lubitel said:
ICE seems to work on BW c41 film, just not on real B&W
That's generally the case (although I note that another poster above noted that it seems to work for him; not sure what's going on there.)
The reason it has trouble with silver-based b&w is that these ICE-type automated dust removal systems are based on an infrared scan of the film.
Think about it: If you just scan the film for visible light, the scanner would have no way of knowing whether that tiny dark speck was a fleck of dust or a small, very bright object in the actual image (such as a candle flame.)
But since chromogenic films form their image from dyes, which are designed to allow IR light to pass through (so they don't heat up during enlarging or projection) using an IR scan makes it easy to tell the difference. IR will go right through the dyes forming the image of the candle flame, but the dust speck (being essentially a tiny "rock") is opaque to IR, allowing the dust-removal software to tell the difference.
The problem with silver-based films is that since the image is formed of metallic particles, these also block infrared very efficiently; there's no way for the dust-removal system to distinguish between an IR-opaque dust speck and an IR-opaque dense highlight in the image.
It may be that some of the dust-removal systems now also try to distinguish dust from image by shape recognition or other means, but I think you'll find that even those that sort of work with silver-based b&w still work better with chromogenic films.
If anyone knows different, I'd be very interested to hear the theory behind it...
