DIGITAL vs. FILM (Round 2)

Discussion in 'Digital Photography' started by Annika1980, Nov 22, 2003.

  1. Annika1980

    Annika1980 Guest

    Well sports fans, you might have a hard time believing this but in a few days
    your favorite Totally Digital photographer will be the owner of a new Minolta
    Scan Elite 5400 scanner.

    Perhaps now we can conduct some further testing to add to the Film vs. Digital
    "discussion" (debate, war, feud, etc.).
    My previous efforts showed the superiority of the 6MP digital images, yet many
    of the Luddites cried that my 2820dpi SDII scanner wasn't of sufficient enough
    resolution to show film's advantages (real or imagined).
    Well with this new 5400dpi monster, there should be no question as to the
    results, whatever they may be.

    Of course, I'll be going into these tests with a Totally Unbiased viewpoint so
    that the comparisons can be done on a level playing field. If someone thinks I
    might somehow skew the results by doing poor scans, I could even let my good
    pal, Alan Browne scan them since he has a 5400 as well.

    So while I await the arrival of the "Fabulous 5400" maybe some of you can offer
    some suggestions of how to best do the testing. I'll be using the Fabulous
    EOS-1V for the film images and I suppose I'll be testing mainly transparencies
    since that is mainly what I'll be shooting, when I give the Totally Digital D60
    the day off.

    I've got a roll of the new Velvia 100 as well as some Astia slide film, and
    I've recently shot a few rolls of Velvia 50 as well. Perhaps I'll shoot a roll
    or two of print film (Reala) just for shits and giggles.

    Stay tuned.
     
    Annika1980, Nov 22, 2003
    #1
    1. Advertisements

  2. Annika1980

    Bob Hatch Guest

    What's film?
     
    Bob Hatch, Nov 22, 2003
    #2
    1. Advertisements

  3. You've got the right idea on films. If you have a good prime, take a few
    shots on a tripod at f/8, take the same shot with the fabulous D60, but with
    a shorter lens so they're framed the same. Landscapes of the golf course
    with trees in the distance would be good. Manually focus at infinity.

    Once you've got the slides scanned, try this.

    In Photoshop, first set the white and black points using levels. Use either
    levels or curves to get the brightness/contrast where you want it. (All
    that, you know.) Then:

    1. Sharpen lightly. Try amount = 100% to 130%, radius = 1.0 to 1.3,
    threshold = 2. (These numbers work for my 4000 dpi scans, you might find
    different values work better.)
    2. Downsample to 2700 x 4050 (Image size, bicubic)
    3. Sharpen to taste.

    Print at 8x10, compare to same shot with the fabulous D60, and report back.

    David J. Littleboy
    Tokyo, Japan
     
    David J. Littleboy, Nov 22, 2003
    #3
  4. Annika1980

    zbzbzb Guest

    1. Sharpen lightly. Try amount = 100% to 130%, radius = 1.0 to 1.3,
    Why sharpen twice and why downsample? Considering the 5400 dpi scan is going to
    be quite a bit bigger than the D60s doesn't it make more sense to upsize to the
    5400 dpi size if you are trying to compare?
     
    zbzbzb, Nov 22, 2003
    #4
  5. I am not Nostradamus, but my feeling is that you will come back to film
    (like I did). In my view, one needs a good scanner to understand that
    the digital revolution means largely a new lease of life for film
    (thanks to the scanner, and prosummer printers).


    nobody
     
    nobody nowhere, Nov 22, 2003
    #5
  6. Scans straight out of a scanner are extremely soft. (Look at the original,
    not the large.)

    http://www.pbase.com/image/22348855 (Image at right is with the scanner
    defocussed.)

    The sharpen/downsample/sharpen trick produces images that are closer to
    digital originals in terms of detail/pixel.

    http://www.pbase.com/image/22348935

    My suggestion corresponds to splitting the difference. He can upsize his D60
    image by 1.35 times to match the downsized film scan.

    David J. Littleboy
    Tokyo, Japan
     
    David J. Littleboy, Nov 22, 2003
    #6
  7. Annika1980

    Darrell Guest

    The real issue is the minute the film is scanned the test is biased, I have
    yet to see a test where both digital and optical are kept in their native
    formats. If I still worked in a pro darkroom, I would have shot a roll high
    quality film, and make a 24x36" or bigger print via photo chemical (analog
    with a good grain focuser) methods, then try to make the same size print via
    the digital route. The digital and chemical method don't require a huge
    print one can print center and corner on 8x10" paper and select the same
    areas for digital output. I would take the images as suggested by David
    Littleboy at f:8 or f:11 as that usually the sharpest area of most lenses.
    Put the same lens or your D60 and EOS 1V. With a 6.3 MP cam up to 11x14
    should be pretty close, after that let's see what happens.
     
    Darrell, Nov 22, 2003
    #7
  8. (Annika1980) wrote in
    You coulda saved yerself some money by simply buying an $80 microscope:

    http://www.mindspring.com/~tony1964/MicroSlides.html

    http://www.mindspring.com/~dreamflier/Films-vs-Coolpix.html

    Film definitely has much more resolution than 5 megapixels. But in the real
    world, my experience is that there are no companies out there that can give
    me an enlargement from film that looks better than a digicam equivalent.
     
    Tony Whitaker, Nov 22, 2003
    #8
  9. Annika1980

    Mark Johnson Guest

    Or just get a rig to fit the digicams, itself - like one of those
    Bower Duplicator tubes. It's just tube threaded 52mm, I think, with
    it's own macro lenss, and a head for a slide holder. You can stick on
    more macro lenses or whatever. And the results are pretty good, and
    obviously 'shake-free', or whatever. It's as simple as could be.
     
    Mark Johnson, Nov 22, 2003
    #9
  10. Annika1980

    Mark Johnson Guest

    I would think that film is superior, but that large format film is
    superior to 35. I would think all the pro equipment geared for film
    cameras still make these far superior, as well. I would also imagine
    that you could use a 1DS, digital SLR, which I think is 10MP, and get
    a clean reproduction for coated paper coffee table books, without
    having to stitch a panorama. But I would also think that the optics is
    important, as well as all the rest for any photo. You need good
    lenses.

    But the advantage of not having to pay so much for digital remains,
    nonetheless. A photographer might try to bracket shots to make sure he
    or she got the proper exposure. And that many rolls of film can get
    expensive. Now, I realize, that the kits, themselves, that they use
    can be expensive. I realize that a 1DS is maybe $4000-5000 street
    price. That's not cheap, particularly if you further need a new set of
    lenses. But it could still add up, and maybe limit the number of shots
    one might feel they can afford. So, just for that, that has to be an
    advantage of digital that adds up as each new memory card is used, and
    reused. And if you try to stitch panoramas, if you have reasonably
    good optics in the digicam, would the tests show that film was still
    superior?
     
    Mark Johnson, Nov 22, 2003
    #10
  11. Darrel hits the nail on the head.
    If we are going to compare Digital to Film the film must NOT be scanned.
    The scan is a second generation image, and it makes huge assumptions about the
    scanner's capabilities.
    You may have to go to a custom lab to have them print an 8x10 section of a 24 x
    36 enlargement.
    Do the same with the digital, print it on your best printer and compare the two
    prints, side-by-side.
    Or send the section of the digital file to Wal-Mart Online (they do excellent
    work) and have them print it on silver halide paper so both prints will look as
    much alike as possible.
    You may be able to spot differences at a lesser enlargement but if there is a
    difference in quality, it will SURELY show up at a magnification of 24 x 36
    The other caveats about using a Tripod, shooting at f:8 and capturing the same
    image are also very important.
    Bob Williams
     
    Robert E. Williams, Nov 22, 2003
    #11
  12. Annika1980

    Dr. J. Smith Guest

    What's the point? The progress and improvements to film are miniscule
    compared to what's happening with digital. Given the rate that things are
    improving it's reasonable to assume that large format quality digital
    cameras will be available in a year or two. Want to argue that? Fine. I'll
    concede it may take two or three years, or three or four, but I seriously do
    ubt it. And that's barring another breakthrough. The point is this
    technology is not stopping, it's accelerating. The way I see it, redundant
    resolution is just around the corner, though let's not forget the snake
    scales in Blade Runner... heh heh heh...
     
    Dr. J. Smith, Nov 22, 2003
    #12
  13. Annika1980

    badgethefish Guest

    Anni, do I really have to remind you about all those endless (totally
    digital) discussions we had about digital vs. film ????

    I am always surprised in human nature and peoples ability to change back
    from completely polarised views, when it suits them.

    Still at least you are honest enough to admit you are going back to film,
    and you can experience & experiment with both forms of technology (unlike
    some closed minded As* holes here) I think this decision needs a greater
    fanfare......and you might well find yourself arguing in favour of film...oh
    boy......

    Badge
     
    badgethefish, Nov 22, 2003
    #13
  14. Annika1980

    andrew29 Guest

    A print is a second generation image too. Every stage involves losses.

    The right way to do the comparison is to get the best possible image
    from both film and digital, and the best possible way to get a print
    from film is not necessarily by a traditional enlargement, as many
    people have discovered. Some labs use a Durst Lambda or something
    similar for their highest quality prints. Why cripple the film by
    insisting that it not be scanned?

    Andrew.
     
    andrew29, Nov 22, 2003
    #14
  15. Annika1980

    johnpower Guest

    And so what if the film looks better. Does the D60 go up for sale?
    Do you relegate it to your point and shoot?

    Are you willing to give up the advantages of digital for an image that
    looks a little better? Because no matter how much better the
    aficionados or "experts" say it looks, show the same shot to the
    typical viewer and they will either not see a difference or won't care
    because they both will look so good.

    I believe I would have set that scanner money aside in a 1Ds fund...
     
    johnpower, Nov 22, 2003
    #15
  16. Film is a first generation information file, just as a memory card is a first
    generation file
    No scanner, Durst Lambda or otherwise, can collect more information from the film
    than is already there.
    Any output from the scanner that is not identical to the information on the film is
    not information. It is artifacts.
    This is not to say that you can't improve the APPEARANCE of an image by manipulating
    it digitally. I do that every day.
    For instance, a levels adjustment or application of an USM may improve the LOOKS
    of an image but it has not added any information to the digital file. It has added
    artifacts. It's just that we like the looks of the artifacts better than the
    original.
    But, if you are truly going to compare a print from film to a print from a digital
    file. You must not introduce any alternate technology between the film and its
    print.
    Bob Williams
     
    Robert E. Williams, Nov 22, 2003
    #16
  17. Annika1980

    Annika1980 Guest

    From:
    I'll never get enough scratch to buy a 1Ds.
    Besides, I have boxes of old slides waiting to be scanned. And who knows, I
    may get into the business of photo restoration and that scanner could pay for
    itself rather quickly.
     
    Annika1980, Nov 23, 2003
    #17
  18. Annika1980

    Annika1980 Guest

    From: "Robert E. Williams"
    I'm not interested in drum scanning or what a custom lab can do since that
    isn't going to part of my normal workflow. And I don't buy the argument that
    says you must print from film to realize it's advantages. If the film really
    does capture that much more information, a scan should show it. Otherwise, the
    extra information that is there is too subtle to make a difference anyway.
     
    Annika1980, Nov 23, 2003
    #18
  19. Annika1980

    Annika1980 Guest

    From: "Robert E. Williams"
    Is it possible to print from slides without using an internegative?
    If so, how is it done?
     
    Annika1980, Nov 23, 2003
    #19
  20. And it does, if only you could get a Nikon 8000, or the equivalent.
    When I look at the scanned image of a slide (with a Nikon 8000), and
    compare it to the same slide I see on a good light box with a 22x
    magnifying glass (admittedly, not a microscope), every single detail on
    the slide has been captured by the scanner. Others in this NG, whose
    credentials are better than mine, have said so as well a number of
    times.

    nobody
     
    nobody nowhere, Nov 23, 2003
    #20
    1. Advertisements

Ask a Question

Want to reply to this thread or ask your own question?

You'll need to choose a username for the site, which only take a couple of moments (here). After that, you can post your question and our members will help you out.