35mm sprocket holes

About my film scanning and post processing workflow

Theirry-Facon asked a few days ago about the scanning and postprocessing workflow I use for film. I referred to it briefly in an earlier post: More about my black and white film workflow, but today I will expand on the digital side of my film workflow.

(Please skip past if this sounds uninteresting, it’s not getting better from here.)

 After drying a set of negatives, I do post processing at my desk. It’s reasonably dust free, I have a HEPA air filter next to my desk which helps reduce airborne particles, and I wipe down the surfaces before bringing in the negatives. Before bringing the uncut film in, I also put clean negative sleeves on top of the work area, then use a small light and white paper background to help see where to cut the negatives into strips with scissors. For frames with dark / black edges, it can be difficult to spot the gaps, so it’s useful to have correctly cut negatives around as a reference for finding the interframe boundaries.

For 35mm film, the sprocket holes are a good guide, since they're  aligned perfectly parallel on both edges of the film. For 120 film this doesn’t work as well, since there are no holes and the edge flashing doesn’t necessarily match. A rotary cutter might also help, but I don’t keep one around.

The length of the cut negative strips will depend on your scanner and how you plan to store them. I have a bunch of very old 35mm negatives cut 6 frames per strip, which was typical for contact printing on 8x10 paper in a darkroom. Most consumer photo labs will normally cut 35mm 5 frames to a strip, which fits in the envelopes for standard 4x6 print orders. I also cut my own 35mm negatives at 5 frames to a strip, because this fits in my scanner’s negative holder, and makes each scanning batch 10 frames at a time, which is convenient for keeping track of progress. You can purchase negative sleeves in different sizes, but It’s important that your negative strips actually fits in your scanner.

The number of frames per strip can vary for 120 film. I shoot 6x6, 6x7, and 6x4.5 format cameras so the strip length changes for each to fit. I didn’t check the first time, and ended up with negatives that fit in the negative sleeves but were too long for my scanner’s negative carrier. For 6x6 and 6x7 I cut 3 frames per strip and for 6x4.5 I cut 4 frames per strip. I don’t have a 6x9 or 6x12 camera at the moment, but those would be fewer frames per strip to fit in the scanner’s negative carrier.

For the past several months I’ve been using a Canoscan 9000F Mark II scanner. These have a built in backlight for film scanning, and can accommodate both 35mm and 120 film. The absolute quality of the scan isn’t quite as good as the Plustek 8200 I used to use, but it is still very good, and it can scan 10 35mm frames or 1 strip of medium format frames at a time without intervention. The Plustek required moving the film holder manually for each frame, which makes it more time consuming. The Canoscan is also much less expensive, about $150 new, under $100 used. It comes bundled with a simple application for scanning and simple photo adjustments.

I don’t try to optimize the scan to make it look best. Instead I try to retain the maximum tonal and feature detail in the scanned image for post processing. This means moving the white, black, and contrast to spread the levels across the full range, and turning off any built in noise reduction, spot removal, and sharpening. The goal is to put as much of the useful negative density levels within the scanned dynamic range as possible, not to create the most aesthetic result.

Just as in darkroom printing, light areas of the negative correspond to dark/shadow areas of the image, and dark areas of the negative correspond to light/highlight areas of the image. Unlike darkroom printing, there is usually a hard limit on the maximum negative density that can be captured in a digital scan. This is because the digital scan just records zero or noise if the input is too dim, which corresponds to overexposed areas on the negative that are too dense for the backlight and sensor. On the other hand, it’s often possible to handle underexposed and shadow detail that would be difficult to print, first because this can usually be adjusted with the scanner’s exposure control, and in extreme cases an underexposed negative can be made slightly denser by putting in a neutral density filter, such as an unexposed or flashed piece of developed film.

The dynamic range characteristics of digital scanners can easily result in disappointing scans of negatives that look a little dense but otherwise  printable in a darkroom. So while modern negative films are extremely tolerant of overexposure, you still need to keep the digital scanning process in mind while choosing exposures, as you will lose more highlight detail than what’s made it onto the negative if it exceeds the maximum sensitivity of your scanner.

Here is a typical scan. Sharpening and grain reduction are disabled. The black point for the scan is moved to “10”. Remember that the “black” part of the image corresponds to the “clear” part of the negative, and the film base is not quite perfectly clear, and there’s no need to capture the tone values beyond the unexposed density of the processed film. If your software supports it, you should use 16 bit TIFF rather than JPEG output files. The scanner sensor will produce somewhere between 12 and 14 bits of dynamic range, and you’ll preserve more detail and tonality for post processing if you’re able to leave the scanned data uncompressed. This is the same tradeoff between RAW and JPEG format output from a digital camera.

 

 

I usually scan to 16-bit TIFF format at 3600+ dpi. Disk storage is cheap, and I’m often interested in preserving film grain and edge effects. I use pec pads to clean the scanner before each batch, and also use a rocket blower on each set of negatives. I have an anti-static brush and Tetenal also, but have never needed these on newly developed film. These are useful if you are scanning old negatives that have accumulated dust and crud though.

Scanning a roll of film at high resolution takes around an hour on my setup, but you only need to manually intervene 3 or 4 times per roll. This means that you can go do something else, like take a shower, get coffee, have breakfast, or run the scanning batch in pieces as you come and go during the day.

For doing any photo editing on a computer, you should be using a calibrated monitor. This is true for both digital images and scanned film images. I use a Datacolor Spyder for monitor calibration and matching. There are color calibration utilities built into Windows and MacOS but you really need a hardware colorimeter to get consistent results.Monitors will typically change brightness and color response over time, so it’s useful to recalibrate regularly. You still need to do color calibration for processing black and white, as you need to remove any color casts and make the luminance respond in a consistent way to use a monitor and achieve predictable output results.

I process and archive all of my photos, film and digital, using Lightroom. Once the negatives are scanned, I store them in sleeves and use the scanned files similarly to digital camera RAW files. First steps are to adjust the overall exposure, black and white points, contrast, and look for dust spots to remove. There are always dust spots, even if you wipe everything down. After basic adjustment, I’m not a purist about post processing film but will tend to stick with things I could have done in an analog darkroom process - crop, rotate, dodge, burn, vignetting, gradients, and limited blurring and sharpening. It’s much easier, faster, repeatable, and doesn’t burn through piles of (expensive, time consuming) work prints.

The classical chemical/optical darkroom process has properties that exceed what I can do with a hybrid workflow. However, sticking with “analog” types  of post processing adjustments means that in theory, I could use the digital version of a given “print” and construct a similar result using an analog process starting with the original film negative.

In my current situation it’s unlikely that I would have time to do this regularly, even if I had convenient access to a well equipped darkroom. It’s far easier and faster to apply sharpening in Lightroom than by constructing and printing with an optical unsharp mask. I enjoy the ability to apply my darkroom knowledge to construct similar effects in digital form.


If you’ve read to this point, feel free to ask questions. This is all specific to me and my setup, you may find other approaches more suitable but the general issues and decisions you encounter if you do your own hybrid film workflow are likely to be similar.