Color issues: Vuescan vs Nikon Scan?

I

Ihmemies

Member
Local time
2:01 PM
Joined
Sep 3, 2007
Messages
18
These are my personal opinions and experiences, so I won't repeat that phase below in my text :rolleyes:

The software I use to scan film:
- Slides: Nikon Scan
- B&W negs: Vuescan
- Color negs: ???

Firstly.. I like Nikon Scan's results colorwise. They look good and natural, a good interpretation of how the film might have really captured color.

For some reason or another slides don't seem to need infrared cleaning, while with color negs it's mandatory. I would scan color negatives too with Nikon Scan if the ICE algorithms were not broken in the software. Vuescan produces much, much nicer-looking scans with its implementation of infrared cleaning.

Thus the problem: scanning color negatives with Nikon Coolscan V. Because of the ICE issue I'd use Vuescan if the software didn't produce bad color.

With Vuescan especially reds, less visibly perhaps greens and other colors too differ in a bad way from Nikon Scan's results.

In these two examples Vuescan's result is left, Nikon Scan's right:
http://hakkarainen.kuvat.fi/tempo/vuescan_vs_nikonscan_color.jpg
http://hakkarainen.kuvat.fi/tempo/vuescan_vs_nikonscan_color2.jpg

Especially the red shirts look garish and out of place. Same noticeably happens with other bright colors like orange too.

I'd like to shoot slides only, but dynamic range, price, bigger variety of choices and other factors make me shoot negatives too.

How to solve the problem? I've gone through various settings in Vuescan, tried to reset them etc. Nothing seem to produce identical results with Nikon Scan colorwise. Is it even possible to get good-looking results with Vuescan?

Those differences might look like minor and innoticeable, but they disturb me to no end and make me wonder why I'm even bothering with these ancient film scanners & software.

Edit: I've also tried Silverfast, but the results were identical with Nikon Scan, so there was no point using SF.
 
Last edited:
I use Nikonscan because I like the color more and ICE works. I'm surprised you find Vuescan's IR cleaning more - I thought it was crap when I tried it. I have to admit I haven't tried it in a couple years. What exactly is wrong with Nikonscan and ICE?
 
I did like Silverfast when I used it with an Epson flatbed. Haven't used it with the Nikon as it's too expensive IMO.
I have to wonder, though, how are the results with SF identical with those of Nikon Scan? SF has loads of different profiles and usually I just use the one that I like best (not necessarily the ''right'' one).
 
I don't really know about Silverfast's color, I only looked at the infrared cleaning system and that seemed to be the same as with Nikon Scan. That would be logical since both are called ICE and probably licensed from same vendor, while Vuescan uses its own homebrew system (which was "recently" redone, maybe 1-2 years ago).

I know that the grain produced when scanning film is not the same grain as in film, it's just an approximation of what the grain might look like. Nevertheless, I like that the grain scanner produces stays "untouched".

Nikon's ICE algorithm has two sides:
- It removes scratches and dust more effectively than Vuescan
- It smudges grain to obscure mess while removing dust & scratches

Instead of only removing what it should be, it looks like it works more like Photoshop's "remove dust & scratches" feature or dust removal feature of some flatbeds where there isn't real infrared light...

Vue: http://hakkarainen.kuvat.fi/temp2/vuescan_noIR_lightIR_medIR_hvyIR.jpg
NS: http://hakkarainen.kuvat.fi/temp2/nikonscan-noICE_stdICE_fineICE.jpg
SF: http://hakkarainen.kuvat.fi/temp2/noICE_stdICE_fineICE.jpg

Also it feels like Nikon Scan applies some kind of basic USM (or something like that) to scans, so that's why I scan b&w with Vuescan since it doesn't seem to do the same.
 
Last edited:
I have similar results with a calibrated Nikon CS9000.

I can't get Vuescan scans to produce realistic colors and the scans usually need a large amount of color correction to remove color casts and produce better colors. But I have seen very nice Vuescan color scans from other people here in RFF so it must work to some level. Nikonscan works much better for me both for slides and C41 negs.

I much prefer Vuescan over Nikonscan for B&W film.
 
I haven't scanned many slides, but use VueScan for all negatives. I've been happy with my colors when I have locked in the black level film base and then made the right white balance selection. I particularly like my Ektar 100 scans. I use Vuescan with a Nikon 9000ED and Epson V-750 Pro.
 
I've been using Vuescan for many years, probably since 2002 or 2003 if not earlier, and have taken the time to understand and explore all the options. I lock the film base and exposure, I calibrated the scanner using Vuescan and a set of reference slides and still don't find it satisfying for color.

While it's true that Nikonscan doesn't have the same flexibility, it still produces colors that are a lot closer to the original scene than Vuescan does (for me). I also don't have many of the problems others report about Nikonscan's crashes.
 
What happens when the color temperature of the ambient light changes? It seems to me that the issue is one of process. All of these programs capture a raw data stream. It is up to us to setup the program properly to get the end result. Now, it may be that I'm far less critical in what I demand and it may be that unknown to me VueScan does not have adequate tools to critically capture the original film. In fact, I'm looking for colors and exposure that I like, which may not be an accurate representation of the subject. I reset my white point every time that the lighting changes. If I shoot at the same location and some shots are in shadow and others sunlight, or some at a different hour, then I reset the white balance. I also do this for every film, even from the same batch and shot at the same time. That said, I also pass every capture through PS for curves adjustment and many time downsizing prior to storing in Aperture.
 
You must log in or register to reply here.
Back
Top Bottom