I have been working exclusively in chromes, having them drum scanned and then printed on a Lightjet 5000 with excellent results. However, I keep thinking that it would make sense to shoot negative film with the large contrast range (i.e. lower contrast) and then have them scanned and adjusting the contrast in Photoshop before printing. I know that there are two problems with scanning negative film. First, it's almost impossible to get an exact color match, in part because there's no color original to compare it to. I am willing to live with that shortcoming. Second, there are problems with resolution and getting a good scan because of interference with the orange layer. Has anyone found a good solution to drum scanning negatives? What service bureau are you using? How would you rate/judge the results you've been getting? Thanks. Howard.

-- Howard Slavitt (info@naturelandscape.com), February 21, 2000

Answers

There are other problems with scanning negatives Howard. I've posted this twice on photo.net, and will post it again here in hopes more people will see it.

Color nagative films, although they can capture a wider contrast range the color slide films, cannot store as wide a contrast range. If you compare color slide and color negative films on a densitometer you will see there is a wider range between d-min and d-max on the slide films.

No matter what you do, you cannot not get as good a scan from color negative film as you can from color slide film. That doesn't mean the scans from negative films are bad, just that the scans from slide films will beat them. Given that, why bother with shooting negatives if your intent is to drum scan them?

-- Darron Spohn (dspohn@photobitstream.com), February 21, 2000.

Howard:

Check with Bill Atkinson... for a while, he used Pro100 (apparently on a trip to New England based on the images) but he is back to Fuji Velvia and now I understand Provia 100F. Joe Holmes recommended using Astia to get a wider contrast range on film, but also suggested NPS works will by having a Pro PhotoCD made (although the max resolution is not real high there). I had Bill Nordstom do a VHC negative that came out fine, but getting the colors right was tough.

-- Glenn C. Kroeger (gkroeger@trinity.edu), February 21, 2000.

Howard, I have considered the same for quite some time. Just a thought, have you considered making a print from the negative and scanning the print? Its not ideal, but in when confronted with a high contrast scene this may be the best solution. BTW, did you get your web site up yet? If so, please forward the address to me, i would love to check your work...

-- Bill Glickman (bglick@pclv.com), February 21, 2000.

Bill:

I'll answer your question for Howard. The answer is as easy as looking at Howard's e-mail address! Check out www.naturelandscape.com.

-- Greg Lawhon (glawhon@unicom.net), February 21, 2000.

If results from scanned negatives are noticeably inferior to slides, then the scanner operator isn't doing it right.

I'll admit that negatives have much more noticeable grain than slides, but the colour rendering should be as good. You might try shooting a colour swatch/grey-scale in the same light and with the same film batch, as part of the session to give the scanner operator a sporting chance.

Darron wrote: "If you compare color slide and color negative films on a densitometer you will see there is a wider range between d-min and d-max on the slide films."

That's the whole point Darron. A real-life brightness range is far too wide to capture on any film, reversal or negative. The negative has a lower density range, but also a lower gamma. This means that very bright tones are still rendered as density changes on negative, past the point where reversal film has reached base density.

A consequence of the lower gamma is also to have lower inherent colour saturation. This is compensated for in normal wet printing by a high gamma in the printing paper, but digital scanning doesn't automatically do this. The colour saturation has to be boosted artificially by software in most cases. Not a problem if you work with 36 bit files up to the final crop.

I notice that most mention is of Fuji films. IMHO Fuji still have a lot to learn about making a decent colour negative emulsion. Purple skies and Lime-green grass are not my idea of realistic colour.

-- Pete Andrews (p.l.andrews@bham.ac.uk), February 23, 2000.

You're reading it wrong Pete. The point is color negative films, although they can record a wider density range, cannot output as wide a density range as color slide films. I got my information from a post Dan Sapper made on the Advanced Photography Forum. The bottom line is that slide films scan much better than negative films because slide films can output a wider density range than negative films.

-- Darron Spohn (dspohn@photobitstream.com), February 23, 2000.

For those who are interested, here is the link to Dan's explanation.

-- Darron Spohn (dspohn@photobitstream.com), February 23, 2000.

The contrast range from Dmin to Dmax is only half the story - you also need to consider the tonal resolution of the film within those bounds. In terms of pure information storage it doesn't matter if negative film has a smaller range of physical densities, provided that those densities can be varied accurately enough to capture tonal subtleties. Printing then magnifies those small tonal differences back up to a level our eyes can distinguish.

Negative film can capture a wider range of scene brightnesses, and the built in masking reduces cross-coupling between the colour channels. If the tonal resolution is up to the job (and I think it is) it is intrinsically a better medium on which to record information.

In order to measure the optical density of negs as well as possible you need to adjust the photomultiplier gain - ideally, to different levels for each colour. You also need to ensure that the pre-ADC analogue electronics have sufficient noise rejection. Do current drum scanners do this, or do they have a fixed analogue gain optimised for for the wider density range of slides?

In any case, because a negative stores a wider range of scene brightnesses, it is often scanned so as to preserve detail that would already have been lost on any slide of the same scene. With a fixed number of bits per pixel this guarantees that the information common to both will be reproduced less effectively in a scan of the neg.

Finally, it is worth pointing out that lasers are cheap and ubiquitous enough that it would be possible for the film manufacturers to write a calibrated grey scale or colour swatch onto the non-image portion of every sheet and roll of film. If you realy care, you could do it yourself with a film-to-film contact printer, a reliable light source and a step wedge.

I realise that a lot of this is theoretical, and that if you don't want you build your own drum scanner you have no choice but to work with the tools that are available. However, these discussions end up implying that slides are somehow better than negatives. To my mind, it's the total workflow from camera to print that currently favours slide film, and for reasons which have little to do with the emulsion.

-- Struan Gray (struan.gray@sljus.lu.se), February 24, 2000.

On the contrary Darron, what Dan says simply paraphrases what I said in my last posting. Dmin and Dmax have little to do with the real-life brightness range that can be recorded on film: That is a function of (Dmax - Dmin)x Gamma, to a first approximation. It is Gamma (delta Density/log Exposure) that is the most important factor.

Struan:- Great idea about the built in greyscale/colour patches. I wonder if that little green stripe down the edge of 35mm film serves any purpose?

-- Pete Andrews (p.l.andrews@bham.ac.uk), February 24, 2000.

Sorry! That equation should have been (Dmax - Dmin)/Gamma. I was going to put Gamma to the minus one, but then realised I didn't know how to HTML superscripts.

-- Pete Andrews (p.l.andrews@bham.ac.uk), February 24, 2000.

Agfa now makes arial camera film (X 100) without the mask (orange color). This may become available in other formats but is currently available in 9" by 125' rolls (could be cut to 8 X 10). This film also has a surface that helps with Newtons rings on flat beds.

-- Paul Weissman (opticalresinc@aol.com), March 18, 2002.