I'm struggling with scanning films with a bit dense base like HP5+ and Delta3200 in 35mm format. They print well on normal paper, but the black and shadow tones are compromised, probably because of the dense base. I am playing with the black threshold, contribution of red channel etc., but does anyone have a good idea of what is causing the problem and how you deal with it?

When I scan APX25 or TMX in 120/220 format, I have absolutely no problem related to the film base density. I use Epson Expression 1600 with its transparency light unit and VueScan program on Linux platform.

-- Ryuji Suzuki (rsuzuki@rs.cncdsl.com), April 27, 2002

Answers

Without seeing the scans I can't say for sure but I suspect the problem is related to grain aliasing rather than the film base. There is a place where the grain size and scanner resolution play havoc with your ability to get a good scan. That precise place is hard to nail down but I think you've found it. The scanner you're using is marginal for making good 35mm scans from any film type. A dedicated 4000 ppi film scanner will give much better results. Making real prints and scanning them on your flat bed may give the very best result from 3200. Real grainy coarse films are hard to scan on any scanner. The best results you can get scanning 35mm negs is probably from C41 process films like XP2, next from carefully exposed and processed films like TMAx 100 or Delta 100. You want negs that are thin, not contrasty and with full information in the highlights and shadows.

-- Henry Ambrose (henry@henryambrose.com), April 28, 2002.

Hmmm, interesting!
Can you give us a bit more info Ryuji?
Are you doing the scans in full colour mode?
Are you scanning in B&W negative mode? (Never a good idea, IMHO, because it uses only one channel of the scanner, and leaves the software to automatically select a black and white point)
Have you tried a RAW mode scan to see if the shadow detail is captured by the scanner at all?

The only reason I can think of at the moment why this should happen, is that the greenish base of the film is acting as a strong filter to one or more of the scanner RGB channels.
Or, on second thoughts, it's just as likely that the black-point detection in the software is confused by the base density, and consequently is setting the black point too high.
I think a colour RAW scan, followed by inversion, level correction, and desaturation in Photoshop should sort the problem.

-- Pete Andrews (p.l.andrews@bham.ac.uk), April 29, 2002.

My Epson Perfection 1640SU Photo scans HP5+ and Delta 3200 really well. In fact I'd say HP5+ is the easiest to scan of the films I use. I always scan in 14bit B&W as my experiments with scanning in colour have not produced anything different. So other than saying , all is not lost, I can't really add anything useful as I haven't had to work thru a problem.

-- Nigel Smith (nlandgl@unite.com.au), April 29, 2002.

The biggest complaint is mottled shadows and blacks. It is unlikely that grain is solely responsible for it because D3200 (120) developed in Microphen didn't have as much problem at the same 1600dpi resolution. This D3200 in 35mm is processed in my metol-ascorbate formula tweaked for T-grain films, producing finer grain than Microphen. On prints shadows are as clean as Microphen version at similar magnification.

When I exclude red channel's contribution to the final image, it seems that I lose readable Dmax range without improving the clarity of the shadow.

I have manual control over the black and white points as well as contrast on my scan software, and I can make more shadow to be black by playing with them, but it doesn't make my shadow cleaner.

I tried to wash out the base dye by soaking it in 2% sodium sulfite solution and then rinsing and drying. I'm not sure how much it helped (can't tell by looking or scanning) so it's not a solution.

I put some minimally processed scans from 35mm D3200 on my web server (unlinked from anywhere). I kept rather large images in high JPEG quality so that you can see it better, so they are about 300KB each.

http://rs.cncdsl.com/Photography/Mochi?.jpg
where ? is 1, 2 or 3.

Continuing to seek solutions... besides switching films or the scanner. To me 35mm is an usable toy and I don't want to invest much resource on it :-)

Note: Those pictures were taken with original black Hexar 35mm f/2 at f/2.8 with 1/15 to 1/60 exposures. The bokeh (out-of-focus blur) is very nice, if you haven't figured out what that term means. (I've seen several confused people when Photo Techniques published an article on the topic.)

-- Ryuji Suzuki (rsuzuki@rs.cncdsl.com), April 29, 2002.

I forgot to say that changing the resolution didn't change much. When the scanner lowers the resolution, it also reduces the optical resolution (lower the spatial cutoff frequency) to avoid aliasing problem.

I realized that those JPEG files don't look as nearly good as other JPEGs of similar file sizes. I think the mottled tones act as "details" to be preserved by the compression algorithm and they inflate the file size!!

-- Ryuji Suzuki (rsuzuki@rs.cncdsl.com), April 29, 2002.

OK. I think there might be two things at work here.
You say cutting the red channel reduces the dynamic range of your scanner? In that case, I suspect that Epson are guilty of using the infrared output of the scanner's fluorescent tube to gently 'fog' the CCD sensor, and so reduce the effective Dmax of the film. This is a cheap trick to make the dynamic range of the scanner seem greater, and I hope I'm wrong about it.
However, if that is the case; the infrared absorbtion of the film base, or residual AH dye, might be responsible for the mottling that you can see, but I must admit, it just looks like grain to me.
Do you know what the Dmax of your negs is? They shouldn't really be much more than 2.4D, and that density should be easily scannable without resorting to any IR trickiness.

The second thing that might be responsible is just plain underexposure. TMZ 3200 isn't really 3200 ISO you know, it's more like 1200 ISO. Exposing it at 3200 will just put most of the image on the toe of the film's curve, and this'll naturally give some strange looking scans. Remember, a scanner is linear, and doesn't have the corresponding 'shoulder' of photographic paper to compensate for the toe response of the film. You should really be applying a curve to scans as a matter of course, to get the same result as a photographic print.

I might well be wrong on both counts above, and the problem could be due to something else entirely.
Anyway. Best of luck sorting out the problem Ryuji.

-- Pete Andrews (p.l.andrews@bham.ac.uk), May 02, 2002.

Your scans look like I thought they would. I think that the reason its not so noticeable with your 120 scans is that the scene is rendered larger on the larger film, so in effect the grain is smaller in relation to the image details. I also think Pete is correct about underexposure and the linear nature of scanners. You will need to do some curve corrections either after they are scanned or in the scanner software. More exposure and a little tweaking will help for sure. Have you printed these scans? I think you will find that they will make nice small prints even with all the noise in the shadows. Still the best solution is finer grain film.

-- Henry Ambrose (henry@henryambrose.com), May 02, 2002.

Thanks for the responses so far, points well taken, but they are already taken care of before posting this question.

I exposed D3200 at 1600, and the neg I showed was from the clip after doing some experiments with formula and processing time, and therefore I think the neg is pretty good. Not too dense or thin, DU probably about 2 range. My scanner can go up to 3.3 on catalog spec, and it handles very dense HP5+ very well. It was a good cost performer when I bought it (spring 2000, about $1k). Scanners in 2002 market has improved Dmax but I think 3.3 is still good considering my negs aren't that dense. The dynamic range is still much wider than any paper I know of. My requirement was to scan medium format films so there aren't many options below $2000 either.

The scanner software I use (VueScan) has a set of film curves, and I just disregard the name they have given to each curve but try to find one that matches the best with the film-developer combination. The software also adjusts the light output (exposure) for scanning. This software is inexpensive, and the guy who makes it is probably not an expert (or as picky as us) in b&w photography, but the way it handles issues like this is pretty intelligent. Again, changing the curve or black point threshold make a big difference in how the image looks but they won't eliminate the mottled shadow unless I make very high contrast image, of course.

Regarding the IR fogging, I think you are talking about different problem. READABLE D-max is determined by combining all color channels (usually RGB) and cutting R simply cuts roughly 1/3 of total readable optical power. However, I'm talking about low opacity region where the optical density is nowhere near the scanner's limit.

Well, if it looks like plain grain to both of you, maybe there is no harm in showing them to other people :-)

-- Ryuji Suzuki (rsuzuki@rs.cncdsl.com), May 02, 2002.