Blotchy blacks on scanned image

[adan]

Advertisement (gone after registration)

Sorry I took so long to get back to this thread - miscellaneous distractions. Here's how I address shadow "noise" and grain these days.

 

First, I try to give my negatives what Ansel Adams called "full" exposure. AKA "expose for the shadows." I try to expose so that every part of the picture has some density above film-base-plus-fog. Or as Steve (250swb) says, "get some more detail/density in the shadows...". So that the blacks within the picture area are a charcoal gray compared to the fb+f density in the raw scan. Sometimes that means downrating the ISO as much as a stop - depends on the specific film. That means more silver grains overlapping and blending together in the "blacks" to eliminate the blotches.

 

The reason that is "controversial" in this setting is that, when 35mm first appeared 80+ years ago, "full" exposure on the thick emulsions of the day resulted in a lot of light scattering in the gelatin and a significant loss of resolution, as well as graininess. Pretty much a non-issue with large and medium format and their "adequate" lenses, enlarged only 2-4x, but a real problem on the "miniature" negatives of 35mm, with potentially much more resolution available from the smaller high-precision lenses, and enlargements of 8-11x or more. From the 30s through the 50s, practitioners strove to get "just barely enough" exposure on small film, to keep the resolution as high as possible. And played with "surface" developers, that could not penetrate the emulsion, but only affected the top layer of atoms, and other chemical techniques. And again, the process of optical printing onto paper, especially through two layers of gelatin, tended to fog over and hide the stand-alone shadow grains.

 

Modern emulsions have been made thinner and thinner over the decades since 1960, so the need for minimal exposure has become less important. Although full exposure on 35mm can still cause problems, especially when one roll contains scenes of widely-varying contrast. A sunlit scene with deep low-sun shadows and white snow (or snow-white clouds) will always be challenging. But by the 1970s, a lot of photographers were shooting Tri-X at 200 again (its original box speed) to get solid exposure in the shadows.

 

As it happens, I scan my film (Hassy negatives) with black borders, which gives me (and Vuescan) a standard against which to adjust the rest of the picture. I Vuescan, using Auto Levels exposure setting and zero scan clipping, to a .dng instead of TIFF, and do most of my adjustments then in Adobe Camera Raw, just as though I had a digital-camera original. And then I just move the "blacks" slider left until all of the border (including the random noise grains), and just a few small parts of the image itself, clip to black.

 

See attached images, which show a picture, and how little of the picture itself actually clipped to black, when the unexposed borders were clipped to black. (I clipped the blacks 45 units in Camera Raw).

 

Occasionally I miss the exposure on the low side, and get larger swaths of "empty" negative. And they scan blotchy. And I try to do better.

 

Other techniques - I stick with lenses "of a certain age" for their lower macro-contrast. 1980s Mandler lenses on Leica, and Hassy C T* rather than newer CF lenses. Allows for a fuller shadow exposure without pushing the whites off the top of the latitude scale. I also make use of Photoshop's "Smart Sharpen" on images with weak shadow exposure - this filter is like "Unsharp Mask", except that the sharpening can be "faded" in the shadows, to de-emphasize those solo grains or clumps that want to show as blotches.

Welcome, dear visitor! As registered member you'd see an image here…

Simply register for free here – We are always happy to welcome new members!

[pico]

Have a look at the roughness of the detail in the dark t-shirt compared with surrounding areas. If I can get smooth brighter tones why not smooth darker tones? 

 

The shadows of the dark shirt appear to be overly aggressive shadow recovery; noise.

Edited December 2, 2015 by pico

[Advertisement]

The Leica Q3 is available for pre-order at B&H Photo Video and Adorama

[philipus]

Thanks for posting this Andy. It's a method I've not previously tried with my b&w.

 

Apologies for hijacking the thread, but why DNG instead of TIFF? ACR can handle both. Oddly, on my system using the settings you mention, the DNG is very bright in ACR whereas the TIFF is not, see below.

 

Thanks in advance

Philip

 

As it happens, I scan my film (Hassy negatives) with black borders, which gives me (and Vuescan) a standard against which to adjust the rest of the picture. I Vuescan, using Auto Levels exposure setting and zero scan clipping, to a .dng instead of TIFF, and do most of my adjustments then in Adobe Camera Raw, just as though I had a digital-camera original. And then I just move the "blacks" slider left until all of the border (including the random noise grains), and just a few small parts of the image itself, clip to black.

 

See attached images, which show a picture, and how little of the picture itself actually clipped to black, when the unexposed borders were clipped to black. (I clipped the blacks 45 units in Camera Raw).

 

 

 

Welcome, dear visitor! As registered member you'd see an image here…

Simply register for free here – We are always happy to welcome new members!

[adan]

Well, I'll have to give TIFF a try - although I am so used to .dng from Leicas, anyway - and see what other differences there are. How "destructive" are changes to TIFFs via ACR, compared to raw .dngs?

 

I also get "light" tone mapping, sometimes - but a touch of the exposure (-), white (+) and contrast (+) sliders, and occasionally tweaking the ACR curves to increase the slope on the highlight half (darken midtones) - fixes that.

[pico]

Perhaps this post should be elsewhere. I have an old flat-bed scanner suitable only for scanning prints and I still have a devil of a time making scans that look like the prints. I print small, regardless of the original format. Nothing looks right in the scans. Shadow subtleties turn into broken barriers, grain does not show. Sure, the monitor can evince more dynamic range than the print, but it does not.

 

I am not giving up yet, but paying a lot of attention to posts here.

[philipus]

Thanks for replying Andy. I've also wondered about how non-destructive ACR really is. I've always assumed that the editing is non-destructive (provided it is done only within ACR) because even if I press "Done" after having pulled on various sliders, re-open the image and press "Default", the image goes back to square one.

 

The little I know about Photoshop is that it is possible to edit in a non-destructive way by using for instance layers (which won't touch the underlying image data). But at some point Photoshop will become "destructive", that is, when one flattens the image. Perhaps there's a way around this, I don't know, but I don't think so. So, it seems to me that if one only edits in and outputs from (as in presses Done or prints or exports to another file) ACR, the image data shouldn't be affected. Even JPGs open in ACR so given the constraints of that file format ACR might be the better tool for editing such files.

 

I'll play around a bit more with the exposure slider in ACR on DNGs.

 

br
Philip

 

Edit: I found this discussion over at Adobe.

 

Well, I'll have to give TIFF a try - although I am so used to .dng from Leicas, anyway - and see what other differences there are. How "destructive" are changes to TIFFs via ACR, compared to raw .dngs?

 

I also get "light" tone mapping, sometimes - but a touch of the exposure (-), white (+) and contrast (+) sliders, and occasionally tweaking the ACR curves to increase the slope on the highlight half (darken midtones) - fixes that.

Edited January 21, 2016 by philipus