film vs. flatbed scanners

Discussion in 'Scanners' started by Mike - EMAIL IGNORED, Apr 7, 2006.

  1. Mike - EMAIL IGNORED

    Alan Browne Guest

    If the original image is very sharp then 5400 dpi will show up. If the
    original image is limited in resolution, then a lower scan dpi will
    yield the same thing.

    The images that you post above indicate that the sharpest tool was not
    used. The quality chain begins with a sharp lens, properly focused,
    high res film and a dead still tripod and high shutter speed (or flash
    as majority light contributor). I'm not sure what your full fram image
    was, but another 'factor' is that long range photos using a telephoto
    are among the worst examples and add other "smoothings" from air currents.

    The great majority of images will scan fine up to 3000 - 4000 dpi. Very
    few will benefit from higher. But those that do have to be be shot
    critically as described above.

    And then, to get the most out of the image, a drum scan will get more
    detail at any given dpi than an "air space" film scanner.

    Alan
     
    Alan Browne, Apr 9, 2006
    #41
    1. Advertisements

  2. Mike - EMAIL IGNORED

    Gordon Moat Guest

    I rarely agree with Roger either, but I think your methods need some
    improvements. While Roger's stuff is becoming dated, and has probably
    been quoted too much, I do think some of his tests are still valid.
    Perhaps one problem is that you are not printing out these images, or
    crops of these images, but that you are viewing on a monitor that has at
    best 100 dpi resolution. If you view an image at 100% in PhotoShop, it
    is still interpolated to your monitor. Easy way to understand this is to
    put an 8x loupe onto your monitor, then you see the square pixels of the
    monitor, not more detail. Put an 8x loupe onto a print, and you can
    often see more detail, though if it happens to be an inkjet print, then
    detail is more limited. In order to properly do this comparison, you
    should be determining the actual output limitation of your printer,
    which you can do with an EPS that you can build in Adobe Illustrator.
    Once you have the output resolution, then factor the magnification of
    the scan to match the output, using larger crops as needed to fit a
    section of paper.

    Quite often scanner comparisons show more about how limited a scanner
    can be, rather than any real information about how scanning resolution
    relates to detail limitations. While it might be true for some scanners
    that there is nothing to be gained from scanning beyond a certain
    resolution, this will not be true for ALL scanners. Scanners are too
    different for us to realistically make such a claim, even if we only
    limited our selection to CCD film scanners.

    <http://www.porteous.net/test/digi.html> Shows a film crop of 1.4mm by
    1.1mm from 35 mm Velvia. While not a technically brilliant test, and too
    simple, at least the scannned photograph through a microscope indicates
    that even Velvia is better than his scanner is capable, either scanner.



    Okay, very simple question, what percentage viewing size in PhotoShop
    would you need to use in order to judge whether an image would make a
    nice 12" by 16" print? It would be nice if you included what monitor you
    are using to make this decision.

    Resampling JPEGs? I don't understand why you don't shoot some film and
    do some comparisons on your own. It is easy to sit on your ass and
    criticize others, but your validity would be better off if you did your
    own comparisons, then posted your own results. Besides, tests are only
    valid for what each individual wants to achieve within the limits of
    their own incomes.

    A nice example of having lots of income and resources, and a more modern
    test, is at:

    <http://www.xs4all.nl/~diax/pages/start_mamiya_nikon_uk.html> Comparison
    of D2X and Mamiya 7, using Imacon and ICG scanners. He also discusses
    using NEAT image to clean up the scans. Tons of links to other relevant
    information, including a comparison of an ICG drum scanner and an Imacon
    848, which if anyone bothers to read it actually reveals some
    interesting differences:

    <http://www.xs4all.nl/~diax/pages/mamiya_boot_scan_compare.html> It
    should be pointed out that ICG is now considered the best drum scanner
    manufacturer, though I do not know exactly which model was used for
    these tests. The reason this test is relevant for a 35 mm imaging group
    is it shows an interesting issue with CCD scanners, since the Imacon
    scan was greater resolution than the ICG drum scan. This quote might
    entice people to read a bit of this, despite the large download:

    "The Imacon is very sharp. Maybe sharper. (i used an unsharp mask) The
    problem is the blooming. At edges, white to dark transitions there is a
    certain halo effect. All edges seem sharper with the drumscan. The
    leaves of the trees are much more defined. Although the Imacon scan is
    very sharp, because of the halo effects everywhere the Drum scan seems
    twice as sharp when you compare the prints. Of course there is also a
    better definition in the dark areas. You can see this clearly in the
    dark paint of the boat."

    It should take about 30 minutes to read the links I have provided, if
    you bother to actually read them. A fast response after my posting time
    would indicate you don't have an interest in learning. There is no such
    thing as a perfect test, probably because people often have a vested
    interest in pushing the gear they own above others, a sort of
    validation. The best way to validate your methods is to compile them in
    a concise manner, then post them for others to read; I have already done
    this, so now it is your turn.

    Ciao!

    Gordon Moat
    A G Studio
    <http://www.allgstudio.com/technology.html>
     
    Gordon Moat, Apr 9, 2006
    #42
    1. Advertisements

  3. SNIP
    It is indeed a small difference because there usually is only a little
    more detail in the film to be extracted. Whether it is
    (in)significant, is subjectivel:
    <http://www.xs4all.nl/~bvdwolf/main/foto/scan/se5400/se5400-5.htm> .

    The difference is most apparent on filmshots taken with good technique
    and quality lenses.

    Another factor is that the graininess is better defined on the higher
    resolution scan, leading to less graininess and it makes it easier to
    control with NeatImage/NoiseNinja/etc.

    Bart
     
    Bart van der Wolf, Apr 9, 2006
    #43
  4. Mike - EMAIL IGNORED

    no_name Guest

    Except that AFAIK, the 9000ED is Nikon's "top of the line" model, at
    about $1800 USD.

    Too bad they don't make a "CoolScan IX" ...
     
    no_name, Apr 10, 2006
    #44
  5. Top of the 35mm line. The 9000ED is "higher" because it supports
    medium format, not especially because it does anything else better.
    Ah, in the cheap consumer medium-format series? I'll look forward to
    it....
     
    David Dyer-Bennet, Apr 10, 2006
    #45
  6. Mike - EMAIL IGNORED

    pioe[rmv] Guest

    Don't even think of buying a consumer class flatbed for 35mm film
    scanning. A dedicated film scanner like the Nikons LS-50 and LS-5000
    will be incomparably superior to any flatbed scanner.

    That is, provided you do not use a flatbed like the Aztek Plateau,
    which by the way can be purchased new for $18 000.

    Per Inge Oestmoen, Norway
     
    pioe[rmv], Apr 11, 2006
    #46
  7. Mike - EMAIL IGNORED

    Alan Browne Guest

    Unless you shoot for sharpness (lens, film, shutter speed/ flash, mirror
    lockup, good tripod, calm/clear day ...) you cannot get maximum
    sharpness. The way most people shoot and print, it does not matter a bit.
     
    Alan Browne, Apr 11, 2006
    #47

  8. So should we only take photos on calm, clear days?


    rafe b
    www.terrapinphoto.com
     
    Raphael Bustin, Apr 11, 2006
    #48
  9. Most of us probably know what the "ls-5000" is, but the actual name is
    "Super Coolscan 5000 ED"; just to avoid confusion.
     
    David Dyer-Bennet, Apr 11, 2006
    #49
  10. Mike - EMAIL IGNORED

    Alan Browne Guest

    I never said that. I said that is a contributor to maximum sharpenss.
     
    Alan Browne, Apr 12, 2006
    #50
  11. If you want to scan them at 5400 dpi, then yes....As well as those other
    things...Tripod mount, stationary subjects, & etc. It's a waste of time to
    do a 30 minute, 5400 dpi scan of your average hand-held snapshot of your
    kids playing with the family dog.......
     
    William Graham, Apr 12, 2006
    #51
  12. Mike - EMAIL IGNORED

    Peter Chant Guest

    At this very moment I am scanning 35mm slides on both a 6 year old Acer
    Scanwitt 2720S film scanner and a month old Epsom 4990 flatbed.

    Both cost the about the same new, but the first is a now elderly dedicated
    film scanner, the second one of the better modern flatbeds, from what all
    the reviews say.

    Using vuescan, with the Scanwitt at 2700 dpi and the 4990 scanning at 4800
    but downsampling on save to 2400 (not too far off 2700) the Scanwitt is
    noticably sharper. However, the Epson has rather a good dmax for a flat
    bed, better than the rather old Scanwitt.

    From the limited experience of the ir channel and dust removal (via Vuescan)
    on the 4990 compared to the lack of it on the Scanwitt and retouching in
    photoshop, I'd suggest that an IR channel is a must have.

    Newer film scanners have better dmaxes than the Scanwitt and the current
    (discontined ?) Minolta Dimage 5400 has better then the 4990, so I would
    hope that others do.

    I really need to get my new more powerful computer for medium format on the
    4990 as I suspect that will beat 35mm on any scanner by a large amount! :)

    I'd suspect buying a 35mm film scanner and then a cheapie flatbed for
    documents and the odd print would be a good bet.
     
    Peter Chant, Apr 12, 2006
    #52
  13. I'd suspect buying a 35mm film scanner and then a cheapie flatbed for
    I agree. A dedicated film user really needs both. However, a good quality
    flatbed might be a good purchase, especially if you wanted to scan film that
    is in larger format than 35 mm. I also wish that I had some way of scanning
    my wife's APS negatives.....I am thinking of trying to modify a KM-5400 film
    holder for this purpose.......
     
    William Graham, Apr 12, 2006
    #53
  14. Mike - EMAIL IGNORED

    Peter Chant Guest

    Hmm, yes, the advantage is that you can scan almost any negative you can
    throw at it. Suppose it depends on the chances of getting any larger
    films. Decided on the 4990 over the 4490 as I thought the extra dmax was
    worth the extra money. Probally not so easy to answer without buying both!

    Of course I was not going to want to scan MF, I had bought a Mamiya C330 and
    a suitable enlarger and was going to stop there but I gave in in the end.

    Just wondering what happened to the large heap of black and white negatives
    that I thumbed through as a child that used to live in my grandmother's
    sideboard. Probally scratched and covered in sticky fingerprints by myself
    as a child I'd bet.
     
    Peter Chant, Apr 13, 2006
    #54
  15. If that's all that's wrong with them, you can probably clean them
    off....Either physically, or with Photoshop. It's the ones ruined by fungus
    that are hard to fix.....
     
    William Graham, Apr 13, 2006
    #55
  16. Mike - EMAIL IGNORED

    Peter Chant Guest

    Don't remember any fungus, but they may have remnants of custard cream on
    them. It would be interesting to see if there were any negatives that we
    don't have prints of, or whether I can get significant improvements by
    scanning negs rather than prints. Doubt it, most photos were probably
    taken on a box brownie, and not the flash new plastic one either. I
    suspect they are contact prints.

    Mind you, interesting to see the quality of a couple of the professional
    portraits, one, taken I would guess 1905 - 1915 at latest has surprising
    amounts of detail (apart from the one child who would not sit still). I
    suppose its a contact print from a large format neg, so not so surprising,
    but a lot of photos leave a bit to be desired in the sharpness department.


    Pete
     
    Peter Chant, Apr 13, 2006
    #56
  17. You will get much better quality from scanning the negatives.
     
    Nicholas O. Lindan, Apr 13, 2006
    #57
  18. Mike - EMAIL IGNORED

    Scott W Guest

    I have been going through this, scanning old negatives where we also
    have the prints. Often the highlights are blown out in the print but
    can easily be captures with a scan of the negative. It is kind of fun
    getting to see a photo that is more then 50 years old printed well for
    the first time ever.

    Also only about half the negatives were ever printed, these where taken
    during my dad's BW phase and he shot way more then he printed.

    Scott
     
    Scott W, Apr 13, 2006
    #58
  19. Photoshop allows some artificial sharpening, but it is limited, and it also
    sharpens, or makes the grain stand out more.....I am torn between using the
    sharpening tool, or the softening tool to get rid of the grain on many of my
    scans. I frequently use neither, and just let the scan stand on it's own
    merits........My reasoning being that some future generation, with their
    newer software and techniques, can do with my scans as they please......
     
    William Graham, Apr 13, 2006
    #59
  20. I am doing the same thing with many of my dad's old photos....Many that he
    never printed are better than my stuff, even though they are 50+ years old.
    I am still learning photography from him, even though he died in
    1968........
     
    William Graham, Apr 13, 2006
    #60
    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.