Dealing with under/over-exposed & -developed negatives; scanning or enlarging

[Yezishu]

“The transmission coefficient of the developed film is proportional to a power of the reciprocal of the brightness of the original exposure.”
You see, this is not a simple linear relationship, so you can’t just invert it directly to get the best result. It will also deviate as the exposure and development stray from optimal conditions.

 

[dcy]

Case in point - my niece regularly feeds me a roll of poorly (variably) exposed Fomapan 400. She enjoys (for whatever reason) shooting film from time to time, but she does not have a solid background in photography, her capabilities in terms of exposure are quite basic and her expertise and interest in the technical sides of photography are superficial at best. So whenever I've processed a roll of her film, I find myself looking at maybe 6-12 frames that are properly exposed, with the rest being generally frighteningly thin in the shadows.

At at that point, I could go back to her, give her my sternest look, set her film on fire with a zippo and tell her she should damn wel do a better job because there's no polishing to these turds. Alternatively, I can gently explain that maybe she could set her camera to 200 instead of 400 for this film and see if she can find the time between changing diapers, getting the kid back from daycare and consulting with a neverending string of diabetes patients to have a look at one or two YouTube videos on exposure (but who am I kidding given her valid priorities). And I'll Wetransfer her images which I've diligently scanned, applied a totally dramatic, outrageous and insane contrast adjustment so that at least my niece will recognize her toddler without having to wonder if it's maybe the neighbor's kid, and guess what - she's happy. Because THAT's what reality often looks like. Whether you like it or not, that's what happens, and we're going to talk about that on a forum as well from time to time.

Can you gift her a camera with a light meter?

My wife's camera is too automated. It reads DX codes. Foma doesn't have DX codes, and if it had one it'd say 400 instead of 200. I went on Amazon and got a set of DX code stickers that read ISO 250 and gave them to her to sick on Foma 400 canisters.

 

[albireo]

I can confirm that this was my intent.



Exactly. My holiday photo is my holiday photo. I can't take it again. If I screwed up the development or exposure, I still care about how much I can salvage.

Ah, but then you'll have no issues at all. Yes, I agree with everyone else saying that scanning those screwed up negatives will in fact return an image, a document, and will 'salvage' your holiday snaps.

I mean, if you wanted, and the purpose was restoration of poorly exposed or processed irreplaceable analogue content, you could skip LR & curves altogether and feed those negative scans to Midjourney or Dall-E or Leonardo and get a great usable, actually more than usable, document. I mean if AI is now good enough to reveal new insights on the Dead Sea Scrolls, it's more than good enough to breathe new life into our thin negatives.

I was talking about something else entirely, and perhaps trying to stimulate a broader discussion on building a densitometric optimality theory for hybrid photography/scanned film, which is currently missing, and which needs to rest on foundations as solid as those existing for the full wet darkroom film photography chain and not hearsay.

So please do carry on, and all the best for your discovery journey!

 

[MattKing]

Kodak claims Ektar and Tmax films are better for scanning. Anyone know why?

At least partially because those films were created after the commercial processing market moved toward "scan plus print from the scan" norm from that which existed previously.
Among other things, the negatives tend to dry flat and to be easily handled.

 

[dcy]

Ah, but then you'll have no issues at all. Yes, I agree with everyone else saying that scanning those screwed up negatives will in fact return an image, a document, and will 'salvage' your holiday snaps.

I mean, if you wanted, and the purpose was restoration of poorly exposed or processed irreplaceable analogue content, you could skip LR & curves altogether and feed those negative scans to Midjourney or Dall-E or Leonardo and get a great usable, actually more than usable, document.

Matt said that certain types of exposure or development mistakes might be easier to manage in the darkroom and others digitally. There is nothing wrong with me wanting to understand why that would be the case. You telling me to use an AI to generate images instead of learning how to handle unintentionally poor negatives can reasonably be interpreted as... uhmm... showing me the door.

 

[MattKing]

Matt said that certain types of exposure or development mistakes might be easier to manage in the darkroom and others digitally. There is nothing wrong with me wanting to understand why that would be the case. You telling me to use an AI to generate images instead of learning how to handle unintentionally poor negatives can reasonably be interpreted as... uhmm... showing me the door.

I would stress that I really was talking about negatives at the margins.
I may as well repost this off-shared image - from a negative that upon first glance looks very, very thin.

That happens to be a scan from the negative. It also prints well in the darkroom.


I don't have a great example of a very dense negative to compare it with, but this one goes at least part of the way:

Where Did They Go?

• MattKing
• Tuesday at 10:00 PM
• 5

From nearly 14 years ago
43c-2011-08-12-res 1200.jpg

That happens to be scanned from a print (If I remember correctly).

I post these two as examples merely to make it clear that both very thin and very thick negatives can be handled well both ways. It is only when I get right out on the edges of usability that I encounter a bit of preference.

 

[Yezishu]

There are also some interesting discussions in a neighboring thread:
"A low contrast scene can easily tolerate 1 stop under and 3 stops overexposure without a significant problem, but a high contrast scene will have 0 stops."

Fujicolor 100

Nice. I am not surprised at all. Somehow this film always felt like being on another level for me. A step above Portras and Fuji's own 400H Pro. I have my last 4 rolls sitting in the freezer and I keep hoping for its return and hopefully even a 120 version. It is indeed an excellent film (so...

www.photrio.com

I’m hoping experienced users to generously share some insights on the exposure latitude of modern films for learning. How much can we over- or underexpose and still without big difference after standard development and scanning? Not only exposure error, but the presence of both highlights and shadow details is also an related issue. @retina_restoration convincingly mentioned that Tri-X can handle up to 2.5 stops of overexposure, which is very informative. Some C41 black-and-white films like XP2 even claim usable speeds from ISO 80 to 800:

Controlling contrast in B&W film

If you want to try DYI you can try divided D 23, D23 will tame the grain, in the divided form will also compensate for high contrast situations. From Chatbot but both AA and Minor White provided the forumal as well. The D-23 developer formula, used in black and white film development...

www.photrio.com

If exposure errors go beyond these limits, can push or pull processing in the darkroom dig more detail? While changing development time doesn’t create new information, it can sometimes reveal details hidden by standard processing. Adjusting the developer can also dig. For example, I recently used a low-contrast developer and extended development time to shoot microfilm rated at ISO 8–10 at ISO 64–100 before tonal compression became too noticeable. With standard development, density would be too low, but the information(not all, but some) was still present on the film.

Similarly, how much can scanning do for over- or underexposed negatives? Understanding the practical exposure latitude under different conditions—knowing when only minor adjustments are needed versus when “extreme contrast adjustments” are required—is very helpful. For instance, if my camera’s meter is normally off by +1 to -2 stops, I’d like to know whether the typical negative film can handle this error range or if I have to bring a separate light meter, or I need to be particularly cautious in low-light situations where I know my meter isn’t so accurate.

 

[koraks]

Can you gift her a camera with a light meter?

She has a Canon EOS 300 with a perfectly fine light meter; I've verified it works as it should. It's just that she lacks the experience to consistently nail exposures under the circumstances she uses the camera in, which range from snapshots while skiing in sunny-16 conditions and snow everywhere to Christmas dinner around the table in EV-sub-zero conditions. Mind you, between the first and the most recent roll there's a world of difference already; she's learning! But she can't afford (esp. time-wise) to practice a whole lot, so I expect there will be many more frames where a scanner and a dramatic twitch of the curves will save aunt E. from the blackness of zone I.

Matt said that certain types of exposure or development mistakes might be easier to manage in the darkroom and others digitally. There is nothing wrong with me wanting to understand why that would be the case.

Certainly; I'm glad we've been able to establish the scope of your question as you intend it and it's a perfectly fine and relevant question to discuss. Anyone who wants to explore a different/adjacent/broader topic is free to create a thread bout that, too.

 

[Mr Bill]

I’m hoping experienced users to generously share some insights on the exposure latitude of modern films for learning. How much can we over- or underexpose and still without big difference after standard development and scanning? Not only exposure error, but the presence of both highlights and shadow details is also an related issue.

Hi, I've posted about some pro portrait/wedding films over the years; here's one such...

What's the exposure latitude of Color Neg film?

Somehow, I don't think Drew photographs weddings. The initial answer to the question posed in the title is a question - "For which photographic usage?" If you are controlling the light, and aiming for extremely accurate and/or dependable colour rendition, Drew is completely right. If you are...

www.photrio.com

If you count down to the 8th paragraph, I said that our "good" exposure range on Portra 160 NC ran from about 1 f-stop underexposed to about 3 or 4 f-stops overexposed.

Within this range, after hand balancing everything to match (optically printed onto an appropriate professional color paper) that a professional color corrector essentially could not tell them apart. These were full-tonal scale portrait scenes, 3 or 4 models with a variety of skin complexions and hair colors. Fabrics from black to white and several of the strongest colors from a local fabric store. And several color test charts. EVERYTHING in the scene was a near identical match.

If one was willing to give up a bit they could say that one more f-stop in both directions was "usable." (But we we would not consider it so.) At 2 stops underexposed the darkest parts of the scene, black fabric in shadow, was no longer solid black, and was getting grainy. At 5 stops overexposed the strong colors were taking on a somewhat pastel appearance.

I should point out that these tests were shot using professional-grade studio flash, which is what Portra film is balanced to. If one were to use a different color temperature for lighting at least one of this film's color-sensitive layers would be offset from the others, and consequently the usable exposure range would be somewhat reduced.

I know that there are a number of internet posts/videos where some "experimenter" finds contrary to this. I don't know what they're doing, but suspect that it's often a scanning issue. Or, for non-professional film processing, perhaps sparse development is an issue.