priscimon blog

In March, I experimented with using a DSLR to digitise negatives and liked the results more than those obtained from my low-end Epson Perfection V200 scanner. However, this sentiment is now reversed because I can produce better images from the scanner.

Previously, I scanned a negative with VueScan (using Lock Exposure and Lock Base Color features to neutralise the colour of the film base), saving the frames as TIFF images, and refined colours with Curves in GIMP. That method relied on my ability to correctly judge neutral colours and was therefore subjective.

The Film Negative module in RawTherapee uses maths to balance colours from film negatives, which is more consistent than human judgment. But until recently, it worked only with RAW files produced by DSLRs. Now the added support for TIFF files in RawTherapee 5.10 enables me to process negatives with it.

My new workflow is to scan each frame of a negative as a RAW image and to save it as a TIFF file. A RAW image is the direct output of the scanner and does not contain any image processing. I can work with the TIFF image in RawTherapee in the same way as I do with a RAW image from a DSLR. But Film Negative does the heavy lifting. Most of the time, it automatically finds the correct colours, but it can also get a good outcome from two user-selected points of neutral colour from a frame.

So far, I am very pleased with the performance of Film Negative. The picture above, which was used in my previous comparison, is the output of Film Negative. It is true to life: The skin, black jacket, grey dashboard, and red car look much more natural than in previous images from both the scanner and the DSLR digitisation. The photos in my previous post were obtained in the same way.

Theoretically, Film Negative works similarly on both a TIFF image from VueScan and a RAW image from a DSLR. However, experimenting with the DSLR method today, I could not get satisfying results. I am now reconsidering the purchase of The Lobster Holder, which I’ve wanted since my DSLR digitisation experiment.

Those pictures mentioned in my post about playing with artificial light? They were a total loss. I messed up the development.

I was not very happy about this costly mistake and needed to know what went wrong. So, I loaded another roll of film into the Nikon FM2 and drove straight to the country park to get some test shots.

It turns out that a 24-frame roll is rather long when you haven’t planned what to photograph. It took me nearly two hours to get the shots, mostly of random nature things, before I could return home for Round Two of developing Kodak Ultramax with Bellini C-41. Now armed with a sous-vide heater, I was confident I could keep a stable temperature throughout the process. I was also going to do inversion agitation instead of twirling the small plastic rod.

The development was successful, but was it a fluke? The only way to be sure was to do Round Three of developing with Bellini C-41. To be doubly sure, this time I used a 20-year-old roll of Fujifilm Superia 400. Again, shooting the 24 frames took some time.

The pictures that came out were as good as can be for expired film. I knew I had nailed developing with Bellini C-41.

I think the most significant change in my process was the use of inversion agitation—which looks like shaking a cocktail drink, except with less vigour—instead of agitation with the plastic rod.

Even P was impressed.

I bought a fresh roll of Kodak Ultramax 400 to test the film in daylight and under artificial light. I set up some LED lamps and flash lights and took pictures with the Nikon FM2. I also shot the same pictures with a Nikon D200 for later comparison. Here are the D200 pictures from the session.