A primer on scanning using a "raw workflow"

Discussion in 'Scanning, Enlarging, and Printing' started by pdh, Mar 25, 2015.

  1. pdh

    pdh FF Regular

    31
    Mar 25, 2015
    UK
    One of the things I frequently read is that scanning film is difficult, something of an art, and takes a long time to master.

    I believe that this does not have to be true.

    What I describe below is a means of capturing and processing negative scans that enables one to avoid having to learn very much about scanning at all, leaving all adjustments to tone and contrast to the "post-processing" stage.

    negfix8
    is a script, available for Windows and Linux, that uses ImageMagick to invert a TIFF file created by scanner software into a form usable for import and manipulation by post-processing software.

    Clearly, this approach will not be for everybody, as we all have our own ways of working, and some will prefer to adjust contrast & colour in the scanning stage rather than learn the intricacies of PS, Gimp, LR or whatever. Nevertheless, for those of us who already use digital cameras and use a "post-processing" package, this has the advantage of keeping all the learning and processing in one place.

    I'm not claiming any originality for this method - negfix8 itself, and the particular approach described, was developed by a Flickr & APUG member (his usernames are JaZ99wro and JaZ99 respectively), and he must take all the credit.

    Nor am I suggesting that it can do things that a knowledgeable user of PS (or similar graphics editor) cannot do. But for those of us for whom PS (etc.) remain a dark mystery, it can be a neat way of doing things.

    It does require however that the user is comfortable with using the "command line" of their operating system, that the user can install the ImageMagick package, and that the scanner software one uses can generate a "raw" unaltered negative scan in tiff format

    I originally started using negfix8 because I found both Vuescan and Silverfast overly complex and with badly coded interfaces. Trying to learn how best to use either from forums produced (as so often) conflicting recommendations. I wanted something simple.

    Using negfix8, I have two major decisions to make in Silverfast: what to call the output files, and where to put them. No fiddling with the image until I have imported the final base image into LR, where I know what I'm doing.

    All my 35mm negatives are scanned and processed in this way. I do not print my digitised negatives, or make use of them for alt-processes, so this method works fine for me for sharing my film snaps online. If I want "hard copy" I go into the darkroom


    The gist:

    The scanner software is set up to generate unaltered "raw" scans, without any adjustment to contrast or other image characteristics; the resultant tiff file is an image of the negative - not inverted to a positive.

    Such "raw" scans can be generated by both Silverfast 6 & 8, and by Vuescan.

    The tiff file is processed using a DOS / Linux script called negfix8, which uses a graphics processing package called ImageMagick to invert the image to positive and set black & white points appropriately.

    Once the script is started, there is no further user-intervention required. negfix8 manages all the parameters that need to be passed to ImageMagick, and again no user intervention is required other than to initially install ImageMagick.

    The output from this process is are positive tiffs that can then be further manipulated in the "post-processor" of your choice.


    Comments & Caveats

    In my own case, I scan a full roll of negatives "raw", using a Plustek 7600i with Silverfast6 in "16-bit HDR" mode; I then process all the resultant tiffs using negfix8 (in a single pass - one does not have to run negfix8 on each individual frame); the positive inverted images are then sucked into Lightroom for cataloguing and editing, where "auto adjust" is usually quite enough to get into the ballpark for a usable image.

    The argument can be made that as the inversion can be done in LR, then inserting the extra step of using negfix8 is superfluous, but my experience of trying that is that negfix8 provides a much greater dynamic range to play with compared to the linear inversion of LR.

    I don't doubt that a sophisticated PS/Gimp user might be able to create a set of actions to perform a similar process to negfix8, but as I say, I have no skill or knowledge in that area; it might be a good idea for a another thread however.

    negfix8 can be downloaded from here: https://www.sites.google.com/site/negfix/, where there are further details of installation and usage.

    ImageMagick can be downloaded here:http://www.imagemagick.org/

    I believe that negfix8 and ImageMagick can be used on the Mac platform, but as I am solely a Windows and Linux user, I cannot describe how that is done.

    If you think that it's too much work, or that everything negfix8 can do you can do better in PS or Gimp or Silverfast or Vuescan, it would be really helpful to start a thread (or threads) describing those alternate methods.

    I hope this is of some aid and interest to others. I can't offer a description of the innards of negfix8 or what it is doing as I long ago lost track of how to read even simple code, but I can try to answer any simple questions if any points in what I've written is unclear in some way.
     
    • Informative Informative x 2
    • Like Like x 1
    • Useful Useful x 1
  2. pdk42

    pdk42 FF Rookie

    15
    May 1, 2015
    Paul
    Thank you - that's a very helpful post. Now all I need is a scanner!
     
Loading...
Similar Threads Forum Date
Scanning Prints vs Negatives Scanning, Enlarging, and Printing Mar 3, 2016
Using a Mirrorless + Macro lens to scan negatives Scanning, Enlarging, and Printing Nov 25, 2015
Post Processing Film Scans Scanning, Enlarging, and Printing Jun 16, 2015