I am currently trying to make some more advanced improvements for my scanning process/set-up and would like to open a conversation about experimentation here.
My end goals is to feel like I don’t need a Frontier scanner anymore to get results that satisfy me (especially ones made by some good labs like Carmencita). I am aware I will probably never get to that level with DSLR scanning but I want to try to get there 95%. I am in no way a technical expert but I want to see how far I can go in copying what a Frontier does, even if the sensor of that machine works differently
For context: I am currently using an A7IV with a sigma 70mm f2.8 macro mounted on a Kaiser repro stand above a basic Negative supply kit and a CS-lite panel.
One thing I keep seeing that is similar to what a Frontier does is using an RGB light to capture the 3 color channels separately and merging them afterwards for better colours. This is something I want to try in the future.
Another thing I don’t see people talking about but I came to because of chatgpt is using cross-polarisation by putting a polariser on the light source and another on the macro lens. According to Chatgpt this is something the Frontier scanner also does and might be worth trying for fun. Though I have no idea if it will do anything for me.
Is there anyone here who has any ideas of how to get results closer to that of a frontier scan? Or else just more consistent results with that rich color, texture and highlight falloff a frontier has?
If I’d target a specific look, I’d take a few well-exposed fotos of a known target and have those fotos developed/scanned/printed in my favourite lab and with NLP. I’d then use Lr to work on a positive copy to make it look like the lab-scans and prints. Save those settings as a preset and test, if it works on other fotos as well. I’d use a calibrated screen and set monitor brightness to about 80-100 cd/m2 in order to make whites on the screen and the photo paper look about equally bright.
Scanning with separate R, G, and B lighting has the potential to make colours look cleaner. Whether that will work with your goal and if it’s worth the effort is something only you can decide. Your decision might also depend on what you intend to do with the converted images, look at them on a screen, a print, in a book etc. Some of the processes involved in printing might not be in your hands, so results might vary.
… is mostly used to eliminate reflections. Many polarisers change colour balance which means that you’ll probably need to create more presets.
Last but not least: Changing film stock, exposure etc. might overthrow all you did above.
Wondered if you’ve ever tried their Noritsu scans, and if so how they differed? Not only has the Frontier the advantage of an RGB light source but it also has software dedicated to that particular machine created by Fuji themselves.
Richard Photo Lab has a good page on the differences between the two:
As someone who previously worked in a lab with both Frontier and Noritsu scanners coughRPLcough, a lot of the “look” that Frontiers end up with are due to experienced scanning technicians. The lab I was at used to create profiles based on certain high-volume customers. I initially thought the profile was akin to a preset, but no, the profile was a photo print pinned to the wall with several reference images from the customer that demonstrated how they wanted things like skintones, foliage, sky, and contrast density to look. Thus, the scanning tech would just sit there and look at the reference and try and dial in that look with each new roll being scanned.
I’m going to be real challenging and controversial here by recommending that you stop your quest and change your goals. The days of Frontier scanners, good as they may have been in their day, are over with. With today’s digital cameras, lenses and software, if you know what you’re doing, you can out-resolve and out-class those systems by “country miles”. Focus on what you want the photo to look like (what your mind’s eye tells you) and use the tools first in NLP, then in Lightroom, then if need be in Photoshop, to get exactly the output that satisfies. It’s like the immersion approach to learning a new language - and just as effective.
On the specific point of protecting highlights, in digital imaging they are either preserved or they are clipped. You prevent clipping by watching the histogram before you make the exposure and adjust tones accordingly (remembering that the highlights are the dark portions of the negative) so they don’t climb the wall. Then in NLP, if necessary, you can bring out highlight the embedded highlight detail first using the White Clip control, then Lights then Brightness and so on. If you need to increase exposure so much (because of very dense negative highlight areas) that the shadows (very thin negative shadow areas) will come out too dark (from being exposed too bright), then you have a case for doing a multi-exposure, protecting the extremes one exposure at a time and blending them in Lightroom using Photo Merge > HDR.
