Color in center and edge

Old film can be difficult to handle and I tried NLP with transparencies that are about 70 years old…with fairly good results:

As for negatives, most of the films I convert are between 30 and 50 years old and most of them convert fairly well. Overexposed or thin (underexposed) negatives are more difficult to convert, but results can be improved if e.g. the overexposed (“blown”) parts are excluded from NLP’s analysis - by cropping said parts off before converting. It takes a bit of time and exercise to find out, when and what crop to apply though.

I mostly set “Border Buffer” to something between 15% and 25%, also in order to not have to crop before converting, and in order to focus NLP on the “important” central parts of the images.