax, for instance).  But in most cases when I choose one paper over another, it’s a matter of what I think of as the physical properties of the paper - the way it feels in your hand, the flexibility and weight and thickness, the surface texture, and so on. 

That’s one of the reasons why I’d like to own an HP z3100 - the built in profiling would allow me access to far more papers than my current Epson 9600 does.  And the built in profiling would let me evaluate papers with a lot less effort (and a smaller quantity of paper) than I currently can.  If there’s a reason I haven’t yet bought a z3100, it’s that I know that the instant I have it up and running, I’ll launch into a frenzy of paper evaluation.

I know that there are folks who feel that digital photography and on screen viewing of photos signals the death of the print.  But my view is that eventually, photographers who don’t print their work will find the on-screen experience disappointing, and they’ll turn to printing.

And I think the current tidal wave of new papers, inks, printers, etc. are the beginning of a great golden age of the photographic print.

One more random thought to throw into the discussion of the (assumed) great longevity of gelatin silver prints.

One of the big issues surrounding inkjet printing has been the use of papers which have optical brightening agents (aka OBAs) in them.  There are huge, longwinded discussions about the problems with OBAs, and I’m not even going to try to recap the discussion here.  But there are apparently serious concerns about the longevity of any print made on paper that has OBAs, because the OBAs lose activity/evaporate/fade over time, and so the print ‘yellows’ as the paper base loses whiteness (and perhaps other bad things happen).

Prior to 1950, no commercially available gelatin silver paper had optical brighteners.  After 1966, most did.  My understanding is that virtually all currently available gelatin silver papers have optical brighteners.

Just something to keep in mind when you’re pondering whether the longevity of prints made 100 years ago has much to do with the longevity of prints made today.

One of the things that’s been most interesting about my friend Rob getting a new HP Z3100 has been that as a result we’ve spent quite a bit of time looking at/comparing profiles, and thus we’ve started to get a grip on just what profiles tell you about what you can expect to see if you make a print on one material or another.

This got me thinking about an easy way to compare one technology (say, inkjet printing) to another technology (say, RA-4 papers).  It seemed to me that if we just had profiles describing the characteristics of an RA-4 paper (like, for instance, Fuji Crystal Archive), then we’d just load it up along with the profile for, say Crane Museo Max on the Z3100, compare the two, and learn a lot.  It turns out that virtually any lab that does custom prints on a LightJet or Lamda provides such profiles to its customers so that they can tweak out their images just so before sending them to be printed.  And those profiles are exactly what the doctor ordered.

The first order observation is that, in the low tones, Fuji Crystal Archive has a much wider gamut.  If you’re printing color, and you want lots of color separation in the shadows, Fuji Crystal Archive will give you better results than Crane Museo Max on an HP Z3100 (and Crane Museo Max was the best in the lowlights of the non-barrier papers we tested).  Conversely, Crane Museo Max on the Z3100 has a broader gamut in the highlights - so if color separation in the highlights is what your prints are about, Crane Museo Max on the Z3100 will give you better results.

Mike comments:

I’ve got prints on my walls that are over 20 years old. They still look as good as the day they came out of the wash. And as far as I can tell no one’s proven that this inkjet stuff will last that long.

Because of my heavy focus on enjoying the process, and the fact that I view prints as being of relatively little importance, I’m probably not as worried about print longevity as some folks are.  I confess that I don’t understand the extremes of passion that are directed toward archival processing, etc. when it’s manifestly true that for all but a handful of photographers working today, the eventual fate of all of our physical prints will be that our grandchildren will shuffle through them, think “Old Grandpa sure was a weird guy”, and tip the entire thing into the trash.  If every print made lasted forever, then we’d eventually be neck deep in prints, folks.  Not everything important needs to last forever.

But one thing I don’t understand is this: the implicit assumption seems to be that, since there are gelatin silver prints around that are 20 years old, (or 60 years old, or 100 years old) we can expect that ALL gelatin silver prints will last that long.  That’s rubbish.  We have gelatin silver prints that have been around for a long time because there were so very many made, and they’ve been processed in a staggeringly wide variety of ways, and then stored in an even wider variety of ways.  And a very small minority of them have survived.  Folks who crow about the demonstrated longevity of gelatin silver prints are conveniently ignoring the millions and millions of gelatin silver prints which have NOT lasted.

Now, I understand that the argument goes “Yes, some GSPs have not lasted very long.  But we now have Archival Processing, and so OUR prints will last forever.”  The problem with this argument is that exactly what Archival Processing is has changed over time.  First, it was two bath fixing.  Then it was the Ilford quick fix process.  Everyone thought that selenium toning would guarantee prints that lasted forever, and then Kodak changed the formulation of Rapid Selenium Toner and eliminated the (previously considered unimportant) trace ingredient that was providing the protection, so lots of prints that everyone thought were archivally processed were, in fact, not.

And, to be honest, I think the whole ‘archivally processed’ thing is more marketing hype than anything else.  A while back, I decided to take a survey of exactly what this much ballyhooed ‘archival process’ was, so I sent email to wide swath of photographers who claimed that the prints they sold were ‘archivally processed’ on their web sites.  About half of those who responded said that they just followed the archival standard, but could not point me to a copy of the standard.  About a quarter told me that they used the old, now discredited minimal selenium toning technique.  And about a quarter said that they’d done residual fixer testing, and anything that passed was by definition archival.  I didn’t consider all this to be very promising.  Face it, there is no ‘archival standard’.  Folks who claim that their processing meets archival standards are, as far as I can tell, blowing smoke and asking you to watch the pretty mirrors.

But the killer is this: because of longevity concerns, resin coated gelatin silver papers have undergone heavy accelerated aging testing.  We have some sense of how long they’ll last.  Likewise the various color processes.  Inkjet printing, too.  But I can find NO such testing for conventional, fiber based gelatin silver papers.  Apparently, it doesn’t exist.  And, to make matters worse, we can’t even extrapolate from the performance of previous years papers.  NONE of the currently marketed traditional gelatin silver papers are the same as they were just a decade ago.  Emulsion formulations have changed repeatedly.  Suppliers have changed, and so the trace impurities (which often have big effects on longevity) have also changed.  So if you’re currently printing on gelatin silver papers, even if you’re using what you believe to be archival processing techniques, you’ve simply got no testing at all to back up your beliefs about the longevity of your prints.  None.  (hint - almost all of the research on things like using toners for permanence were done to preserve (wait for it…) microfilm records, not continuous tone photographs)

At least with inkjet prints, you can fall back on the accelerated aging and fade tests.  They may not be much but they’re better than nothing.  I don’t trust the Wilhelm accelerated aging tests farther than I can throw them (which is just about to the end of the decade) but I trust them more than nothing at all.

And a little perspective is helpful, too, when you’re deciding how much energy and worry to invest in the longevity thing.  At a workshop, John Sexton told a story about how his earliest photographs were of drag racing.  And, he said, they weren’t really all that good.  The good news is that when he made them he knew next to nothing about proper processing, so most of his embarassing work is now gone.  From this I take the lesson that worrying about print longevity is, for me, putting the cart before the horse.  When I can make photographs that are good enough to make them worth preserving forever, I’ll start worrying about making them last forever.  Until then, I’ll work on the art part.