Faded negatives: normal aging or problem with technique?

I’ve been scanning my old negatives with my camera (Sony A7R III and a Canon 100 Macro lens with a 35mm film carrier). I’ve noticed that some of the negatives come out looking very faded, with nothing to compare them to, I don’t know if it’s a problem with my technique or because they’re old (some of them are 40+ years old and kept in poor conditions).

I recently found a roll of negatives taken by a very cheap waterproof disposable camera from 2014. Apparently, I also had the negatives developed & scanned at Walgreens at the time.

The color/contrast different between my camera scans and what Walgreens gave is quite stark. Is this because the negatives have degraded in the intervening 10 years? Or because there’s something wrong with my technique?

I realize I can probably do a better job if I manually edited it with curves in PS to eliminate some of the color casts, but that’s going to take way too long, and I have 1000s of negatives to process

So if there’s any way I can systematically improve my technique, that’d be the easiest way

Here’s the link the raw file from my camera (scan) if anyone is interested: https://1drv.ms/i/c/499d6900d1cab18c/EbRaKUUkM3xChbamyKhKf1wBlXU8JjNto6aUJ6QWPWVMVw?e=YwTicq

2014 Walgreens Scan:

From NPL:

Added some curves in PS

I don’t have an answer for you but giving this a bump…

But my semi-educated guess is that the Walgreens scans have a curve applied to them.

Definitely. The Walgreens photo definitely had a curve applied and has much more contrast. But for my camera scans, beyond just lacking contrast, there’s also an an ugly color cast. E.g. on the white cabinet door. I wonder where that’s from?

I tried getting a few rolls scanned at a local place that has a Hasselblad X5 scanner. It’s also scanning aged negatives, so any differences can’t be because the negatives deteriorated because of the time difference.

And again, the difference between the X5 scans and my at home camera scans are radical - much larger than I imagined. The shadows are completely crushed despite lowering contrast to the maximum I can in NLP. Preformatted textAm I doing something wrong?

Settings I used to convert the camera raw file in NLP

Left is camera scan; right is X5 scan

You can clearly see the detail around the steering column, the key etc on the X5 scan so you must be able to see it on the actual negative as well. Can you see it on your copy of the negative? Maybe you could post a link to that also, it might be a better one to use than the link above.

Have you also tried copying positives/transparencies, if so how do they look?

How did the lab deliver your X5 scans - as a positive or as a negative or both? If it was as a positive then they will have processed it in their own software, probably Flexcolor or the later Hasselblad version (Phocus?), but possibly they have their own method, even NLP.

The lab gave me the original FFF files. After trying to read it “properly”, I gave up on using the ancient Hasselblad software and just renamed the extension to TIFF. Then I ran it through NLP

But anyways, I discovered what the issue was. Turns out I was converting in NLP before I cropped and I wasn’t setting a large enough border size. After I cropped the 35mm negatives first, and then converted in NLP the results turned out to the almost as good (but still not quite) as the X5 scans (the X5 scans were still much more neutral)

That’s good news, similar level of shadow detail? I’d expect so as the X5 is quite old now (2006), 10 years older than your A7R3 but very much the top of the tree of the Imacon/Hasselblad ‘virtual drum’ scanners. I actually like the Flexcolor software for converting colour negatives from 3F files, it was one of the first pieces of imaging software to use non-destructive editing, possibly the first, I’ve not tried Phocus.

Yes, crop first or set a proper border size. Other things that help:

Be sure your macro lens is spotlessly clean on both sides, to reduce flare.

Mask off all extraneous light sources that might flare into your macro lens.

Be sure your film is held completely flat. Underexposed negatives, in particular, are easily flared by light bouncing off the edge curvature of the film.

Be sure your light source is completely diffused with Perspex or milk Plexiglass.

Be sure your light source is 96 CRI or higher, flicker-free, and made for photo and video applications.

Do as much color adjustment in NLP as possible. The controls are better and easier to use on negatives than most of the controls in LrC.

One thing I learned while running a huge scanning operation in a pro photo lab back in 2000-2005 is that every film emulsion batch of every film type, and every film brand, is different. Film AGE at both exposure and development matter, as do storage conditions before and after exposure. Film development matters a LOT, especially if the lab is experiencing low demand and the chemicals are not as fresh as they should be. Negative AGE and storage conditions matter a lot, too. I have some negatives from the 1960s and '70s that scan acceptably. I have some negatives from the 1990s and early 2000s that won’t yield decent color, so I converted them to B&W files.

Particularly important are negative EXPOSURE conditions. Daylight is usually fine. Dawn, dusk, and incandescent sources are amber. Fluorescent sources are generally green, but can be all over the map. LEDs can be all over the map depending upon quality. Film exposed in buildings with 277-volt fluorescent “troffers” in the ceiling will be ALL over the map in both density and color! Underexposure and overexposure will change both contrast and color balance. The shift is usually across the cyan-red color axis.

In the end, yes, patience, experience, and expertise all matter, so “tune for maximum quality” and notice it becomes easier. Save settings for particular film types when you get something you like. I have a bunch of them. If you currently use film, remember the exposure conditions and batch scan according to them.

Here’s an overview comparison. The contrast was lowered on the Sony (left) shot and default on the right (X5 shot)

Here’s a closeup of the shadow areas. Note the X5 scan was only done at around 2000dpi, so much lower resolution than the Sony
![Screenshot 2025-01-19 at 2.35.45 AM|1380x758]

(1) If metering from the camera, is it better to under or over expose? Or just ETTR and not clip the highlights? Should I be changing exposures when scanning the same roll of film?

(2) Is there any easy way to determine the CRI rating of my light source? It’s an LED panel