Wondering why I still shoot film

[Thad Gerheim]

Out of curiosity, did you mistake an IQ180 print for 4x5" or vice-versa?

I mistook an IQ 180 for a 4"x5" film shot. It was a shot of redwood trees, and it had very good intricate detail and sharpness. I wondered if that one was mistakingly labeled

[Pere Casals]

Just one final comment - it's just as likely this test is flawed in some way, giving an advantage to film, compared to the LL article. I skimmed over things but one thing that jumped out at me is that the sharpest 8x10 shot was at f/16 or so. Well yeah, of course 8x10 will have more resolution at wider apertures when not fighting diffraction. This is why theoretical lp/mm is irrelevant - at common shooting apertures 3-5 stops lower due to DOF concerns in many situations, your resolution suffers. I can't comment on the LL test as it's behind a paywall these days, apparently.

At f/22 not much is lost from /16 as tests indicate, a viewer observes some image quality only percetual if pixels are 2x, and beyond 60 perceptual MPix viewer cannot see more, just we have a limited number of sensor in our retina.

Then a 4k Tv/monitor has 8MPix. Even my Sony Xperia Z2 mobile fulfills that. And the D610 I'm using exceeds that by a wide margin. Even enlarging 2m a DSLR shot still renders decently.

If you look those tests you need to enlarge 4m and see it at reading distance to observe that detail.

What I say is that such resolutions are an overkill, no need to compare, but if we are curious and anyway we want to compare still a 60 years old CAMBO with a 1978 lens beats down a $23k top notch digital system in terms of optical performance. Of course Digital is convenient, you can shot 1000 shots in an afternoon for free... after investing in that gear.

I do not shot LF because resolution power. I shot it because movements vs narrow DOF 8x10 delivers for portrait, if I want more DOF I shot 4x5 or MF... Also to me is very important spectral signature, Portra, Ektar, Provia, Velvia have unique spectral signatures (conversion from a continous SPD to 3 rgb values) that cannot be matched later with PS. With digital I feel tied to the on pixel dyes, that are generalist, not good for portrait, not good for landscape.

Not much people use film today, these is good because I can peek excelent glass with my limited amateur budged. Not only inmutable, every year some film disapears and little is new, no problem, if all disapears we'll shot dry plate...

Just one final comment - it's just as likely this test is flawed in some way, giving an advantage to film, compared to the LL article. I skimmed over things but one thing that
Not to mention the 8x10 system is pretty much immutable - likely no new, improved films are going to get made, or new lenses, etc., while digital will keep improving. Since there's already a new IQ model out apparently, this test is already out of date and now irrelevant no matter how you slice it. Digital in the last few years has really exploded in terms of resolution and abilities, and it just keeps getting better. Which brings us right 'round to the original point I made way back in post #3 I think - does it matter?

With a bare consumer DSLR I can beat 8x10 (and IQ 180) in landscape, I can stitch 100 shots in PS, no viewer will realize it. Film MF exceeds the needs about resolving power (IMHO), LF offers a lot of additional aesthetical tools, first if the particular DOF look. The 800 "MPix" overkill of a 8x10 comes as a bonus...

But.... what about a 8x10 velvia in a light box ? underexposed, with some 5000 cd/m2 under... what about that !

Rent it.

Now off to develop a bunch of film I shot today.

I would be a nice experience, but I think in my remote location is not easy...

IQ280 is an amazing device, also it has some image quality advantage in front of a D810, not much, but this is irrelevant for most of practical applications, simply because today's devices have resolving power in excess.

The edge I see for IQ280 is not "sharpness" but just the counter: defocus. Larger sensor uses a longer focal for the same framming, so DOF has the nature of MF cameras. Then we can use sweet ancient glasses, like Pentax-Takumar 105mm 2.8, with incredible Bokeh that most or today's perfect lp/mm based glasses lack.

[Drew Wiley]

Big deal. Someone could just come along, take 100 shots on real 8x10 film, then scan and stitch them all together. Or if you have the budget, you can precisely
synchronize a whole array of giant telescopes over a square mile and stitch that, like they do in the Atacacma Desert. Now that's what I call large format! As
far as I'm concerned, stitching has always been a job for sweatshops. I'd rather spend that time outdoors taking shots that don't need to be stitched. Hypothetical -this/hypothetical-that. Nobody on the planet can beat even 4x5 film with any kind of 35mm DLSR, maybe not even 6x9. That's marketing make-believe. But you have hit a sensitive nerve of accurate hue reproduction, esp warm neutrals - it might be hell with film if you're as fussy as I am, but damn near impossible digitally. Easy with a basic assortment of watercolor pigments and a little brush. So technology seems to be going the wrong direction. But if you actually enjoy crunching numbers, that's fine. I don't. Reminds me of all these MBA bean-counters who drive companies out of business.

[tgtaylor]

From a May, 2007 study:

How much information can be captured on one frame of 35mm color negative film? How does
that compare with the best digital image capture systems? We put those and other related
questions to Nestor Rodriguez, senior principal scientist for the Kodak Entertainment Imaging
division.
Question: How much image information can be captured and stored on a single frame of 35mm
color negative film and how does that compare to today’s best digital cameras?
Color negative layers before development
Answer: A single frame of color film scanned at 4K by 3K resolution with
10-bit depth contains about 50 megabytes of data. However, there is
actually a lot more information than that on each frame of 35mm film.
We have conducted tests where we have scanned film at 6K by 4K
resolution at 10-bit depth, resulting in about 100 megabytes of data,
or twice as much image information. In comparison, a typical CCD or
CMOS RGB three-sensor 1920 by 1080 electronic camera with 10-bit
depth records 8 megabytes per frame, assuming that there is no subsampling
or data compression. A single CFA sensor 4096 by 2048
camera records about 10 megabytes of data. So the simple answer is that
today’s best film technology enables you to record 5 to 10 times
more picture information on a single frame than the best
contemporary digital cameras.

For the full study see http://motion.kodak.com/KodakGCG/upl..._film_info.pdf

Thomas

[Drew Wiley]

I should have qualified those remarks, lest I seem like merely a curmudgeon once again. Here's how I approach the whole subject. I go up town to see a friend who has it all, who's done it all, and done it superbly on a commercial basis, even become a multimillionaire as a studio photographer, and has absolutely no problem hauling six hundred thousand dollars worth of studio gear to the dump and spending a million more if he thinks its going to improve his workflow. But he's getting old and no longer wants to pay the hazmat expense of maintaining a full chemical lab. So he's gone full digital, even though he doesn't like the look of it himself. So I look at output, the final product itself. Neither of us give a damn about the math. It's pretty remarkable, but with a Phase One on a Sinar he can damn near arrive at the store window display quality he used get printing 4x5 film optically, with an enlarger. Certainly not to the same level of sheer detail, or overall "gallery" quality, and nowhere near the league of 8x10 film. Same goes for people I know who now serve as guru preachers or hired guns for the
inkjet industry. Good stuff; but they did better back in their darkroom days. So what do you mean by "better". If you can make more money firing a bunch of
assistants because you can streamline the workflow right from the point of capture to prepress, that is unquestionably better in terms of cost efficiency. But if
your goal is optimal print quality, the subject is a lot more involved and decidedly more subjective. For me, the apex of photographic quality might just as easily
be something taken with a box camera in the 1860's.

[unityofsaints]

I don't mind much if I scan or print optically - it mostly depends on my mood and whether it's night or day as my "darkroom" isn't dark enough when there's daylight outside. I wouldn't dream of using digital for capture though - it's just not in the same (highly subjective) league as film in my opinion.

[John W. Browning]

Oddly enough I have just had a resurgence in the use of film. All of those coveted medium and large format cameras that were out of reach before the advent of digital are now finding their way into my possession onto my tripod and being put to use in my business. Film still has something about it. I love shooting E-6 stuff and displaying it on wall mounted light tables. I do wet plate portraits and clients love them. The old has put the fire back in my passion for my work. I do shoot mostly digital for a living but I truly believe that film will no more totally disappear than oil paints will.
But if you have lost interest in film you probably should abandon it.

[Peter Collins]

I am 71 years old and do not want to learn digital, if I could learn it. Brain not what it used to be.

[John W. Browning]

I am 71 years old and do not want to learn digital, if I could learn it. Brain not what it used to be.

I felt the same way. All of the same principals apply and we get to cheat with the ISO (ASA) by moving it around. I have learned to live in both worlds and am loving it all over again.

[Pere Casals]

Big deal. Someone could just come along, take 100 shots on real 8x10 film, then scan and stitch them all together. Or if you have the budget, you can precisely
synchronize a whole array of giant telescopes over a square mile and stitch that, like they do in the Atacacma Desert. Now that's what I call large format! As
far as I'm concerned, stitching has always been a job for sweatshops. I'd rather spend that time outdoors taking shots that don't need to be stitched. Hypothetical -this/hypothetical-that. Nobody on the planet can beat even 4x5 film with any kind of 35mm DLSR, maybe not even 6x9. That's marketing make-believe. But you have hit a sensitive nerve of accurate hue reproduction, esp warm neutrals - it might be hell with film if you're as fussy as I am, but damn near impossible digitally. Easy with a basic assortment of watercolor pigments and a little brush. So technology seems to be going the wrong direction. But if you actually enjoy crunching numbers, that's fine. I don't. Reminds me of all these MBA bean-counters who drive companies out of business.

Hello Drew,

Of course... stitching LF pictures... don't think that something like this has not been done !!! NASA had "weird" ideas Moon was mapped by stitching a lot of 5" film images, by Apollo program (15-17 missions), and using a KA-80A aerial camera...

https://www.google.com/moon/


No Full Format 36x24mm DSLR comes close to 4x5", but a very good shot from a MF digital camera (Hassy HD5), with multishot and a very good lens... this comes close or surpasses a bad 4x5" shot,taken with a not best film or glass. Still a good 4x5" shot can have more information than the HD5. And also it's a very different type of result.

I guess that looking those numbers it can be interesting, but not much useful... at the end an LF optical print can deliver a clearly superior output, as long one has the knowledge to do it well... (I'm trying... )