←back to thread

251 points QiuChuck | 1 comments | | HN request time: 0.212s | source
Show context
felixfurtak ◴[] No.45893633[source]
RGB LED backlight is a terrible choice. Wide gamut but terrible color rendering.
replies(1): >>45894230 #
Brian-Puccio ◴[] No.45894230[source]
Can I ask why it will produce terrible color rendering? In addition to commercial scanners that used narrowband trichromatic (RGB) light sources, hobbyists are creating their own RGB light sources to digitize color negative film claiming superior results and putting forward arguments why this is better:

https://jackw01.github.io/scanlight/

(NB: Most film I shoot is slide film, which I’ve been told doesn’t benefit from RGB light sources because it’s intended viewing was projected with a broad-spectrum white light [likely a warmer than daylight (but color temperature isn’t much of a concern for digitizing slides)] so I haven’t dug into this much.)

replies(2): >>45894777 #>>45895676 #
1. roblh ◴[] No.45895676[source]
I got one of these from the latest batch last week and I’m not entirely convinced by it yet. I need to experiment some more but I went back and did a couple rolls from this summer and so far I think the cslite warm setting + negative lab pro results are better and more consistent. I’m still getting some wonky colour casts with it. It’s nice that the control app lets you change the power of each LED colour separately, so that’s the next thing I’m going to experiment with.

I’ll also note that negative lab pro hates negatives that are scanned with it. They don’t turn out at all. If you’re using it, you should expect to be inverting them manually, which is kind of a pain. I was quietly hoping (but not expecting) to still see some of the benefits of it when passing them through NLP.