"at least equal or better image quality than Canon EOS-1Ds"

Discussion in 'Canon' started by Mike Henley, May 21, 2004.

  1. Mike Henley

    Mike Henley Guest

    "I've no doubt that a technically optimal Fuji Provia or Astia 100
    frame from, say, the sub-$100 Olympus µ [ mju: ] II, if professionally
    printed chemically or scanned on a drum scanner and properly
    processed, can produce at least equal or better image quality than the
    same frame shot with royalty like the Canon EOS-1Ds." (the Canon
    EOS-1Ds is an 11mp $8000 dSLR)


    What do you guys think of this?

    (p.s. please read the article if you have intentions of forming
    negative opinions of him; he prefers digital over film in practice.)
    Mike Henley, May 21, 2004
    1. Advertisements

  2. Mike Henley

    Justin Thyme Guest

    Perhaps to some extent it is a bit like the vinyl vs CD debate in the music
    world. CD is technically better, but sometimes vinyl just sounds better,
    because it isn't as good. I have a pretty good sound setup that is based on
    early 80's technology, and i think it whips the pants off modern digital
    hifi, even if technically it isn't as good. Anyway... back to the topic -
    For the price of the equipment, film still outperforms digital and by a long
    way. (note, approximate prices are average street prices here in australia).
    a $20 disposable camera will give a much better print than the basic $200
    kodak CX6200 point and shoot digital. A $300 basic SLR will give a much
    better print than a $300 digital (eg fuji S3000). A good film SLR of around
    $1000 will give a much better photo than a $1000 digital (eg fuji S7000). It
    really is only when you start getting into the $2000 and above (nikon d70,
    canon eos300), that you even start to rival film quality. Of course however
    the $200 kodak CX6200 can do something that even the most expensive film
    camera can't, and that's instant preview of photo, and a print within a
    couple of minutes of taking the photo.
    What he says about the development process is true too - in a shopping
    centre across the road from me, there are 2 mini-labs - one kodak and one
    fuji. I always used the kodak lab and kodak film, simply because i figured
    kodak had the best name. I had always thought 400 film was too grainy for
    anything useful, even when printed at 6x4, so I only used 100 speed film.
    Every time i used 400 speed film for the extra speed, the prints were so
    grainy i regretted my decision to use it. I was then starting to think my
    camera's focus alignment was up the creek because all the prints were out of
    focus. Anyway, a friend started working at the fuji lab so i got a roll
    done there and was blown away by the picture quality. A roll of 400 from the
    same batch as one I'd previously had developed that was too grainy, came out
    just as clear as 100. So I got them to do a reprint of one of my 400 negs
    that kodak did, and it came out grainy, but in focus. Conclusion - the
    local kodak minilab develops negs badly resulting in the extra grain, and
    their enlarger is out of focus. I now use 400 speed almost exclusively, and
    the results are excellent.
    I posted in aus.photo about an experience I had yesterday comparing the fuji
    s7000 and eos300d - the s7000 over-exposed the highlights and under-exposed
    the shadows - basically it just wasn't capable of recording sufficient
    contrast for the shot. plus it had horrible green halos everywhere there was
    a change in contrast (Eg between the mag wheel and tyre on the car). The EOS
    was much better, fairly close to film in ability to record the contrast. We
    started playing with enlargements (cropping the picture and printing the
    crop as a 8x6), and concluded the eos would enlarge to about a 24x18
    reasonably ok - putting it about on par with 100 film. However when film is
    over-enlarged it still tends to look ok because of the randomness of the
    grain. Digital on the other hand forms horrible square pixelation which
    looks disgusting.
    Justin Thyme, May 22, 2004
    1. Advertisements

  3. Mike Henley

    jriegle Guest

    Image quality encompasses many things. Resolution wise, yes, I can believe
    it. My own tests show 15mp - 24mp is needed to match most films in 35mm. It
    depends on many factors. Print film has the wide dynamic range advantage,
    slide films have a narrow advantage.

    Considering were only talking about the large sensors common in
    interchangeable lens SLRs, digital has the advantage in producing virtually
    noise free images. High res film scans will always reveal the grain
    structure that makes up the image.

    In the end, it is what you need from the medium. I'm happy with my 6mp SLR.
    I rarely shoot film anymore. If I do, I shoot ISO 100 speed. ISO 400 and 800
    speed grain shows on anything bigger than a 4x6 print while the digital
    camera retains the smooth colors so I have no use for these speeds in film.
    I will have no more use for film when a good 10mp+ SLR comes along at a
    reasonable price.

    jriegle, May 22, 2004
  4. Mike Henley

    Mark Weaver Guest

    Digital only pixelates if you fail to up-res before printing. If you up-res
    properly, there's no pixelation (and no grain either) -- it just looks soft
    on close inspection.

    Mark Weaver, May 22, 2004
  5. Everything else he says in the article is spot on, but I think he's
    underestimating the 1Ds. If he's talking about _percieved_ image quality,
    the 1Ds gets close to 645 with ISO 100 film. IMHO, it's not quite as good as
    645, but amazingly close. 35m is a _long_ way from 645. (As I understand it,
    the 1Ds antialiasing filter is not as strong as it should be: if you don't
    want to see aliasing, you'll need more like 15 or 16MP with the same filter
    for that quality imaging.)

    Film has a long "tail": it "resolves" extremely fine detail at extremely low
    contrast. But this "resolution" isn't of much use for imaging, since it's
    extremely high noise. People who claim 35mm is equal to 24 or 35MP are
    talking about that tail. They're not wrong, it's just that it's academic. In
    the real world, you need to capture with a minimum loss of contrast to
    retain textures. The world isn't just signs and license plates. (There was a
    3MP vs. consumer color negative film comparison here a couple of years ago.
    The street signs in the film image were clearer, but the rest of the image
    was a gross disgusting mess, while the rest of the image in the digital was
    quite usable.)

    David J. Littleboy
    Tokyo, Japan
    David J. Littleboy, May 22, 2004
  6. Mike Henley

    Bill Hilton Guest

    From: (Mike Henley)
    What's funny is how people quote snippets that give a totally biased view from
    what the guy REALLY said ... LOL.

    Here's the exact same quote from your source except I've begun the quote two
    sentences above and extended two sentences below, giving it a totally different
    emphasis :)

    In my considered opinion, at this writing, digital exceeds 35 mm film in
    practical image quality at every level... with a very few exceptions.

    The keyword here is "practical." I've no doubt that a technically optimal Fuji
    Provia or Astia 100 frame from, say, the sub-$100 Olympus µ [ mju: ] II, if
    professionally printed chemically or scanned on a drum scanner and properly
    processed, can produce at least equal or better image quality than the same
    frame shot with royalty like the Canon EOS-1Ds. However, I can't see what
    practical importance this has. I have never heard of anyone using a µ to create
    poster-sized prints, which is where the difference between it and a 1Ds (or
    even 10D or D60) would become readily apparent.

    Back to his quote
    I have to disagree with this. I'm able to make much better large prints with
    files from my 1Ds than with any Velvia or Provia 100F or Astia 100F slides I've
    ever taken. I've tested film vs digital a fair amount with the exact same
    Canon "L" lenses and the 11 Mpixel digital files give better results than 4,000
    dpi film scans on large prints (assuming you know the proper techniques for
    rezzing up and edge sharpening). I've had several medium format trannies drum
    scanned (in the dark days before I got my LS-8000) and quite honestly at print
    sizes up to 20x30" the 1Ds files print almost as well as 645 drum scans, and
    far better than 35 mm scans at this size.

    The guy who made the quote has a list of his gear and he's using a 10D, which
    is only 6 Mpixels ... my wife has one of these and I agree that prints from it
    are not quite up to 35 mm quality on 12x18" prints, but the 1Ds is a different

    Bill Hilton, May 22, 2004
  7. This is the standard/common pro view. The Canon 1Ds and Kodak 14n
    were the first digital SLR to approach 35mm color film's resolution,
    but the dirty little secret is that this is only when using a B&W
    target, since the monochrome Bayer sensor can return black values at
    every photosite, even after the color mosiac is placed in front of it
    to interpolated color. To put it simply, color interpolation is 100%
    accurate when no color is involved. With color targets, both fall
    short of 35mm by about a factor of 4.

    Foveon is roughly a 56mm film equivalent in full color...

    Though Discover Magazine rated Foveon much higher than that after
    grueling real life scientific testing...

    "Digital cameras have relied on image sensors that can't do what color
    film does: record all three primary colors of light at each point in
    the image. Instead, each light-sensitive point in the sensor measures
    just one color—blue, green, or red—and complicated software in the
    camera calculates the missing colors. Foveon's breakthrough X3 chip
    solves the problem with a three-layer design that captures red, blue,
    and green light at each point. To demonstrate quality differences, the
    monarch butterfly on this page was photographed with three cameras: an
    $1,800 Sigma SD9 with an X3 chip; a $300 Nikon Coolpix 2500; and a
    $2,300 Nikon 35 mm F5 film camera. Insets show magnified detail from
    each camera's image.


    This photograph, taken with the Sigma SD9 using a Foveon chip, attains
    levels of sharpness and color accuracy usually seen only in
    medium-format cameras that use 120 mm film, which has a resolution
    about twice that of 35 mm film."

    --Discover Magazine
    George Preddy, May 22, 2004
  8. The whole problem with all of this is that quality is whatever people agree
    that it is. Quality cannot be quantified, of course, and so there is no
    objective standard of reference. One can discuss pixels and grain, and
    that sort of thing, but one is still using arbitrary quantities to describe
    issues of quality; one defines the relationship however one wishes, and
    there is no possibility of objectivity in this regard. In science, one
    tries to limit one's metrics to those intrinsic in natural (as opposed to
    artificial) phenomena, which is why science does not address issues of pure
    quality. You simply cannot make one serve the other, in either direction!

    For instance, when CDs came out, I can remember just how excited people were
    about the clear clean crisp highs, and they tended to define quality by the
    perceived presence of these. Nowadays. this same attribute is regarded as
    artificial and unpleasant; hence the move to vinyl. It simply is a matter
    of perception.

    Now we have the argument about (for instance) pixels versus grain. Which
    defines image quality? Who is doing the looking and criticizing? Some
    people like one and some another, and the arguments are basically species
    of techy entertainment, in the main. Or so it seems to me.

    As I pointed out in a recent thread, albeit rather obliquely, some define
    image quality according to the style of brush strokes used! They claim, in
    effect, "a pox on both your houses!".

    So please understand that, while these threads are entertaining, they have
    little enduring merit. At the end of the day, we all have our own opinions
    based on our own experiences, and I suspect most of us at some point find
    ourselves thinking, "Who cares?!?!?"


    Bill Tallman
    William D. Tallman, May 22, 2004
  9. Mike Henley

    Lionel Guest

    Kibo informs me that (George Preddy) stated
    ...you're an obsessional, lying idiot who doesn't know what he's talking
    about, & who just makes this shit up.

    Where are the photos you claim to have sold, 'Preddy'?
    Lionel, May 22, 2004
  10. Mike Henley

    Lewis Lang Guest

    Subject: Re: "at least equal or better image quality than Canon EOS-1Ds"
    If you are getting grainy 4x6" prints from 35mm you probably are setting (or
    letting the camera set) the I.S.O. at 400 speed. Color negative films benefit
    from slight over exposure, not box speed rated exposure. Over-exposing by 1/2
    to 1 stop allows the dye clouds in the C-41 negative emulsion to overlap more
    therefore causing the image to look less "grainy". It also adds shadow detail
    and may improve color saturation.

    I don't know if 24x18 refers to centimeters or inches, but either way, if you
    had some real experience with large enlargemnts from 100 speed film blown up
    large and/or looked very closely at the resulting non-digital photographic
    prints from 100 speed film you would see that digital comes in a very sad
    second. The two (film and digital) are equivalent by no means to anyone with a
    _critical_ eye. Its like saying that an ice cube is equal in mass to an iceberg
    merely because both contain ice and both look about the same size when you hold
    the ice cube up close to your eye and the iceberg is far away.

    I have seen 5.7 MP images blown up to 16x24" and beyond. They look like sh*t.
    Its not merely a matter of pixilization at large enlargements with digital.
    There is no comparison between film and portable 35mm full frame (and that
    includes the 1Ds) and sub-frame digital at any price... yet. Especially for
    large enlargements. There simply is not enough small detail (resolution)
    rendered and contrast range and hue range/bit depth in digital, regardless of
    lack of film grain (which just makes an image look smoother not "better" in all
    the other aspects just described). I've even seen stuff from the Canon 1Ds at
    10x15" (Dennis Reggie's wedding photos) and larger and there simply is not
    enough resolution and subtle differentiation of tonalities/hues to anywhere
    nearly match film's ability. Plus, blown up to about 25x38" or so (full frame
    35mm on 30x40" paper) film shows detail to spare. The best digi portable DSLR
    images I've seen already start to fall apart at about 10x15" and at 30x40" they
    become a sad joke.
    Lewis Lang, May 22, 2004
  11. Mike Henley

    Lewis Lang Guest

    Subject: Re: "at least equal or better image quality than Canon EOS-1Ds"
    Exactly. No amount of upresing or Genuine Fractals can add detail that wasn't
    there to begin with.
    Lewis Lang, May 22, 2004
  12. Mike Henley

    Jim Davis Guest

    I don't think many can afford a good drum scanner, nor do many want to
    scan anymore. that's the beauty of a DSLR, convert the RAW file and
    you're done. You just can't compare the colour from digital to film.

    And the 'technically optimal' is only a wetdream of film purists. It
    doesn't happen outside a lab.

    Not to mention the many other advantages of a DSLR.
    Jim Davis, May 23, 2004
  13. Mike Henley

    Jim Davis Guest

    Pleez do not reply anymore to this idiot. Killfile him so I don't have
    to read his second hand crap, nor comments like yours to him. Just
    plonk him now.
    Jim Davis, May 23, 2004
  14. Mike Henley

    Lewis Lang Guest

    Subject: Re: "at least equal or better image quality than Canon EOS-1Ds"
    Then get others (local lab, Kodak, etc. to scan it for you). Many like to scan
    and tweak just as many like to tweak for best results with non-scanned DSLR raw
    (and jpeg) files. Drum scanners aren't necessary all the time, especially for
    color neg use which doesn't require a great gamma range due to shadow detail
    being recorded as the barest of densities. Shadows on Kodachrome are another
    story and may (or will, depending on the scanner) pick up noise and

    that's the beauty of a DSLR, convert the RAW file and
    By no means are you done. Those who care about their work will continue to
    tweak contrasst, color balance and even dodge and burn as necessary regardless
    of whether the file is from a scan of a film camera or a DSLR. I took a seminar
    with Lewis Kemper on RAW file conversion/"tweaking" and like everything else,
    you get out of it what you put into it. If you are satisfied with nothing
    special that's exactly what you get. If you adjust all the parameters until
    they are their best, then you get a much better result. RAW files are like
    negatives or the scores to music, its what you do with them that is the
    performance. Don't delude yourself - good enough is not always good enough,
    unless you are too willing to settle for mediocrity *whether from a film source
    or a digital source) as oppsoed to excellence. Go for the excellence...

    You just can't compare the colour from digital to film.
    Because film has much greater bit depth -- there is no comparison, film "wins",
    period, exclamation point.!
    You mean advantages like lack of media longevity, blown out highlights, lack of
    resolution, lack of subtle tonal transitions and all for the sake of what,
    chimping and no grain? No thanks, you can keep your advantages and your
    self-rationalizations/justifications. Digital is an inferior medium, and
    especially so for my fine art needs (and Henley (sp?) seems to be just another
    amature trying to justify the inferior technical qualities of DSLRs compared to
    35mm as good enough/advantages - and, yes, I have shot with 4x5" and my
    technical and artistic standards are far above his and most amatures)... and
    many others too - esp[ecially for large blowups (above 11x14"/10x15").

    Look, use whatever you want (and I'm sure you will) just don't tout
    disadvantages as advantages - your needs and level of excellence (as well as
    the authors) seem like mediocrity to me not any attempt at fine art. The one
    area where I do agree with him is that there are different equipment types
    better suited for specific needs but DSLRs only make 35mm redundant for his low
    quality needs, not mine. I expect and demand more from my work. 35mm barely
    meets all my needs. DSLRs, regardless of convenience but because of piss poor
    quality, especiallu for large blow ups, don't even begin to meet my needs/uses.
    Lewis Lang, May 23, 2004
  15. Mike Henley

    Bob Hickey Guest

    Agfa publishes, or used to, charts showing the effects of over/under
    exposure on the colors for their Optima line of films. Of course, I can't
    find the stupid thing. The idea obviously, is to expose to give each layer
    equal light. Not only is Optima one of the nicest films I've used, but Agfa
    has always been extremely helpful w/ any questions I've had. But their
    marketing stinks. Bob Hickey
    Bob Hickey, May 23, 2004
  16. You know, I keep reading, perhaps not often but regularly, that both Agfa
    and Konica produce some remarkable color negative film. Here, you speak of
    Agfa, and I presume you use it and know it reasonably well. So I've some
    questions for you:

    Why do you choose Agfa over other brands when you do make that choice
    (presuming that you do also use other brands)?

    What is your custom for processing, or getting processed, this brand of
    film? What can you say about your experiences in that regard?

    To what other brands and versions of film do you, or can you, compare Agfa?
    Do you only use Optima? <..he's to lazy to Google to find out, so he asks
    the potentially stupid question... lol!>

    Any other comments?

    Oh, and have you tried Konica's film, and if so, do you have anything to
    share in that regard?

    Thanks much!

    Bill Tallman
    William D. Tallman, May 24, 2004
  17. Mike Henley

    Mike Henley Guest

    You see, this is quite interesting, because often in the arguments
    i've seen from people who prefer digital over film, there is usually a
    digital step involved, such as scanning film and digital pictures and
    then comparing the scanned images either on screen or reprinted. I
    think such digital step is unfair to film. I am more interested in a
    *purely* analog (optical-chemical) process for film, where images are
    taken through a lens onto a film and then exposed from a negative onto
    a print... no scanning, no digital stuff in between.

    I think only such a *pure* analog treatment of film would provide a
    fair comparison with digital. And IMHO, personally, i *imagine* film
    will definitely win in such a case, though im' no expert.
    Mike Henley, May 24, 2004
  18. Mike Henley

    Bob Hickey Guest

    The reason I used some Agfa Optima is, I was given some. Up till
    then I used Fuji Superia. But the Optima is much kinder to skin tones, and
    generally much warmer. I don't know anything special about processing except
    that I go to a guy who actually looks at all his work. If I sent it to
    Quallex, I don't think it would matter what I used.
    Bob Hickey
    Bob Hickey, May 24, 2004
  19. Mike Henley

    Bill Hilton Guest

    From: (Mike Henley)
    Many people have moved over in steps, buying a film scanner first so they could
    digitize their film and work on digital images, and eventually (not inexorably)
    switching to digital for some of their work. So comparing digital prints to
    prints from scanned film is what these guys are used to doing. That's how I
    decide whether 6 Mpixels matches film (it doesn't) or 11 Mpixels (it surpasses
    it easily).
    Most of the guys in my area who make a living selling large landscape prints
    would disagree with you. 10 years ago most of them printed Ilfochromes (often
    from large format) and that was considered top of the line. 5 years ago the
    switch to digital prints started, once guys saw what good scanners and a
    printer like the LightJet 5000 could do.

    Today I'd guess 80-90% of the fine art landscape photographers I'm familiar
    with have gone digital. There are a few holdouts, like Chris Burkett, but
    these are usually guys who do their own Ilfochrome printing. Guys like Jack
    Dykinga, Tom Till, even Robert Glenn Ketchum (a Velvia - Ilfochrome advocate
    for years) switched and never looked back.

    I've seen a couple of images that were printed 20x24" on several different
    processes for comparison while attending workshops ... one I remember was
    printed Type R, Type C (via interneg), Dye Transfer by one of the best known DT
    guys in the country, straight Ilfochrome (too contrasty), custom masked
    Ilfochrome (now you're talking), LightJet 5000 from a drum scan and Epson 9600
    from the same scan. The two best prints were the custom Ilfochrome (which took
    most of a day to get right, with all the masks) and the LightJet 5000 print
    (which prints in minutes once the prep work is done)... the LJ print cost about
    1/3 as much and was rated to last almost 3x as long before fading. The Epson
    9600 print was third best, I thought, not too far behind the LJ and Ilfochrome.
    Listen to the experts ... the Canon 1Ds beats 35 mm film for large prints, as
    pretty much everyone who ever actually used one will tell you.

    Bill Hilton, May 24, 2004
  20. Bob Hickey wrote:

    Oh, okay.


    Bill Tallman
    William D. Tallman, May 24, 2004
    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.
Similar Threads
There are no similar threads yet.