-1
\$\begingroup\$

Medium format and 35mm film cameras have a distinct look that even film emulators in image editors for digital photos can't replicate.

Putting a digital photograph of Scene A next to a film photograph of Scene A, are there quantitative ways (e.g. color spectrometry, light tables) to verify whether any photograph file, when presented, was indeed taken with film camera rather than digital?

When opening the file in an image editor, without peeking at file properties, what are some dead giveaways to prove that an emulated picture was actually shot digitally?

Improve this question
\$\endgroup\$
3
  • \$\begingroup\$ "Prove" just means someone wins or loses an argument. Even in a death penalty trial, all it means is that the jury believes a picture was or wasn't made with film or digital and that the judge believes the question can plausibly be decided either way. Ordinarily, there's nothing on the line one way or the other and when there is something on the line, it's inconsequential. \$\endgroup\$
    – Bob Macaroni McStevens
    Commented Feb 11, 2022 at 21:21
  • \$\begingroup\$ What photographic problem are you trying to solve? Is there a specific use case for which you are trying to decide if digital would be just as appropriate as film? Or vice versa? \$\endgroup\$
    – Michael C
    Commented Feb 13, 2022 at 0:53
  • \$\begingroup\$ Film has greater aesthetic appeal? You're taken more seriously? \$\endgroup\$
    – user610620
    Commented Feb 14, 2022 at 19:44

2 Answers 2

Reset to default
0
\$\begingroup\$

The perceived dynamic range, the blacks/shadow areas and the highlights/whites and grain are the obvious pointers. There is also a perceived softness, even though it might be considered sharp. Also, especially when people talk about the perceived 'pop' from the film formats (not just large) is usually more evident in film photographs.

Grain is often hard to replicate faithfully in digital, and often in older digital cameras there was not enough head room on highlights leaving areas of clipping. Similarly in shadows digital noise can be evident if not treated.

Just an easy guide, quite generalised. Things are getting better in terms of dynamic range and have superceded in terms of sharpness..

Improve this answer
\$\endgroup\$
1
  • \$\begingroup\$ grain vs pixels? whats the difference \$\endgroup\$
    – user610620
    Commented Feb 12, 2022 at 7:07
0
\$\begingroup\$

The simple way to prove an image was made on film is to show the original film image -- if you're displaying digitally, take a photo of the negative, with rebates and a hand or similar, on a light table or held up to a window. A detailed image done this will will allow confirmation that it's the same image as your high-res scan, and no digital emulation will fake this (short of Hollywood or government budget CGI).

Giveaways for a digital image might include sensor issues (hot or dead pixels mixed in with the fake grain), image contrast edges that don't match up with "grain", pixellation at high magnification (that again doesn't coincide with "grain") or compression artifacts (thanks, @xiota) (though if highest quality or lossless compression is used, you won't have these).

Improve this answer
\$\endgroup\$
6
  • 2
    \$\begingroup\$ A negative could be produced from a digital original. Then the physical negative may need to be microscopically examined for digital artifacts. \$\endgroup\$
    – xiota
    Commented Feb 11, 2022 at 18:35
  • \$\begingroup\$ But then you'll have those, and sampling artifacts from their spatial frequency interacting with the digitizing method, in the final digital image. If you can examine the negative, however, you'll also see unmistakable artifacts from the physical camera -- microscratches, wrapper offset from backing paper, sprocket hole bending on 35 mm, pressure fogging, etc. \$\endgroup\$
    – Zeiss Ikon
    Commented Feb 11, 2022 at 18:40
  • 1
    \$\begingroup\$ If the original digital is presented, it won't have moire or whatever from being redigitized. Also, a photo of someone holding the negative would be fairly easy to fake nowadays. Live stream of someone handling the negative and writing a code communicated in real time would be better. \$\endgroup\$
    – xiota
    Commented Feb 11, 2022 at 18:56
  • \$\begingroup\$ A recreated negative would, however, have "fake grain" from the film emulation superposed on the actual film grain. Sure, if someone wants to spend enough time and money, almost anything can be faked; Unreal Engine could fake the live stream in real time at low resolution, I think. Possibly not the interaction, though. Not this year, anyway... \$\endgroup\$
    – Zeiss Ikon
    Commented Feb 11, 2022 at 19:03
  • 1
    \$\begingroup\$ That's why microscopic examination of physical negatives may be needed. Live stream isn't perfect, but prevents fakers who don't have equipment that can render fast enough with believable realism. \$\endgroup\$
    – xiota
    Commented Feb 11, 2022 at 20:47

Not the answer you're looking for? Browse other questions tagged or ask your own question.