Recommended Scanning Technique for B&W Negatives

[shel_belinkoff2]

Hi Gang ...

 

Over the next few weeks I'm going to attemp scanning a lot of

conventional B&W negs. I have heard a number of conflicting opinions

on the best way to do this. Most comments center around whether to

scan in RGB or greyscale. Greyscale would be nice as I could save

some space, but if RBG will give higher quality results, I'll bite the

bullet.

 

I have scanned using both methods before, although with the help of a

friend who did most of the work and setup, so I'm still pretty much

uneducated and inexperienced wrt the subtleties.

 

Scanners used will be an Imacon Flextight 626 (I believe that's the

number) and the Nikon Coolscan IV 4000dpi unit, if that makes any

difference.

 

Thanks for any help,

 

shel

[lex_jenkins]

I'm no expert and my scanner is a lowly Minolta Dimage Scan Dual, version uno. But I seem to get the best results scanning at the highest rez in color mode. Then I'll desaturate as a final step to restore b&w images to their original state and to avoid any risk of a color cast.

 

You might find more expert help on the Digital Darkroom Forum, tho' some of those folks also weigh in here occasionally.

[gerald_widen]

Lex I have the Minolta Dimage Dual-Scan III (which I mostly use to evaluate B&W negs). I'm not thrilled with the results. Maybe I'll try your idea to scan in the color mode.

[sander_pronk]

Regarding the Scan Dual III: it seems (to me) that this scanner suffers from massive grain aliasing problems with conventional 400ISO films (HP5+ and TriX). Slightly underdeveloped T-grain films, however, scan very well.

 

I've been using Delta 400 in XTOL 1+2 (13:00 at 20C) and the results have been good (very sharp; detail to the individual pixel level, with reasonable grain). Neopan 400 has slightly less grain, but is a little less sharp; TMax 400 seems a lot less sharp and I don't like the tonality.

 

Any 100 speed T-grain film should give you very good results with this scanner (practicaly no grain and very sharp scans). I liked the greys from Fuji 100 Across most.

 

I'm under the impression that the exact film+dev recommendations are different for each scanner model and vary according to taste, so you should try for yourself.

 

I could post some examples if you want, so you know what you can expect from scanning B&W.

[john_shriver]

The Coolscan IV has not given me any grain aliasing problems in B&W. It has an "optical low pass filter". I've found it to work just fine scanning to "greyscale".

 

I do have to play with the grey scale. Partly, you need to twist things the way paper does, to balance out the heel and toe effects. Also, there are some artifacts where scanners do have a non-linear response in dense areas of the negative (highlights) due to a diffraction issue.

 

The last scanner I used that I had to scan in color and convert was my HP 6350. The less said about that lousy scanner, the better...

[sjmurray]

I would try scanning a black and white in both rgb and grayscale and compare the results. With my Minolta scan dual 3 using Vuescan, there is no appreciable difference, and I have no problems with "grain aliasing" even scanning Tri-X, as far as I can tell. The advantage to scanning using grayscale is you can use 16 bit and the file is much smaller than rgb even in 8 bits per channel. There is a big advantage to using 16 bit because of the much increased number of levels, so when you make adjustments, you don't get gaps in the histogram, thus losing smoothness in the transitional areas.

[roger krueger]

The big advantage to scanning as color neg (NOT color slide) is that, on some

scanners, Vuesan at least (and maybe others) give longer exposures for the green

(2.5x) and blue (3.5x) channels, to eliminate the expected orange mask.

 

You can then combine the red and blue channels to get a scan with considerably less

clipping than a "straight" scan--the red channel has the shadow detail, the blue

channel has the highlight detail.

 

For normal negs, this may not be that much of an issue, but for TMY pushed to 3200

or overexposed Pan F, it's been a lifesaver. Note that this may not work quite as

expected with some staining developers (pyro), since the highlights will actually be a

different color than the shadows.

 

Re: your steup, Nikon Coolscans have a reputation of causing highlight blowouts in

true B&W due to their highly collimated light source. Not sure how the Imacon is set

up, although in general it is a significantly better scanner.

[david8]

Not to be a heretic, but if scanning, I think the C-41 B+W materials are the best.

 

I use a polaroid SS4000+ and found traditional silver based materials very difficult to

scan. In speaking with a B+W pro lab owner, he recommended it, even though he was

shooting himself in the foot.

 

I find the Kodak TCN material to be fantastic in the scanner. Long tonal range and

almost invisible grain. I must admit I used to love the grain of HP5 in Rodinal. Things

change.

 

FWIW

[tonghang_zhou]

<blockquote><i>... but if scanning, I think the C-41 B+W materials are the best.</i></blockquote><p>

Is this any better than C-41 color negative scanned and converted to grey scales? I ask this because color negative film is more commonly available.

[david8]

I like to scan XP or TCN rather than color C41, but .... if....

 

 

1. You have only 1 dye layer with C41 B+W, and thus a finer "grain" with C41 B+W, if

that's what you're going for...

 

or

 

2. You can get a feeling of "traditional" B+W by using high speed color and going

B+W, but it's not the same, just the feeling. The other issue is mastering the channel

mixer in order to get the right B+W image: takes some work and practice. I would

suggest not a simple gray scale conversion...

 

Again, so no one hates me too much, nothing beats the real thing if you have the

facilities to really do it.

[eric merrill]

<i>I use a polaroid SS4000+ and found traditional silver based materials very difficult to scan.</i>

 

<p>

 

What was the problem? I have a Polaroid SS4000 and have no problems at all scanning traditional B&W films. <a href="http://hamrick.com/">Vuescan <http://hamrick.com></a> is the best thing since sliced bread, in my opinion. I'm wondering if your scans were attempting to use the IR channel for "cleaning." That's a feature the ss4000+ has that mine doesn't. And it won't work with silver-based films.

[david8]

Oh I don't know Eric...

 

I use silverfast and consider myself very adept with it and the scanner, but silver

based B+W just seems to loose something compaired to the traditional darkroom:

perhaps its me. I find the lower contrast and "density" of C41 B+W far easier to cope

with in the scanner and to manipulate (not to mention the thinner film and flatness of

the negative as well).

 

I wish I still had a darkroom and the rest, don't miss the smells and mess to be

honest....

[robert simpson]

Try this: Scan 16 bit greyscale and use multipass scanning, if your scanner has it. Compare to a single scan 16 bits and then another at 8 bits. Then cut the resolution in half for anther 8 bit. Then go into your image editor for a close-up to check the things that are important to use - grain, tonality, shadow and highlight detail.

 

Multipass provides more data, or so I'm told. In 16 bit you can also make adjustments like levels, etc. during the scanning process without losing real data the way you would in 8 bit.

 

Let us know how it goes.

[george_collier]

I have scanned only prints, with a very good flatbed scanner. I

tested extensively when I first started doing this, and found:

Grayscale scanning seemed sharper, but marginally. RGB

recorded very light (highlights in the print) detail better, although,

again, marginally. A good friend of mine who is an expert

scanner told me that the reason for both, in his opinion, is that

the triple pass for color will pick up some subtle values that the

single grayscale pass might miss, but that "registration"

between the three color scans might not be perfect, resulting in a

slight loss of sharpness (verrry slight).

I went with RGB because I'm scanning portfolio prints, and the

color scan will pick up subtle color casts from toning, and the

difference in sharpness will probably not be supported by most

output devices.