Monitor calibration and default hardware white point

Discussion in 'Digital Photography' started by Dave, Sep 29, 2004.

  1. Dave

    Dave Guest

    I am trying to generate a color profile for my monitor using the Adobe
    Gamma that comes with Photoshop (I have version 7). I have already
    spent a lot more time on this than I intended -- searching the Internet
    for tutorials and resources on the subject, and I just want to generate
    a general-purpose color profile; e.g., my needs aren't specific to
    print, web, video, etc.

    I have downloaded canned ICM profiles for my monitor targeted at 5000,
    6500, and 9300K. Most resources seem to indicate that a color
    temperature of 6500K is desirable for most purposes. However, the
    default hardware color temperature that my monitor is set to is 9300K.
    Am I supposed to change my *monitor* temperature to 6500K in addition to
    the software settings? Because when I do, the screen looks awfully dark
    and yellow -- not at all what I am used to. And I would assume that the
    factory defaults would be a closer baseline to what is most comfortable
    to the eye.

    But anyway, could someone please tell me if this is the correct
    procedure to calibrating the monitor through Adobe Gamma for 6500K:

    1.) Change my monitor from the default setting to 6500K
    2.) Load the 6500K canned ICM in Adobe Gamma as the starting point
    3.) Calibrate from there

    When doing so, I noticed several strange things:
    1) The phosphors are listed as "custom" but I am almost positive it is
    Triniton, and 2) the Gamma value is listed as custom instead of "Windows
    Default" with a value of 2.50. 3) The hardware white point is listed at
    5000K. Oddly, I repeated the same steps using the default 9300K monitor
    setting and 9300K ICM, and the values are the same.

    FWIW, the monitor I have is a 21 inch CRT, IBM 6558 P202; it's
    manufactured by Sony from what I've read.

    Any help would be much appreciated. I am really new to color management
    and right now I don't have much time to invest in learning, but would
    like to at least generate at least a *reasonably* optimized color
    profile that is better than not nothing, which is what I have now.

    Thank you.

    Dave, Sep 29, 2004
    1. Advertisements

  2. Dave

    Rick Guest

    You must allow a few days for your eyes to adjust. If 9300K
    is all you've ever used, 6500K will look horribly yellow for a
    few days, but your eyes will eventually adjust.

    Follow Ian Lyons' excellent procedure, available here:

    Rick, Sep 29, 2004
    1. Advertisements

  3. Dave

    Dave Guest

    It's a relief to know I'm not crazy.
    Thanks for the tip. Believe it or not, I had actually been using that
    guide, which generated further confusion when I attempted to use my
    manufacturer supplied ICMs as a starting point for calibration. As
    noted in my previous message, when using the vendor supplied ICMs the
    following differences were observed (by comparison to creating a new ICM
    from scratch):
    1) The phosphors were listed as "custom" even though I know they're
    Triniton. Should I leave them at "custom" anyway?
    2) The Gamma value is listed as custom (2.50) instead of "Windows
    Default"? I don't know if I should change it to "Windows Default" or
    leave it at 2.50.
    3) The hardware white point is listed at 5000K instead of 6500K.
    Repeating the same steps using the default 9300K monitor
    setting and 9300K ICM profile yielded the same results. Should I change
    this to 6500K?

    In short, the calibration process seems to produce the results
    anticipated when following the guide *if* I start from scratch, but when
    I start from the vendor supplied ICM profiles, the results leave me
    wondering whether I should avoid using the baseline profiles for my
    monitor. I compared the monitor ICC files from the original monitor
    installation disk (file timestamps of '97) with a set I downloaded from
    the IBM site for a similar but newer monitor, and they were binary
    identical; I'm not sure how "fine-tuned" they could possibly be for my

    Dave, Sep 29, 2004
  4. Dave...

    after messing around with software, todo's and howto's for two
    years... I have finally given up and purchased a Colorvision Spyder
    when price was reduced from something around 400,- to something around
    150,- I use it with the allegedly amateurish PhotoCal solution (rather
    than the professional Optical that let's you play with the settings
    more but costs a lot more).

    I plug it in and it will calibrate my TFT monitor, my Powerbook and my
    CRT screen in about 3 minutes each. It generates a profile for Windows
    and Mac and my results have become much more accurate - the screen now
    "good-enough" matches the professional output I get from my printing
    service. It's so easy, I calibrate my equipment everytime I start a
    Photoshop session (these usually mean: 4+ hours with Photoshop).
    Actually, this is what you need to do: calibrate often
    Bernhard Mayer, Sep 30, 2004
  5. Dave

    Rick Guest

    The purpose of Ian's procedure is to create a custom ICM
    from scratch. Don't use canned ones, they are intended for
    customers who for one reason or another cannot create
    their own.

    Rick, Sep 30, 2004
  6. The canned profiles are usually a good starting point, but only
    to get the correct phosphor values. Ignore everything else,
    and simply calibrate using Ian Lyons' technique.
    Graeme Cogger, Sep 30, 2004
  7. Dave

    Jim Waggener Guest

    I use the spyder with my TFT as well. I wonder how often you need to
    calibrate a LCD as opposed to a CRT? It would seem a lcd has no way near
    the kind of light fall-off or aging a tube based monitor would have.

    Jim Waggener, Oct 1, 2004
  8. Dave

    Rick Guest

    Wrong. LCD backlights like all other fluorescent lamps
    typically lose 30-40% of their brightness within the first
    two years, three tops. And eventually they fail altogether.

    Rick, Oct 2, 2004
  9. Dave

    Jim Waggener Guest

    Ah, ok, thanks for that info.

    Jim Waggener, Oct 2, 2004
  10. Dave

    Skip M Guest

    That's the primary reason I got a 5 year warranty on mine. The failure rate
    within 5 years is very high, so when mine does (not if) it will get
    Skip M, Oct 2, 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.