I have a large number of black and white negatives from the 1970’s. Most were taken at live rock/pop concerts, and a number of them are underexposed.
Does anyone have suggestions for how to best scan and process these with NLP, particularly those that are very underexposed? My hardware setup is a Nikon Z8 with ES-2 film scanner, Nikon 60 mm macro lens and FTZ adapter.
Some of them are also dusty and scratched, so any pointers to taking care of that would also be appreciated.
I understand large volumes as that is what I am dealing with myself. How many are we talking?
That said, others may have better advice but my other question: Do you have a budget here?
Lots of solutions to help automate the volume of images if its truly large, from hand turn rollers to full mechanical automation.
There are tools to help with dust and scratches from as simple as brushes to buying a system with Digital ICE or creating a Digital ICE system
As for underexposure, it depends on how bad. If just a little normal digital editing works wonders, dodge and burn. If more is needed, they can benefit from a layering system to build up the values you are looking for which is primarily a post-process. Extreme, impressive example: https://www.diyphotography.net/andy-saunders-spent-years-remastering-thousands-of-apollo-mission-photos/
Just a few starting point thoughts before I run out the door.
Cheers
The volume isn’t huge - in the hundreds, not thousands, so that’s not a problem. I’ve already digitised several thousand slides. I’m more interested in learning how to extract the most out of severely underexposed negatives. I don’t plan to invest in more hardware, but any suggestions for NLP settings, capture exposure etc. would be welcome.
The cleanup would be the next step assuming I can recover a decent enough image from an exposure point of view.
The Apollo work is pretty amazing, couldn’t see any details of how he went about achieving the results. Is there an overview anywhere that you’re aware of?
Thanks for replying.
Places where he has talked at length in the past:
As for your images and others: It really depends on the image—sometimes there’s a lot of usable data, and other times not so much. I’ve been curious about this for a while, and it’s clear there are techniques to apply - and for more than just black and white film. I’m thinking this could actually turn into its own post sometime soon.
For now, I wrote up some thoughts and then got some help from an AI to tidy them up and add a bit more. The focus here is mostly on underexposure, since the techniques for dealing with overexposure can be a bit different.
I do not know enough about the nitty gritty of NLP to point you to the places it can help do some of these things. I use it fairly basically.
Others should chime in!
This guide is a good place to start without getting too deep or technical. It’s not a perfect fix for every situation, but it offers plenty of options you can try depending on what you’re working with.
Primarily for B&W analog negatives — but many techniques also apply to color negative and slide film.
Re-scan or duplicate and stack images:
Scripts > Load Files into Stack → Auto-Align → Convert to Smart Object → Stack Mode: Mean or MedianCheers
Test a typical negative by photo scanning. Exposure bracketing can get you an idea of how far to ETTR the remaining negatives. Hint: NLP is fairly tolerant to exposure when the histogram is narrow (should be the case here) and not too close to the right.
Test a conversion with NLP vs. a manual inversion of the RGB tone curve. Move the black- and white points inwards for higher contrast. Don’t clip either edge because structure (and the other sliders) can push the edges outwards.
I noticed that NLP B&W conversion tends to clip blacks and whites. This can be avoided by a custom preset that sets the points accordingly. I use a value of -10 each.
That’s much appreciated. When I get some time I’ll work through these steps and see how it goes.
I was wondering how to use NLP settings to best effect, so this is very helpful. I was also going to try just using just “normal” inversion in Lightroom to give me a reference point before jumping into NLP.
Converting with NLP results in a starting point. Individual adjustments from there on can be anything from nothing to complicated. And that is normal and to be expected.
On a positive note I’ve found that getting a good result from under-exposed negatives is much easier than it would have been in the darkroom, I was just using Lightroom, inverting of course but playing about with some custom S-shaped inverted curves which I’d saved. You can always do what NLP suggests and save a 16 bit Tiff when you get close to get the sliders working correctly again.
Blemishes, dust, scratches etc. do inevitably show up more than they would on a properly exposed negative but again they are much easier to deal with digitally of course.
From my experience with this, whether monochrome or full colour, the most important step for rescuing under-exposed media happens right at the film exposure stage in your digitizing set-up. Use a high enough shutter speed on the camera so that only the right amount of light is allowed to hit the sensor that shows-up the most detail you can see in the photo before exposure. (This avoids the main risk of over-exposure obliterating whatever detail there is in these negatives) It helps to tether the camera to a computer screen for doing this kind of work. The result will look kind of dull and grey at first instance, but that is exactly what you want as a starting point for the other kinds of software adjustments others have mentioned here.
