FILM vs. DIGITAL (Round 2,739)

Discussion in 'Digital Photography' started by Annika1980, Aug 27, 2004.

  1. Annika1980

    Annika1980 Guest

    Here's an image I took recently with both the Fabulous EOS-1V and the Totally
    Digital D60.
    http://www.pbase.com/image/31516270

    Now here are actual sized crops from both the 5400dpi scan of the 35mm slide
    and the digital image from the D60. I used the same tripod-mounted lens and
    settings for both pics. (Super 70-200 f/2.8L @ 200mm, 1second @ f/22)
    http://www.pbase.com/image/33062580/original
     
    Annika1980, Aug 27, 2004
    #1
    1. Advertisements

  2. A lovely demonstration of my point: the scan looks like [email protected] But it's got
    lots of pixels, so it does have _some_ amount of information.

    So why don't you bicubic (or some other sensible method) upsample the dSLR
    image to about the same scale as the film image, and then compare???

    David J. Littleboy
    Tokyo, Japan
     
    David J. Littleboy, Aug 27, 2004
    #2
    1. Advertisements

  3. Normal failing.. These tests are comparing dSLR with scanners. Print both to
    the same relative enlargement factor in their native formats. And then scan
    those for website display. You don't need a 30x40" Ciba, just an 8x10 crop
    will do. I haven't gotten around to doing that, yet.
     
    Darrell Larose, Aug 27, 2004
    #3
  4. Scanning technology has long been to the point that projection prints simply
    aren't any better. (There's argument as to what final printing technology is
    best, but the wet chemical processes can all be driven from pixels.)

    David J. Littleboy
    Tokyo, Japan
     
    David J. Littleboy, Aug 27, 2004
    #4
  5. Annika1980

    Mark M Guest

    I think you might be in for a shock under those circumstances.
    I think the differences will lean LESS toward film in that test.
    Hmmm... I'll have to try it sometime, but my Nikon film scanner is on the
    blink...
     
    Mark M, Aug 27, 2004
    #5
  6. Annika1980

    YoYo Guest

    This is why so many went digital, the
    cameras are better then the scanners.

    in message
    both the Fabulous EOS-1V and the Totally
    both the 5400dpi scan of the 35mm slide
    used the same tripod-mounted lens and
    f/2.8L @ 200mm, 1second @ f/22)
     
    YoYo, Aug 27, 2004
    #6
  7. Annika1980

    Bowser Guest

    Why a different angle of view?
     
    Bowser, Aug 27, 2004
    #7
  8. There is only one image at that URL.
    So which is which? I assume the top one is the film scan, but you don't
    say. Also not sure film camera was focused.

    Gary Eickmeier
     
    Gary Eickmeier, Aug 27, 2004
    #8
  9. SNIP
    I normally avoid multiposting but, since you did, I'll repeat my
    response to your earlier post about that image:
    Interesting, but why is there so little grain visible in the film
    scan? I know that the DSE-5400, or rather the software that goes with
    it, can have difficulties with focussing, and this looks just like it.

    Also, how sure are you that the focus/aperture/tripod settings were
    optimal in both cases? The reason I ask is because I've measured the
    capability of the 5400 to resolve fine detail, and in my sample
    (http://www.xs4all.nl/~bvdwolf/main/foto/scan/se5400/se5400-1.htm) I
    measured 76.1 lp/mm (total lens+film+scanner image chain). The D60 is
    physically limited by it's Nyquist frequency to 67.8 cy/mm and that is
    even before the lens and AA-filter reduce that to approx. 53-58 cy/mm.

    Then how can 58 cy/mm be sharper in your example than 76.1 cy/mm, even
    despite the magnification difference?? Process error seems the most
    likely explanation to me..., and those are not hard to achieve because
    of the number of variables.

    It would be interesting to conduct a little experiment when your D20
    arrives in 1-2 months, and you can compare it to your current D60 (so
    you can objectively quantify the difference).

    For those who want to do their own testing, I have created a target
    file that is also better suited for Digcam sensors, not only for
    analog film, and you can make your own target from it at home with a
    decent inkjet printer.
    For HP/Canon inkjet printers (3.8MB):
    http://www.xs4all.nl/~bvdwolf/main/downloads/Jtf60cy-100mm_600ppi.gif
    For Epson inkjet printers (5.3MB):
    http://www.xs4all.nl/~bvdwolf/main/downloads/Jtf60cy-100mm_720ppi.gif

    Print it at the indicated ppi without colormanagement (e.g. with
    IrfanView) or printer enhancements on glossy Photopaper which should
    produce a 100x100mm target, and shoot it with your (digi)cam from a
    (non-critical) distance like between 25-50x the actual focal length.

    Take several shots of it, while re-focussing between shots (to avoid
    variations in auto- or manual focus), and a sturdy tripod and
    mirror-lockup will help to get the best out of your system. Then
    choose the sharpest shot (shooting digital is cheap, so shoot several
    and be amazed by the differences between them). Also revealing is
    shooting handheld, your average resolution will probably drop quite a
    bit, due to camera shake. All are easily detected with my test target.

    The resulting center "blur"diameter is a measure of "on-sensor
    resolution" of the whole optical chain (lens+AA-filter+sensor), and
    can be expressed as cy/mm after calculating "(60/pi)/diameter". The
    diameter can be expressed as number of pixels multiplied by the sensor
    pitch.

    For comparison with different sized sensor arrays or film, all you
    then need to do is relate it to the physical sensor size difference,
    because a larger sensor needs less magnification for equal sized
    output.

    Bart
     
    Bart van der Wolf, Aug 27, 2004
    #9
  10. information.

    In that case your point is [email protected] ;-). The scan was not well focused, it
    seems (lack of sharp graininess), and how do we know whether the lens
    was optimally focused between bodies?

    See my other post to Annika for more detail, and a suggestion to cut
    the [email protected] and do an objective test. The digital test is easiest because
    it costs no money to shoot multiple re-focused shots and pick the
    best, film is more costly and requires careful scanner focusing as
    well..

    Bart
     
    Bart van der Wolf, Aug 27, 2004
    #10
  11. Annika1980

    Annika1980 Guest

    From: "Bowser"
    ???
     
    Annika1980, Aug 27, 2004
    #11
  12. Annika1980

    Annika1980 Guest

    From: Gary Eickmeier
    Yes, that's a shot showing the full frame from which the crops were taken.

    http://www.pbase.com/image/33062580/original
    There was nothing wrong with the focus on either pic (same lens, same focus,
    only the bodies were changed).
    Remember that you are viewing a highly magnified tiny (1/8" or so) section of
    the 35mm frame.
    That's why I included the full frame pic for reference.
     
    Annika1980, Aug 27, 2004
    #12
  13. Annika1980

    Bowser Guest

    The film scan is cropped tighter, or the angle of view was smaller. Either
    way, the right way to do this is to show the exact angle of view for the
    crops, as well as the original shot. That way, the extra info captured by
    the scanner can be compared to an image that's exactly the same, and the
    digital image will begin to pixelate. Right now, you probably have both of
    them at 100%, and the scanned image is bigger.

    Other than that, it shows that scanners suck, and if you shoot film, get it
    done another way.
     
    Bowser, Aug 27, 2004
    #13
  14. Annika1980

    Annika1980 Guest

    From: "Bowser"
    I've heard that argument before and I still don't agree with it. The point of
    the exercise is to see the images at their natural output sizes. Upsizing the
    digital makes no sense. Downsizing the film image gives away any advantage
    that the film scan has. So the images aren't the exact same size. So what?
     
    Annika1980, Aug 27, 2004
    #14
  15. Annika1980

    JPS Guest

    In message <>,
    Especially if you shoot in RAW. Shooting in JPEG is like bringing a
    roll of film to a lab and only getting back the prints.
    --
     
    JPS, Aug 27, 2004
    #15
  16. Annika1980

    Bob Guest


    ummmmmm


    which one is which??

    could you label them???

    I'm 'guessing' the top is film?
     
    Bob, Aug 28, 2004
    #16
  17. It's interesing that a lot of people haven't seen what high-res film scans
    look like. They're really really bad compared to images from dSLRs _on a
    per pixel basis_. Take a look at

    http://www.terrapinphoto.com/jmdavis/

    for a lot of examples.

    David J. Littleboy
    Tokyo, Japan
     
    David J. Littleboy, Aug 28, 2004
    #17
  18. Annika1980

    Annika1980 Guest

    http://www.pbase.com/image/33062580/original
    Nothing gets by you.

    "Scanned 35mm vs. 6.3MP Digital
    These are actual sized crops from a 35mm slide scanned @ 5400dpi on the Minolta
    5400 and a full-size image from the D60."

    Now which would you think is the first one?
     
    Annika1980, Aug 28, 2004
    #18
  19. Annika1980

    Annika1980 Guest

    From: "David J. Littleboy"
    Maybe the guys just don't know how to focus?
    Hey he's even got one of mine on there!
     
    Annika1980, Aug 28, 2004
    #19
  20. Annika1980

    Bowser Guest

    Unless you view the images with the same field of view, you can't really
    tell what resolution differences, if any exist. The digital image looks
    sharper, but is it? Blow it up so the bug is the same size as the scan, and
    it may look different as the digital image pixelates.

    Tell you what. Now that I have acquired the world's best digital camera,
    I'll do a test and show you what I mean. I'll shoot an 8M digital, 35mm, and
    645, scan, and compare the results. I'll put a a web page, too. Three
    cameras, same angle of view, same magnification for the images. You (and I)
    may be surprised. In the meantime, I suggest single malt scotch. It won't
    help your photography, but after a few hits, you won't care. And it tastes
    good, too.
     
    Bowser, Aug 28, 2004
    #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.