Resolution (Film AND Digital)

Discussion in 'Casual Photo Conversations' started by wogears, Aug 7, 2011.

  1. When I was a young sprout, back in the Sixties and Seventies, I used to buy the Nature calendars. (Still do, sometimes.) Looking at the marvelous images, I found I could--almost always--tell the ones shot on large format film from those shot on 35mm or 6x6 on Hasselblad (those were rare). This was on the equivalent of 8x10 prints made with (probably) a 150-line screen.
    Now flash-forward to a few years ago. I was looking through a copy of Lenswork, with its beautiful state-of-the-art printing at (roughly) 5x7 size. I found I could very easily tell the large format originals from the digital, even when the latter was from high-end DSLR. I turned the page, wondering if I shouldn't sell the 4x5 after all, even though I haven't used it in a long time. It puzzled me, though. At 5x7 inches, I just shouldn't be able to tell and difference. Yet on the very next page, I saw that wonderful lucid quality that I associated with LF. Nope. It was MF digital--'Blad with a 50(?) MP back.
    So what, precisely, is going on here? What is there about high-resolution originals that comes through even at very small reproduction size? How is the brain able to distinguish what should not be visible?
  2. I would honestly question the parameters of the "test". Testing like this must be performed in a double blind fashion meaning neither you nor the person presenting test prints to you have any access to hard information about the source. Further, the imagery would have to be controlled to isolate various factors from each other (resolution; tone; movements; contrast; DR; DoF; photographer's technical skill and style; etc). And the entire reproduction chain would have to be carefully studied, quantified, and controlled to insure differences were not introduced at that stage.
    Large format is absolutely beautiful when printed large. I love it. But it should not be possible to observe resolution differences at 5x7, all other factors being equal, especially given the limitations of a printing press. Technically speaking prints in the best art books I have fall short, regardless of source, of prints off my Epson 3880 from crop DSLRs, and this is at larger sizes. That suggests a DSLR should not be a limiting factor in a 5x7 press run.
    I don't have the most current issue of Lenswork, but I do have some older issues. While photographer styles certainly jumped out at me, I can't say equipment used was apparent to me at that reproduction size.
  3. I am inclined to say I agree with the OP, that larger format images have a look to them that is identifiable. Obviously the difference is not just resolution. But I think there is something, and it's not simply observer bias.
    Don't ask me what it is, though....
  4. I have a print on my living room wall that I did with my D70--6mp camera of a dead tree stump. It is smooth in texture and highly detailed. I cannot tell it wasn't from a 4x5 color transparency, which I have many of as well, of different subjects though. It has the same "qualities" that I would expect in a large format print of that size, about 11x14 inches. Go figure. I have thousands of 4x5 negs and trans over the years as well as thousands of images from D70 and D80. The majority of my digital images do not look like they were from a large format camera, but some do. Many could pass for medium format though. I think it is the smoothness and lack of grain, which in my film days took at least a medium format camera to produce easily in a print.
  5. The look of film is different to digital. I shoot in several formats - Digital (4/3 with FD lenses, APS-H, APS-C and full frame), 35mm film (Canon FD, Contax G, Canon EOS), 645 (Mamiys Pro and Pro TL) and 6x8 (Fuji GX680). I am also able to scan using a Nikon 9000 and a Nikon 5000 scanner and I have a B&W dark room set up. Here is what I have concluded on film vs digital. I will use the 5DII as the digital standard as it is the best image quality of the bodies I own.
    In terms of 35mm the 5DII will produce a higher resolution image than a scanned 35mm negative or positive. That said the look and feel of the film image is different - especially with Fuji Velvia (which I like a lot) and the Contax lenses have a quality that my FD lenses and EOS lenses cannot match. I should state that in both FD and EOS I have extensive collections of the best lenses Canon makes (e.g. 85 F1.2, 17 F4 TS, 135 F2, 100 f2.8 LIS, 85 F1.8, FD80-200 F4L, the full set of EOS F2.8 zooms etc...). You need to print at 11x17 or larger or pixel peep to see any resolution difference with digital.
    With the Mamiya M645 the resolution of a scan and the 5DII are very similar but differ in the details. My own conclusion is that the 5DII is reaching the limit of its optics - especially at the edges whereas the film is limited by the scan and film performance not the lenses. The difference in image appearance is obvious but to see any resolution differences you need to pixel peep.
    With the Fuji GX680 (which has some amazing lenses - such as the 180 F3.2) the resolution is clearly higher than the 5DII. You can see the difference here on a 17x11 or similar size print. It should be noted that a scanned Tiff from the Fuji can be 300 - 600 MB for one shot! The fuji image scans at 4000 dpi at about 12,000 x 8,800 pixels or over 100 MP.
    This is why the film images from larger cameras look so good - they are incredibly detailed and the large film size means that the lens is not being stressed to resolve. When you compare a state of the art MF digital body I suspect that you need very large prints or the ability to pixel peep before you start to see the difference. I also suspect that the weak anti alias filter of the digital MF backs is one of the things that makes digital seem much closer to film in terms of colour, contrast etc...
  6. It's not the resolution. It's the larger frame and the lenses. I can get scanned frames from my blad and digital shots
    from my DX Nikon to come out with the same resolution but they don't look the same.
  7. Are we talking about tonality? In other words, does the larger sensor/film size give you the capacity to distinguish more tones and more subtle tonal distinctions?
    Roger Hicks, who no longer posts on, used to say that was due to something called the 'halftone effect'. I didn't really understand that, but then, I haven't read his books where he explains it.
  8. I read a photography book not long ago in which it was very clear which images were shot on 35mm film and which came from medium format - the resolution of the 35mm was visibly inferior even in a 4" square print, although I guess it could have been the reproduction technology that was limited (I'd hope it was a drum scan, for a commercial photography book...) In landscape photography magazines it's often clear when a larger format has been used, but that may be because f/22 and below don't cripple the resolution as much on larger formats. Shot at a large aperture to rule out diffraction (probably f/8 and larger for modern sensors), there's no reason a small digital sensor shouldn't be able to produce as sharp an image as its low-pass filter and post-processing allow, especially if the lens is good enough to retain decent contrast at the sensor resolution at that aperture. It's quite common for an image to look absolutely fine until you compare it with a higher-resolution one, though - perhaps a lot of images are being over-enlarged.

    That said, the highest resolution (terrestrial) images I've seen were taken with small-sensor digitals - but stitching a vast number of images.
  9. How could you "always tell" what camera a shot came from? That information isn't disclosed on most calendars. Are
    you sure that you're didn't see a sharp image and imagine that it came from a view camera when actually it might
    have been shot on 35 mm?
  10. With small prints made from film shots it is not the extra resolution that makes the differance, it is the lower grain visibility you get with a larger format.
  11. Last year I had the opportunity to exhibit at an art museum. As the curator checked in my prints she commented "I misunderstood. I thought you worked with all digital." She thought the prints were from large format film. I only shot 4x5 for my landscapes for years, but all of these prints were from a Canon 20D and 5D.
    My walls are hung with large prints from a variety of medium format cameras and 4x5, as well as digital ranging from 8mp to 21mp. I ask visiting photographers to pick digital from film. My experience has been they are not particularly good at it. If the film prints show grain (some of the medium format prints) they quickly pick up on that, but they often choose the 4x5 as digital, and the other way around.
  12. I suspect it's going to depend a lot on the style of images and the quality of the equipment used. Many people say that digital is a test of good glass; while there a few digital-specific requirements (notably telecentricity and nonreflective rear elements) it's usually the ability to pixel-peep that makes previously acceptable lenses look bad. As I mentioned before, stopping down and causing visible diffraction will cause a more visible effect on smaller formats.

    I dispute that the problem is always film grain. I've seen very grainy 5x4 and very smooth 135 film (at different speeds and with different processing). Unless the softness was an attempt to de-grain it, the 35mm prints in the book I mentioned - which was at least twenty years old and must have had limited digital processing - did not appear grainy, but did appear soft.

    I would claim that the distinction might have more to do with the contrast rolling off near the resolution limit of the lens, and that this effect might soften 35mm images. WIth larger sizes, while the lens may not be any sharper and while the limit of the resolution of the smaller format may not be large enough to be visible in the print, the differing enlargement may give higher-contrast edges to the bigger format. Since digital has a hard resolution limit, it may behave more similarly to a larger format (until the resolution limit is reached) if processing is applied - or if lenses that are deliberately designed to give contrast rather than resolution are used.

    That may be a lot of irrelevant waffle, and I could be completely wrong. Just my impression.
  13. How could you "always tell" what camera a shot came from? That information isn't disclosed on most calendars. Are you sure that you're didn't see a sharp image and imagine that it came from a view camera when actually it might have been shot on 35 mm?​
    No. The Nature calendars, unlike the ones from the Sierra Club that were also popular 'back then', had the photographers' names and technical details. These were NOT printed on the page with the image, so it does represent a kind of single-blind test.
    Please, let's not do film vs digital. The calendars were produced long before digital--even scanned film--and the MF digital had all the appearances of the LF film images published in the same magazine. It is correct that at the small sizes I ought not to be able to have told the differences I saw, but I could.
  14. I didn't realize that Nature was a brand. I assumed that it was a category. I've seen a lot of nature calendars over the
    years. None of them ever had camera information listed, and in many cases they don't even mention the
    photographer's name.

    "let's not do film vs digital" - Okay, but your subject and your original post both imply the comparison, especially your
    own comparison of LF and a 50 MP Hassy.
  15. Digital keeps getting better & better. I shoot film & digital & I think the digital often looks better.

Share This Page