« RIP #2 (Not OT) | Main | Panasonic Dual I.S. »

Thursday, 10 March 2016


I started shooting 4x5 in the 1950's. 8x10 came along in the 1990's- Digital came along in the early 2000's. Fact: it is a hell of a lot more difficult to make a really good print from film than digital, no matter what size either is.
I shoot digital because I am old and don't have the energy. My darkroom building is now converted to a print/frame operation.
I sold my 8x10 enlarger a few years ago, got a lot of floor space back.

When you push toward the edges of dynamic range or lens rendering, medium format film will most definitely win. I just watched the documentary "Finding Vivian Maier" and the magic of her medium distance Roleiflex portraits is astonishing.

So, is it fair to conclude that you choose convenience over quality?

The bridge and moon photo from the earlier post echos back to Michael Reichmann's infamous test of the Canon D30 vs slide film. The D30, at 3.1 glorious megapixels, was "better" than 35mm slide film. A later test showed that the D60 was "better" than medium format.

The important thing to notice about Reichmann's comparison skyline shot as well as Ctein's bridge photo, is that both photos are of something scenic very far away. The conclusion of digital beating out its film counterpart was judged at infinity on very flat, technical grounds.

Take a look at Vivian Maier's medium format street portraits. There is just a... shape to them. And the tonality is amazing. There is just something organic and real to film, especially medium format, that frequently does not come out to play in digital, let alone small sensor digital. Why is that famous portrait of Lincoln so 3D? It's an 8x10 portrait. try to get that look on a GH4.

As for a color shot of Yosemite Valley? I happily shoot digital and the print turns out great.

So which is better? "Depends what you mean"

"Image quality" is a phrase that is fraught philosophical difficulty because, as you say, you can't pin people down by what specific parameters they are looking for. And, even if you could no one would be able to agree on which parameters are the important ones.

You might say that nothing in the digital world looks "as good" as 120 Tri-X shot with a nice lens and then printed on fiber.

I might reply that what you really mean is nothing in the digital world looks "the same" as 120 Tri-X shot with a nice lens and printed on fiber.

Both, I think, are true statements, depending on where you are coming from.

Other people might be more worried about edge sharpness, or low light noise, or color saturation. There is really no way to have this discussion except in personal generalities.

So here is my personal generality.

There is a restaurant in Pittsburgh that I go to a lot, and have gone to a lot since the early times for digital cameras. The place is pretty dark, and it used to be really hard to get a good shot that wasn't just all blotchy noise and noise reduction. But these days, with the iPhone, it's dead simple You just point and shoot and you get a decent picture. I'm pretty sure that the JPEGs out of the phone look better than even the last Nikon DSLR I brought in there (D200). I'm pretty sure that it is better than any color film you might have brought in there if you were not allowed to set up a tripod.

It's easy to find other stories like this. You can take hand held pictures with a phone these days that would have been impossible to get without a tripod before... and they mostly look great. As good as film. Maybe better? Hard to say.

So my personal generality is that the best phone cameras are now about as good, or better, depending, than most 35mm color film in a point and shoots used to be. Assuming that technical 'quality' scales by sensor size, I would not find it surprising if you can make m4/3rds pictures that are as good or better, whatever that means, than 6x6.

But ultimately this is an empty exercise. If you want pictures that look like 120 Tri-X shot with a nice lens and printed on fiber paper, there is really only one way to get them.

But that also does not mean that the digital stuff is "not as good". IMHO.

It all depends on what you want.

"... and counted the bricks on a distant building."

Oh my, is it just me or is anyone else thinking this will suddenly become a thing over at that DPR websight all of a sudden? Thanks Mike ;)

And it is also said, 'Go not to the elves for counsel, for they will say both no and yes.

JRR Tolkien

For me, IQ is mostly comprised of four things: sharpness(as in spatial resolution); low distortion; accuracy(correct representation of colors and/or shades of gray); and range (as in stops of gray scale). The significance of each is image dependent. Sharpness is usually less of an issue in an atmospheric foggy day image than in a macro of a complex mechanism. Color accuracy isn't involved in a B&W image, but range may be. Distortion is rarely desirable, and if I want it I can add it in postprocessing. But these measures are more about the ability of the camera and lens to perform. When looking at the image, how they relate is more about the overall impact on the subject. And that is what usually defines a good photo.

"for my work and my taste . . . I'd pick 120 Tri-X 400 in D-76 1:1 in a 6x6-cm camera with a beautiful lens, printed on gorgeous fiber paper, every single time."

So, you disagree when Ctein says "Image quality is a multidimensional thing . . . the aggregate image quality of Micro 4/3 today, in film terms, falls midway between 6x7 medium format and 4x5-inch large format."

I'm trying to understand what makes medium format film and camera better for your work and taste.

I think you were saying (in the entire article) that digital has a larger "sweet spot" because it has fewer tight limits, compared to film.

Am I on the right track?

"... have you ever heard of a medium-format film camera that offers IS? I'm not aware of one."

I can reliably hold a 3.5F Rollei at 1/2 second -- due to its Mass and Ergonomic design. No "artificial IS" needed.

People ask me why I still prefer to shoot film when digital gives you near instant results. My reply is usually, " Who says art has to be efficient?"

I've seen a lot of medium format pictures in my life, and some of them certainly had something special - cannot be replicated digitally, not even with the so-called "full frame" sensors.

But what would interest me: you're using Fuji, and you once wrote it's because of the beauty of their B&W outputs. So this comes closest to Tri-X? Or is it some Nik plugin, or can't we "emulate" that film look until today?

I wonder how the new Pen-F would compare to the Fujis, and to Tri-X. Somewhere I read that it overlays "real" (photographed) grain. I can test the .orf files from sources like DPReview with the OV3 raw converter, but I don't have any comparison to Fuji's cameras...

Well said, Mike. I think many, including me, could learn a lot by including the *whole* work experience as part of "quality" of anything.

BTW, thanks for saying that you simple don't care about results in really big prints, because you don't ever make prints that big. I think I could have saved some money as well as heart-ache if I'd realized such an attitude much earlier.
(I've been working on it for years, but I'm *still* in the process of unlearning that ultimate sharpness is only as important as you make it.)

I wholeheartedly agree with the total experience measure. It is that Image Quality that impels the photographer to make more images and the viewer to appreciate them.

Thanks for illustrating the subjectivity of image quality. If you hadn't already stated as much, when I read `The best digital B&W I've ever made does begin to approach the IQ of medium-format B&W film expertly crafted. Emphasis on "begins to approach."' my gut reaction was to ask `so what about the best digital B&W anyone *else* has made?', which is perhaps rather cheeky.

Still, I share much of the same ill-suitedness to MF. Working on the principle of the format being as good as what you can get by walking into a certain Highland camera shop and buy for an amount of money before the plastic starts to complain, MF gave me the *least* depth of field in landscape work - sharing the 50mm-e normal focal length with equivalents in multiple formats, compact and dSLR could achieve nose-to-infinity before encountering diffraction; LF has movements to work around the narrow focal wedge; stuck in the middle, MF has you griping about needing f/22 and reciprocity-failure at 4am. No thanks.

I'm back in m4/3rds now and finally going to grow up and call it my home format.

[0] *there's* the downside of MFT. I have no clue what model designation means what spec even within Panasonic, let alone Olympus. The only way to buy something is to choose the newest and most expensive. Oops.

By nature of the smaller sensor you can use a shutter speed three or four stops faster for the same depth of field with Micro 4/3 compared to 6x7 film. So right there you have a huge advantage as far as handheld sharpness in marginal light and for moving subjects. Add in IS and you gain even more for static subjects.

As far as resolution goes my experience is that medium format (6x7) film yields more resolution when shot in the "center" (compared to my 16MP Olympus). That's either darkroom printed or scanned. It's visible in a 16x20 print, and becomes very obvious when you print larger than that. But printing larger than 20x24 in my darkroom is very difficult, so it's not much of an advantage.

For me the reason Micro 4/3 wins out over medium format (and even FF digital) is the efficiency, and sufficiency. I don't need huge prints of most subjects, and being able to make many very good images is sometimes much more important than being able to make one big and perfect one. The smaller camera size and less need for a tripod also helps a lot.

Like you I prefer the look of fiber based darkroom prints.

"So, is it fair to conclude that you choose convenience over quality?"

That's kind of like asking a father if he chooses money over family because he works 40 hours a week.

I presume Mike's answer would be that he chooses a compromise between convenience and quality that suits his needs; not one over the other. And furthermore, the point of the article is that, depending on your needs, convenience factors might be significant contributors to quality.

Dear Mike,

Very well said!

If I may add to it, both confounding and clarifying… [grin]…

Confounding the discussion is the problem that “medium format” is very ill-constrained. It covers everything from 645 to 6x9 formats, a bit more than a factor of two in film consumption.

(Normally I swat down folks who erroneously compare film and sensor areas, because it is not a correct metric of image quality, but the proportions are so different between the different formats that I hope I shall be permitted to do so for what is a very superficial and qualitative comparison.)

That's almost 3/4 of the “distance” between 35mm and 645 formats, and it's roughly the distance between 6 x 9 cm and 4 x 5 inch formats.

That is, it covers a hell of a lot of image-quality territory and is further confounded by usage (e.g., folks who liked 6 x 6 format primarily because it let them crop to a vertical or horizontal rectangle without having to reorient the camera).

This confounds the discussion.

On the clarifying side, though… Print size doesn't much enter into it once you get above 17 x 22. Up to that size, folks tend to pick prints up in their hands (if they're allowed to do so) and look at them from what we would consider a normal close viewing distance. They hold big prints a little further away than small prints but not proportionately so.

Above that size, people tend not to pick up prints, even if they are allowed to do so, and reflexively (and almost automatically as a result of body posture) stand back from them. The bigger the print, the further back they are inclined to stand. It's not fully proportional to print size, so “bigger” really is bigger, but it's only a small fraction of the scaling of the physical dimensions.

I'm talking about the typical. There are, of course, exceptions both ways. This, though, is what one can usually expect of a viewing audience. (Subject matter and photographer's aesthetic intent also, obviously, affect this.)

That is, discussion of print size as a factor is rather overblown once you hit the “16 x 20” point.

pax \ Ctein
[ Please excuse any word-salad. MacSpeech in training! ]
-- Ctein's Online Gallery http://ctein.com
-- Digital Restorations http://photo-repair.com

Dear Robert,

I think that would be entirely fair, so long as one says the same thing about every single photographer who doesn't work with large format film and contact-print their negatives in a properly designed motorized vacuum frame (seriously).

In other words, 99.999% of the photographers out there choose convenience over quality at some (quality) point.

We're a blessed group of readers/commenters in that we have several of the 0.001% here.

pax / Ctein

Hi Mike

You really covered it. Well said.

I spent some years - at least six - chasing after how to do with m43, what I did with MF. I learned a lot about light, depth of field, lens contrast, distance-to-subject vs. angle-of-view, format sizes, the zone system, the history of lenses, depth cues, developers, histograms, the history of portraiture, and, of course, cameras.

It was a fool's errand, and a fine education. If I'd done an MFA, I doubt I would have learned more. (Though it would have taken me 1/3 the time!)

For me, all the advantages you listed, and more, make it more sensible to use m43 for pretty much everything. There are a handful of portraits that might benefit from the rendering of a few legacy lenses* on a focal reducer, but even there, lack of AF makes that moot. I've learned that common, humble, flawed lenses, can easily provide great results if I get to know them. (Something I learned from you, by the way. Thank you.)

*a Nikon Series-E 50mm lens that has a certain kind of low contrast at certain frequencies, and the Nikkor 45 gn, a Tessar design that flares in an interesting way, but also provides a tremendous amount of "sharpening" even when visibly out of focus.

I can't help but wonder what percentage of your audience who have read your and Ctein's article have ever even used a medium format film camera. And what about the people who seldom if ever make prints? This must be a rather abstract and esoteric discussion for them. What I would say to either group (or anyone else who's interested) is this: It all boils down to whether the camera and lenses you're using do what you want them to do, whatever that might be.

There is no right or wrong answer to this question; it all depends on your individual needs and expectations. If you're pleased (or dissatisfied) with the results then no amount of opinion-sharing by anyone else should convince you otherwise.

I personally could not care less whether Micro 4/3 is a good as medium format film. What I care about is that for my needs it's almost as good as full-frame 35mm digital, yet at a fraction of the weight, size, and cost.

Insightful words Mike! I was searching the web on 35mm printing and another of your articles came up; rather topical I thought.


Aesthetics is part of quality, but it's a subjective part. I kind of agree that from an aesthetic POV, digital falls a little short in the monochrome department, but I wonder if that is not a printing problem rather than a camera problem.

I am pretty sure that there is enough raw information and depth available in digital output to emulate just about anything, but printers have their limitations too, especially 8-bit printers.

However I would say that for colour (and I have been a colourist since Eggleston and Shore) the difference is far greater. Colour photographers have never had it so good, and that is partly the result of modern printing too. Colour developing, enlarging and printing were extremely difficult to get right.

Quite apart from anything else, the sheer flexibility of the digital medium offers almost unlimited scope to use colour in a more deliberate and active way.

The "edges" are always the most interesting places to explore, be it with film or digital. Right now, still, I'd give the edge to film. Film pushed to its technical limits can be beautiful. "On the edge"-digital is, more often or not, just kinda ugly.

"I think you were saying (in the entire article) that digital has a larger "sweet spot" because it has fewer tight limits, compared to film."

Dave, I think that's basically it. M43 will get you 90% of the way 90% of the time. A nice film setup can get you that 100% but under very limited conditions. I constantly battle with whether to try to get the "optimal" image or have the flexibility to capture more NEAR-optimal images. On our next trip I'd love to bring the M9 but practically I think I'll capture a lot more very good shots with the LX100.

So obviously I'll bring both.

No-one except you and your client cares — no-one! Whats right for you is wrong for someone else. You like bokeh and I use a Tilt & Shift lens at f/11 or f/16 to be sure I have no bokeh.

You like 120 Tri X and I like 35mm Kodak BW 400 CN, that uses color technology, and is about as grainless as you can get in B&W.

In HCB's Man Jumping Over a Puddle, the man is blurred, so is sharpness really that critical? HCB said: “Sharpness is a bourgeois concept” and Ansel Adams said: “There is nothing worse than a sharp image of a fuzzy concept.” I agree with both of them, YMMV.

BTW have you noticed Apple's Back Cover ads on Rolling Stone magazine? I'm impressed with the iPhone 6's quality. YMMV.

"If that technique better suited the things I wanted to do, I'd probably still be using it today."
That is the crucial sentence, I think, in this article. If you get the hardware that best suits the things you want to do with it, that's it, you're there.
After all, there's nothing better than the best, by the very definition of the word.

I somewhat agree with Mike when approaching it from that angle. But if you start out knowing the equipment being used you will carry with that knowledge your own opinions and this will influence your resulting opinion.

Consider a blind test. Examine 12 photos, any size you choose, without knowing in advance what camera took what. All you're told is that 12 different cameras and lenses (film and digital) were used. Color and black and white. You would be lucky to get 4 cameras out of 12 correct.
Never mind the lens. What does that tell us?

I have seen this test repeated in several different fields. Photography is not an exception. All that matters is if you like the end result.

That is all that matters. Everything else is suggestion and voodoo. Of course, a little voodoo is not a bad thing. :)

Three guys walk in to a bar. The printer photographer, the pixel peeper photographer and the theorist photographer. Which one is going to see what they need and how much is enough (two of these will eventually turn to drinking?).
IF printing requirements are the true measure of required quality, as Ctein is ample proof of, we are in a happy and blessed time. The other two camps may never be happy.
There was an enlightening article on Luminous Landscape a few months ago on the difference between 8, 20 and 50 mp FF Canons in printing conditions. Surprisingly small in real terms.
My poison was Agfapan 25 and Rodinol at 1:200 with minimal agitation for maximum physical development. Medium format results, but at medium format effort.

Good post. I agree there's something magic about prints made from medium format B/W film.

I used a lot of Panatomic-X exposed at ISO 12. Because of the slow film, I always had to use a tripod.

It didn't seem to be a popular film when I was using it in the 80s. It was probably just too slow and a bit inconvenient for most purposes.

Exposed in a Hasselblad with a Zeiss 80mm lens, and printed on Agfa Portriga, it produced some impressive print "magic" including lens rendering quality and tonality to die for.

I sometimes long for the days when those materials were readily available. Reading your post just triggered the nostalgia again.

This an interesting conversation, but it's not the only relevant conversation.

Digital cameras are good 'action' cameras.

They've somewhat solved the problem of high shutter speeds in so-so light.

So a relevant question is what is Micro Four-Thirds like at high ISO against, say, a Nikon D610?

I've also thought about why black and white prints seem to have a special magic. My main reading chair faces an original (and excellent) print of Ansel Adams's "Moonrise." After looking at that thing daily for a couple of decades, I finally concluded that part of magic isn't in the scene or the camera or the lens or the film, but in the printing paper itself. It might be measurable, if you could figure out what to measure. I think the fact that old b&w paper relies on grains of silver has something to do with it -- that the silver itself adds something to the way the paper reflects light. And I think that's why silver-print paper seems to have a touch of magic that the best carefully printed dye-based black-and-white doesn't; and why digitally printed color seems superior to traditionally printed color, because neither has silver, and with silver taken out of the equation, the other aspects of digital printing come to the fore. (Even traditional chromogenic processes eventually dispose of the silver component.)

An interesting experiment would be to take "Moonrise" out of its frame and then attempt to copy it digitally with a hi-res digital camera in carefully calibrated light, to see how much of its magic you could capture...and if intelligent, photographically aware people not familiar with the experiment could tell the difference, or would prefer one over the other.

Maybe if Ctein ever gets back to my house again, we could figure something out.

There also exists the possibility that I'm talking through something, but in my case, it wouldn't be my hat.

Ctein wrote: "...every single photographer who doesn't work with large format film and contact-print their negatives in a properly designed motorized vacuum frame (seriously).

In other words, 99.999% of the photographers out there choose convenience over quality at some (quality) point.

We're a blessed group of readers/commenters in that we have several of the 0.001% here."

But for the fact that I use a Condit spring back contact frame and not a vacuum frame, guilty as charged. Even when just starting out in this hobby during the mid-1970s, I almost always had my OM-1 on a tripod. Using up even (the then available) 20 exposure rolls was a chore. I've always worked like a large format photographer, long before knowing what a view camera was.

Within the last couple of weeks, I made the most satisfying (to me) print ever. I must thank several people, in addition to my own hard work, for it. First, in recent years, correspondence with Oren Grad and Dick Phillips helped liberate me from slavery to fiber-based paper. Their praise for Ilford Multigrade Warmtone RC, combined with a newly acquired visceral appreciation of being finite (I'm three years older than Mike), motivated a serious attempt to achieve high quality images on it. Then, six months ago, I contacted Ctein to ask for an update on the test of displayed RC permanence described in his book Post Exposure. When he replied that, 20 years after framing and hanging those prints, there was still no silvering out or bronzing evident, he sold me. California's drought and the vastly lower water consumption associated with RC prints means I can continue to practice my pastime without guilt or worry about deterioration during my actuarially probable lifetime.

The 8x10 contact print I made from a December, 2015 Tri-X negative is the best one (to me) I've ever produced, in every IQ aspect. I always work "at the center," so it's not likely that Micro 4/3 -- or even FF -- will be "better" (for me) in the near future. However, I will continue to monitor technology improvements with an open mind.

"...stuck in the middle, MF has you griping about needing f/22 and reciprocity-failure at 4am. No thanks."

Although Tim is referring to medium-format film photography, I ran into the same issue(s) after I spent a lot of money and moved from the m4/3 format to medium-format digital with a Contax 645 body and lenses, and a Phase One P30+ back.

One-minute exposures with an M4/3 body became six-minute exposures with a medium-format digital back when the lens was stopped down to achieve similar DoF.

Five minutes might not seem like much more time, but when you consider that I photograph outdoors late at night, it's plenty of time for someone watching TV to realize they've run out of beer, get up, put on their shoes, find their car keys, get in their car, drive down the block, make two turns and then drive right through the scene I'm photographing, ruining the photograph I started taking several minutes earlier.

Although I loved the image quality I got from my medium-format digital outfit, my yield-per-outing fell precipitously after I made the switch, from typically a dozen photos to as few as three or four, assuming I was lucky. On bad nights, it might be even as few as just one or two, especially during the summer, when I had to let the digital back cool several minutes between photos to keep the noise level down...

The bottom-line is that, for me and the scenes I photograph, medium-format was unworkable, despite having noticeably better IQ than the handful of m4/3 bodies and lenses I had been using previously.

Ultimately, it turned out that full-frame bodies, with their 24x36 sensors, offer the best compromise for my purposes, despite their files being not quite as good as those from my medium-format outfit and the bodies and lenses being somewhat larger, heavier, and more costly than similar m4/3 outfits.

I suppose I would have to try using film now and printing now to see, but I'm quite sure I'm getting better prints than I did in the darkroom days. I still have a lot of those around, so I can look at them. I printed a lot in some years back then, and only used large or medium format.

I'm printing a lot now. I'm mostly pushing my digital files to relatively large sizes, and they hold up generally very well. I didn't print very big in my darkroom for a couple of reasons. When I print my digital files at the size of my darkroom prints (I just recently printed about 25 smaller different black and white prints to send off), they are really something.

In the film years I didn't appreciate grain, and now I appreciate it in hindsight. So there's that.

The thing is, I've become a much better photographer in a number of ways since I used film and printed in the darkroom. Maybe I would be making better prints using film if I were doing it now, but I think the curve of improving was a whole lot slower with film. All I can do is look at my current prints, many of which come from MFT files, and know I didn't print anything that good before.

It's interesting that Mikes' post "Gone Fishin, Back Soon" immediately preceded the "Definitive Answer" post as he has certainly opened up the can-o-worms. When he first posed the question a few days prior I searched and reread Cteins' post about m43 print quality as I recalled a reference to 6x7 in that article. I appreciate his further commentary on this topic. I had moved from 35mm to 6x6 (Mamiya C220) in the mid 90's and was saving up for an RZ67 system in the early 2000's when digital started gaining traction. I held off until the Nikon D90 and now use E-M1 and Pro zooms. So basically I'm where I was trying to get to 12-15 years ago. Point of sufficiency for me. I'll rephrase a Thom Hogan comment along the lines of, if you can't get an excellent print from a desktop printer out of any crop sensor camera then it isn't the camera that's the problem. I believe if you can't get an excellent print of any size imaginable with current technology, film or digital, any format, then you should probably take up a new interest.

Great point by Ctein on handheld print size.
I never thought about it that way as 16x20 silver and 17x22 inkjet are the largest I have ever made in 50 years of printing.
When I want to make a larger picture I use paint and canvas, may seem odd but it works for my creative fulfillment.

The final result of any photographic image capture is to ensure the final result is satisfactory to all concerned.

The method/hardware part of the conversation matters not. The acceptance of the iPhone as a photographic rendering device proves that theory; how many images satisfactory to "you" were done with an iPhone? You probably would have difficulty knowning what hardware was used
for any resultant image in this digital age.

First we must accept that digital printing and silver printing are two very different things. It took me quite a while to accept it, I was always trying to mimic the look of silver prints with k3 pigments, and I can say that, after around $10,000 of inks and paper, to me, was not possible. I thought, maybe I don't have enough skills, but then I realized that some of my K3 prints were more attractive and better than the equivalent silver prints. I have a nice darkroom, still today, with a Durst medium format enlarger fitted with a superb 105mm f5.6 Apo Nikor. I can do decent prints, at a high price though, because usually 8 to 10 intents go to the garbage can. I'm not a skilled silver printer, but I have a good idea of how a good silver and platinum print looks. I did a PhD. at the University of Arizona in Electrical Engineering, just one block from the Center of Creative Photography. There, I had the opportunity to watch, many, many original prints from top photographers, including Adams, Caponigro, Eugene Smith, the three Westons, and many, many others, including Dean Brown, one of my favorite landscape photographers, who did small 4x6" dye transfers that were really a jewel. After my PhD. I returned to my country with the urge tu pursue printing. I did, and when I was starting to be happy with my results, the digital revolution started to take over. I swore to never do digital photography. Well, I swore in vain, I bought my first camera and an Epson 2000, just to try. I was locked even though my first prints were like shit.

Around two years ago I started to do digital prints that I fill are pretty good. My latest printer is an Epson 7880 and I'm using the Imageprint RIP software by Colorbyte, which I found esencial to get first class results, especially for B&W. It is quite expensive, but to me it doesn't have competitors.

Let me add more noise to the discussion:

"The best digital prints I'm making today with my setup are those using matte papers, specifically: Museo Portfolio Rag Extra Smooth Fine Art Paper, and Canson Infinity Rag Photographique. After all my exerience and waste of money, I firmly believe that pigmens are made for matte papers and that it is a mistake to use them on luster or shiny papers, except with images that quite a lot of blacks on it.

This image of my dog "Scar" grabbed with an Oly OMD EM1 and 75mm f1.8 (It looks a little bit to contrasty in Flickr)


prints superbly on the Canson Rag Photographique. It looks very 3D, the dog's head jumps out of the paper plane. I tried many luster and shiny papers (Museo silver, Canson Baryta, Harmann, Hahnemuhle, etc.), but they can't offer the same impression of aliveness, in a word, they look flatter. The sheen on the paper precludes 3D, unless you hit a spot of light over the print and look at it from the correct position, which I don't like. Light spots at exhibitions add an artificial vignette that I just don't like, they do the same with famous paints. Prints and paints are made to be seen in a well illuminated room with diffused light, and in these conditions, pigments over matte papers are just unsurpassed.

This other image of a rural tin shack in a poor region of southern Chile (also a little bit to contrasty in Flicker), taken with an Olympus Pen EP2 (12 mpix) and a chip 14-42mm f3.5-5.6 lens


Prints excellent on the Museo Portfolio Rag paper and looks very alive and natural. Again, there is no comparison with the prints from the top luster and shiny papers.

It is quite a bit more difficult to achieve a good matte paper print than it is with a luster or gloss baryta, you have to work the file a lot more, but when you get it, the result is much better, at least to my taste and that of the many friends I show my images.

Last year 32 of my prints were exhibited at the art center of a University. Previously the curator ask me to bring some prints to check. I showed him about 10 8x10" prints made with Museo Portfolio Rag and also with Canson Baryta photographique, he took two seconds to tell me to do the exhibit prints in matte paper. He took the prints with him to show them to other artist and every one of them choose the matte paper prints.

They don't mimic silver images, but to me, a lot of them look better. I strongly believe that we must forget to copy silver printing look (to me is not possible), and do the best with the tools that digital printing offers today, which are extraordinary.

Great post, Mike. I like your conclusion "Depends what you mean." exactly what I would say.

So it's not "definitive" after all. Then why did you say it was?

[Of course it was! --Mike]

I think this confirms the comment I made on the previous thread - that for colour, I agreed that digital is probably better than film now but it can't match a good, ooptically printed photograph.

Colour's main attribute is... well, colour. And digital sensors do that very well now.

Black and white deals in tonality and dynamic range. Get the colour bit out of the way and the film wins this contest.

Twas Brillig. Beware the Jabberwok, my son.

Mi dos pesos

Very nice. As you know, I am an admirer of Ctein's skills in photography and printmaking, both in dye transfer and inkjet printing, and I couldn't really think of how to respond to his post, even though I wanted to. In the end, you said it far more eloquently than I could. Bravo

John, speaking of "Moonrise", I tend to think that its magic derives primarily and essentially from a single fact: behind the groudglass, at that instant t, there is Uncle Ansel. Without Uncle Ansel, no Moonrise. It's so simple.

I think I get better pictures out of my Olympus m4/3 than my bigger Pentax DSLR for several reasons. It's easier to get good exposure first time with the live histogram, so I'm more likely to get the shot. It's easier to fine focus using focus peaking. It's much lighter so it sits more steadily on my tripod. All these reasons are nothing to do with sensor size and everything to do with being able to get the shot right.

A superbly concise and thought provoking article. When the issue of quality came up prior to your meditation Mike, I had 'quality is a concatenation' at the back of my mind.

An obsevration @ John Camp: I've been experimenting with silver prints from digital negs (in the manner of the platinum / palladium printers). This is precisely because I suspect so much of the 'magic' I see in darkroom prints is to do with the silver.

Ctein is absolutely right: Print size is crucial.
In film days, I never did enlarge beyond 11/14'' but at that size 35mm even with high contrast copy film and Rodinal 1:200 could not match the print quality achieved with the Rollei, Agfa 100 and Rodinal 1:100.
And in digital, as long as I could not make prints larger than A3+ (12/19?), I saw no difference between M43 and FF. But now, with an A2 (17/24"?) printer, that size has come to define my aspiration level, and now I do see a difference between the OLy EM-5 and the Sony A7R/RII. And even though I love the Oly in many ways, I do regret not having taken the Sony when the shot seems great in all other respects.

Much of our human existence is experienced at low light levels and in relatively close quarters. This is not an area of strength for medium format. Olympus' pixel shift high resolution mode is a nod to people who like to shoot in the medium format fashion. But I would contend that most micro 4/3 equipment is designed for low light and close quarter use. In that arena, the image quality is pretty good - but it clearly does not equal what the larger sensor cameras can capture.

"So a relevant question is what is Micro Four-Thirds like at high ISO against, say, a Nikon D610?"

I went from a D700 to m4/3. I would say that at equivalent ISO the noise/sharpness performance is about "1 stop worse". So while on the D700 you could shoot at 6400 in a pinch you really don't want to go past 3200 on an E-M5 or whatever.

The newer CaNikon bodies (APS and full frame) have of course improved by a stop or three over the D700, so this gap is probably wider now. But for practical purposes, esp. with image stabilization, I don't find it to be a huge limitation.

I mean, you can do this sort of thing hand-held from a ferris wheel bubble in London and it looks pretty good.


(two image panorama, handheld and then auto-stitched in Lightroom).

I imagine that at some point they'll make the sensors even better.

Mike, those are important points, and as you suggest, they apply to all sorts of discussions of quality: sensor, lens, printing... I find many discussions about image quality to be frustrating because people talk as if there's one ideal standard that applies to every need, which simply cannot be true.

When I shot and exhibited a series of winter landscape large format prints, bare tree branches against an overcast sky way out in the corners of the frames revealed lens quality issues that aren't normally an issue for people shooting portraits or street photos or a hundred other subjects. Crappy quality in the corners stuck out like a sore thumb.

So when people offer blanket statements about quality, you have to know that there are people out there for whom those standards are irrelevant.

And I think that applies to more than just camera discussions.

Interesting. I remember your articles on MF, and still have your review of the Rollei 6008, which put me over the edge for that camera 24 years ago. Still shooting with those less, and would rather stay with MF and go for optimal quality. While the percentage of "keepers" goes down, that is because of tougher standards. The keepers are more worthwhile, and there is nothing wrong with that.

The simple answer in all of this is that the subject dictates the acceptable outcome. If you want a top-quality B&W portrait of your grandmother, and you can choose between your OM-D E-M1 or your Mamiya RB67, which are you going to choose? But subject needs context too. Your grandmother may be a very slow, wrinkled old lady who pretty much stays at home. Your grandmother could be a 45yo who zips around and is very busy and doesn't have a lot of time to spare. You never get to choose the camera. The subject sets the terms. Even so, the old adage is that the best camera is the one you have with you. THAT even trumps subject. And in some cases, you simply can't take a good picture with what you ahve. Life goes on.

"In terms of the portability of the cameras, Micro 4/3 slaughters medium format film."

On this one small point, have you ever used a compact 6x6 folder like a Voigtlander Perkeo II? I can slip mine easily into a jacket pocket with a few rolls of Tri-X in the other. Amazing.

Of course m4/3 will beat it hands down for speed, IS, and just about every other metric. But if you want portable MF, it's hard to beat.

A couple of years ago I took a grab shot looking down a spiral staircase in a lighthouse in Oregon. It was taken with a Panasonic G5 and the kit zoom. Exposure was a half second handheld at ISO 400. My daughter saw it and wanted to make a big print to hang over her sofa. I looked more closely at the file and began to bite my nails. There was a lot of motion blur, poor depth of focus, some noise, etc.

I hunkered down at the computer and began to edit. A quick download for a free trial and the motion blur and depth of focus were fixed. Careful noise reduction and sharpening were applied in Lightroom. The file was converted to B&W using Silver Efex, enlarged in Lightroom and sent off to her printer.

I recently got to see the 30x40 inch print for the first time. From a cross the room I was blown away. I moved closer expecting it to fall apart. Knees pressed against the front of the sofa it still looks good. Only when I pressed my nose against the print did I begin to see the "flaws".

M4/3 is just fine for me!

Dear Sal,

Well, seeing as your one of the Good Guys, we'll make you an Honorary 0.001-Percenter [g].

Okay, more seriously, back when Mike was a dead-tree magazine editor, he hit me with the task of finding out just how sharp printing papers were (in order to end some rather magical-thinking debates about the merits of various paper formulations). I discovered that it took extraordinary pressures to get maximum sharpness in a contact print––tens of PSI. That's far more than you'll ever get from a spring-loaded contact frame.

(In case anyone cares, the answer was that every single print paper on the market, black-and-white and color, was capable of far more resolution than would ever make a visible difference in a print. As I said, magical thinking.)

Lest people think that this doesn't much matter, it's not just about sharpness. Analog photographic media are extremely nonlinear. Changing the balance of spatial frequencies changes the tonal placement. An example of this is a trick that old-time news photographers knew, which was to print badly-underexposed negatives slightly out of focus. This produced about a half grade harder print then you could get otherwise. The reason was that defocusing just a bit spread out the very finest film grains (and spaces between grains) enough that the print paper wouldn't respond to them at all. In other words, a little unsharpness clipped both highlights and shadows.

Weird stuff happens in analog prints. Understatement. Which leads me to John's question (next posting).

pax \ Ctein
[ Please excuse any word-salad. MacSpeech in training! ]
-- Ctein's Online Gallery http://ctein.com
-- Digital Restorations http://photo-repair.com

Dear John,

Off the top of my head, I would say that there wouldn't be any problem duplicating the tonal placement in your Adams print. The density range of that print is about eight stops, well within the comfortable range for any modern digital camera. My Epson P800 printer can produce nine stops of density range. Finally, an eight-stop print doesn't convey more than 256 distinguishable gray levels to a human eye, under absolutely optimum viewing conditions. Hence, “eight bit” printing wouldn't even be a problem, and the P800 supports 16-bit data flow.

And… my bet would be that if I performed this exercise to perfection, the result wouldn't look like the Adams print. Mind you, I wouldn't bet a lot, because it's just an educated hunch. But that's how I bet.

That's setting aside the obvious difference in appearance of an image embedded in emulsion and ink sitting on top of the paper. Which, it happens, I do NOT prefer. It's the one thing that rather bugs me about digital prints (and, for that matter, about carbro and tricolor pigment prints). I just don't happen to like that look. In case anyone cares, which they probably don't, I prefer the look of dye transfer prints to digital prints. I live with it because the benefits outweigh the deficits, and there's something to dislike in pretty much every print medium, and one needs to be realistic about that.

As I said to Sal, analog prints are very nonlinear beasts. That's not just in the making, but in the portrayal. The way light scatters, is absorbed, and is reflected within an image embedded in an emulsion is extraordinarily complex––scattering theory is one of the cutting-edge subjects in optical physics. It's very difficult to analyze. I suspect there is stuff going on in the way the light comes back to you from the print which isn't reflected in a simple description of density levels. Or, possibly more correctly, I might be able to make a print that exactly matched your Moonrise under one particular lighting condition, viewed from one particular distance and angle… and it would not match under different viewing conditions.

I'm totally guessing, here. But it's a semi-educated guess.

I can dispel one myth, though. It's not something special about silver, or any embedded metal image. That's more superstition, supported largely by sample bias and magical thinking. Whatever is going on, the data says that's not it. (And while we're at it, let's dispel the notion that “silver-rich” papers produce superior images. Yes, you can have too little. But that doesn't make more better, and it's more about how you use it than what you have.)

Here's only one data point, but it's informative: A considerable number of years ago, at Pierce's behest, I made some monochrome dye transfer prints from black-and-white negatives. They were glorious, just marvelous. Every bit of magic people think they see in a silver gelatin print, it paled in comparison. It would be the absolute premier way to make black-and-white prints, and there wouldn't be a bit of silver in the image, except for two things:

First, dye transfer prints aren't very sharp. They have lousy acutance. Second, monochrome dye transfer prints exhibit huge photometamerism. A print that is exactly neutral under sunlight looks distinctly brown under incandescent light and a rather unpleasant green under skylight. You just don't want to know what happens under fluorescents.

Nonlinearities, it's nonlinearities all the way down.

pax \ Ctein
[ Please excuse any word-salad. MacSpeech in training! ]
-- Ctein's Online Gallery http://ctein.com
-- Digital Restorations http://photo-repair.com

Don't forget the MF shots that never exist - the ones missed.

"So, is it fair to conclude that you choose convenience over quality?"

I choose the shot made with small, light, quick camera and lenses over the one not made at all with large, heavy, slow. I might even consider it infinitely better in IQ. \;~)> I have so many, many images I love that I simply could not have captured with MF gear, let alone MF film gear; too far away, too small, moving/evanescent.

Take yesterday; we were going to have dinner with friends. I didn't expect much in the way of good photo ops. So I had a GM1, the tiniest µ4/3 camera, with 12-32 mounted and 45-150 in my little bag. As we got out of the car in SF's Cow Hollow neighborhood, I noticed some great architectural detail on the top of a building up the hill. A quick change to the 45-150 lens and a shot did it. (400 g. for the two.)

Lest you think this tiny camera, small zoom and µ4/3 sensor short circuited quality, look at this 100% sample. Its size on my screen is equivalent to about a 16x18" print @ 300 dpi.

Had I a commission or the personal need, sure, I could have done this with MF, but we are talking extensive and expensive equipment, at least 15 minutes, probably much more, to also check out the view from a couple of blocks up the steep hill, in hopes of a better angle and in hopes of not needing a FF 300 mm eq. lens or a big crop. But there are trees . . .

In practice, were I carrying MF gear, I'd have not taken the shot, or perhaps taken it, then been disappointed with the crummy IQ of a huge crop taken without tripod or IS.

Most of my more serious photography is made where MF gear either won't work or wouldn't come along without a Sherpa.


...Okay, more seriously, back when Mike was a dead-tree magazine editor, he hit me with the task of finding out just how sharp printing papers were (in order to end some rather magical-thinking debates about the merits of various paper formulations). I discovered that it took extraordinary pressures to get maximum sharpness in a contact print––tens of PSI. That's far more than you'll ever get from a spring-loaded contact frame...

Dear Ctein,

I can't find my copy of the March/April 2002 issue of Photo Techniques magazine, but seem to remember that you tested both fiber-based and RC papers. While your discovery of a need for extraordinary pressure is consistent with the inherent curl of fiber-based paper, I have great difficulty imagining that a motorized vacuum frame would make any perceptible difference when contact printing Multigrade Warmtone RC. The latter is thin and dead flat, even when simply placed under the influence of gravity on a hard surface. I'm six diopters myopic and, when evaluating sharpness of my contact print by "putting my nose on it" while peering over my spectacles, cannot conceive of more resolution. It's far beyond your "perfect sharpness" threshold. I'm happy to live with whatever loss of contact might result from 'only' using the Condit frame with its 1/2-inch thick dense closed-cell polyethylene foam pad and very heavy pressure spring.

...it's not just about sharpness. Analog photographic media are extremely nonlinear. Changing the balance of spatial frequencies changes the tonal placement...

While that might be so, whatever tonal placement impacts one could attribute to not using a vacuum frame are built into my total system. When I make a print, it's the final result that counts. Burning, dodging, masking, etc. are part of the complete process loop. If I'm happy with the end product's tonality, it is "in the center" of image quality for me.

Thanks for making me an Honorary 0.001-Percenter. I'm looking forward to receiving the certificate/badge. :-)

Convenience or ability to work in a certain domain to me is not even remotely close to being part of Image Quality. No, I don't see how m4/3rds (and I have one) is equal to medium format film image quality.

That's almost like saying that my Pentax Q has better image quality than a laundry machine..

The comments to this entry are closed.