Page 1 of 2 12 LastLast
Results 1 to 10 of 14

Thread: High Lights scanning problem

  1. #1

    Join Date
    May 2014
    Location
    Cote d'Azur France
    Posts
    109

    High Lights scanning problem

    Hello all,

    I am fairly new to scanning and encountering problems with high lights.

    I am using a Epson V850 Pro scanner, Silverfast 8.8 software to scan 4x5 Tri X negatives in Grey scale 16 bit. Many of my negatives are high contrast with highlights that show detail (just) in the neg and print OK in the darkroom giving a good strong print.

    I run the pre scan, go to the histogram to adjust each side to the limits of the histogram - but I find that the right side slider cannot be adjusted sufficiently to bring the high lights down. So effectively blown out...? Once scanned, I can bring the high lights down in the editing software (Affinity) but just end up with a light grey tone which is flat and not a good look.

    Appreciate any comments on what I may be doing incorrectly or is the just the limitation of the scanner / software?

    Thank you
    Peter

  2. #2

    Join Date
    Sep 2021
    Location
    Scotland
    Posts
    77

    Re: High Lights scanning problem

    I am no scanning expert, but with SilverFast I scan negatives using the HDR Raw output, I then deal with the curve in post, in my case CaptureOne, which offers far better control then SF.

  3. #3

    Join Date
    May 2019
    Location
    Stockholm, Sweden
    Posts
    372

    Re: High Lights scanning problem

    I look at scanning as catching raw data, meaning setting no limits in white and black, acquiring the full scanner range. Or at least leaving an ample headroom not trying to optimize the white and black point .
    Post scanning I do all the processing, either in Capture One or Photoshop

    16-bit, or perhaps more accurate 12 to 14-bits for most scanners, is enough for stretching and cutting the levels. Anyway, that's what the scanner program is doing anyway.

  4. #4
    Resident Heretic Bruce Watson's Avatar
    Join Date
    Sep 2003
    Location
    USA, North Carolina
    Posts
    3,335

    Re: High Lights scanning problem

    Quote Originally Posted by 12pmc View Post
    Appreciate any comments on what I may be doing incorrectly or is the just the limitation of the scanner / software?
    I studied this in a fair amount of depth about 15 years ago. What I found is that B&W negatives optimized for darkroom printing (Grade 2 paper) typically will scan just fine. But if you're never going to print in the darkroom you can usually get a little better results with a little thinner negative. In Zone System terms, I was developing to get highlights (Zone VIII) to max out around a density of 1.0. For darkroom printing, it's more like 1.2. And I was scanning with a drum scanner (ColorGetter 3 Pro).

    The thing to notice about B&W negatives is that they form density through overlapping grain clumps. The grain of course being metallic silver. And the silver of course is completely opaque. This means that scanning, like darkroom printing, is subject to Callier Effect. The more density your highlights have, the more they scatter light, which in turn decreases the contrast in the highlights that are scattering that light.

    So, to finally answer your question, what you might be doing "incorrectly" (if there is such a thing) is feeding your scanner a negative that it is having difficulty "reading through".

    Of course, much depends on your scanner / software, your exposure, your development methods, etc. But as a first guess, I would say to try a sheet of film exposed / developed to have a little less highlight density and see if that's an improvement.

    A word of warning to those who start down this road -- yes, you can go too far. If you take the highlight density too far down, and try to "stretch" the tones back out to make a full range print from the scan of a "thin" negative, you may find that you don't like the result. But the only way to know how far is too far, is to do the experiments and find out. You are the only one who can decide what works, and what does not, for you.

    Bruce Watson

  5. #5

    Join Date
    May 2014
    Location
    Cote d'Azur France
    Posts
    109

    Re: High Lights scanning problem

    Thank you all for your input appreciated.

    And Bruce your comments regarding the every high density neg and the opaque nature of silver make sense and helps me understand what is going on. With negs ranging from Zones 3 / 4 to 6/7 I do not have a problem.

    All part of the learning curve!

  6. #6
    Alan Klein's Avatar
    Join Date
    Jun 2015
    Location
    New Jersey was NYC
    Posts
    1,630

    Re: High Lights scanning problem

    Either you clipped the highlights in the original negative, or there's a problem with SF. (I have never used SF). Have you tried Epsonscan on manual, flat scan with the histogram set at 0 and 255?

  7. #7

    Join Date
    May 2014
    Location
    Cote d'Azur France
    Posts
    109

    Re: High Lights scanning problem

    Alan,

    Thank you, I do not believe I clipped the highlights in the original scan , the problem I have is that I moved the right side maker on the histogram as far as it would go (to the right) and still could not bring the highlights down.

    I have not tried Epsonscan but will do so as you suggest.

    Thanks again

  8. #8

    Join Date
    Sep 2021
    Location
    Scotland
    Posts
    77

    Re: High Lights scanning problem

    Epsonscan doesn’t do anything SilverFast doesn’t, as much as I dislike SF, Epsonscan is a rather mediocre piece of user interface to the drivers, which I suspect exists primarily for in-house testing of them. SF exposes all the functionality provided by the Epson hardware and its drivers and some on the top of that (it would be a great piece of software if they spent some effort on QA and got rid off the crippleware licensing model).

    The scanner can resolve a decent dynamic range, you should not completely loose highlights even at density of 1.2. Make sure you are exporting 16 bit values for BW / 48 for colour, assuming the negatives are not clipped, the loss is most likely happening in the muscling of the raw scanner data into the file format.

  9. #9

    Join Date
    Dec 2020
    Location
    EU
    Posts
    11

    Re: High Lights scanning problem

    I've found that the most important setting in scanning, is giving the correct amount of exposure. This is especially true working with consumer grade scanners.

    These scanners don't have a linear response. The histogram may show no clipping, but if the exposure isn't enough the denser parts (highlights) will be compressed anyway. Then you'll loose detail in the highlights and the file will require a lot of post processing to look OK (if at all possible). Even though the scan didn't 'clip'.

    Sometimes the negative has too large dynamic range and you'll have to choose whether highlights or shadows are more important.

    This is less of a problem working with PMTs, which are linear over a larger range. And this is IMO the largest advantage of using a drum scanner. The response is much more like working with an enlarger.

    In Vuescan you can set the exposure manually with the Epson scanners, but I would be surprised if you can't do it i Silverfast too.

  10. #10
    Alan Klein's Avatar
    Join Date
    Jun 2015
    Location
    New Jersey was NYC
    Posts
    1,630

    Re: High Lights scanning problem

    Quote Originally Posted by _tf_ View Post
    Epsonscan doesn’t do anything SilverFast doesn’t, as much as I dislike SF, Epsonscan is a rather mediocre piece of user interface to the drivers, which I suspect exists primarily for in-house testing of them. SF exposes all the functionality provided by the Epson hardware and its drivers and some on the top of that (it would be a great piece of software if they spent some effort on QA and got rid off the crippleware licensing model).

    The scanner can resolve a decent dynamic range, you should not completely loose highlights even at density of 1.2. Make sure you are exporting 16 bit values for BW / 48 for colour, assuming the negatives are not clipped, the loss is most likely happening in the muscling of the raw scanner data into the file format.
    Epsonscan does at least one thing better than Silverfast. There's no extra charge for it.

    I've been using Epsonscan for ten years on a V600 and V850, and it works great. I do most of my editing after the scan. Check my Flickr link for samples of 35mm, medium format, and 4x5 in BW and color.

Similar Threads

  1. Problem with High Sierra, iMac PS CC 2019 freeze +crash
    By jnantz in forum Digital Processing
    Replies: 3
    Last Post: 27-Feb-2020, 17:55
  2. high contrast problem with Epson V700 scanning
    By Linhof in forum Digital Hardware
    Replies: 6
    Last Post: 15-Feb-2013, 09:39
  3. scanning negs of high key subject
    By Roseanne in forum Digital Processing
    Replies: 2
    Last Post: 3-Oct-2011, 12:55
  4. high quality scanning
    By Carl P in forum Digital Hardware
    Replies: 12
    Last Post: 10-Mar-2010, 15:55

Bookmarks

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •