I have no idea which calibration software to get

Discussion in 'Digital Darkroom' started by glenn_carroll, Aug 30, 2011.

  1. Hello I hope you all great people can help. I have a brand new imac 21.5inch desktop. All I want is to calibrate this thing because I want prints ordered to match what I see on the monitor. Is this so hard? I found these settings to plug in: gamma 2.2, 6500k and 110 luminance for macs. Look, I'm not technical guru, I just want to be able to calibrate this monitor. After reading call sorts of reviews and other posts on photo.net, I am at a loss. All I can find is confused people and contradicting customer reviews. Just take a look below to see what I mean. I am looking for something around $100 but will pay 150 if I have to. I just dont see why one would be so much more than another -- they do the SAME tihng right??? what am I paying extra for with respect to the munki device? PLEASE HELP -- maybe an amazon link to something would be great.
    SPYDER 3 ELITE
    Review 1) strightforward installation on a mac, calibration is a breeze, it lets you see before and after differences and the prints come out perfect. It even takes ambient light readings while plugged into the usb port. I highly recommend this.
    Review 2) I switched from Spyder2 Pro to Spyder3Elite for work. Mistake. The new software for the S3E is buggy and the UI is awful -- it's like they forgot everything they did right with the S2. The Big One: poor calibration! I ran both wizard mode and expert mode, then used the colorimeter to check the color temperature of gray patches, and the S3 calibrated colors were all over the map -- different by thousands of degrees. The S2Pro calibrated results were nearly spot on from near black to full white
    SPYDER 2 PRO
    review 1) I had no problem understanding the procedure from color vision. HUGE difference from before to after. Great job color vision!
    review 2) I had better results without using this thing. I was looking forward to aligning my computers with proper calibration and this product failed. It was a waste of time. This is just terrible.

    I1DISPLAY 2
    Review 1) great product! easy to use, no previous learning required, and simply does exactly what it says it will do: Calibrate.
    Review 2) this is an obsolete unit. It is discontinued and not usable by furture apple users.
     
  2. i'v felt the same way . . . that said, no matter what system you use, you will have to work with your printing
    company, or your own print system to get the prints right, it's not out of the box, click the button simple . . .
    I'll be interested to see what the "experts" say about your question. . . I kind gave up and follow some
    basic rules that keep my unit dialed in for my needs . . . I print often, every week, my images are
    published all the time, at least a new one or two every week, i send material to stock agents, etc . . . and i
    never have issues . . . I'm working in a color managed environment but calibrate about once a month and
    i'm never sure it's doing anything . . . I do do so much printing that if things need adjusted i tweek the files
    not the monitor . . . i know that's not the right way to do it, but the whole color calibration for my computer
    and monitor then calibration for camera, printers etc gets a little crazy . . .
     
  3. 1_you dont want the spider2 pro (dont think it exist anymore anyway) If you go the spider road.. get the spider3 pro. The spider2 give mix result, and inconsistent one from time to time.
    2_i1display pro (the newer model, the display 2 is a old unit) excellent product, this is what i use. If you get the display 2.. it work on mac with snow leopard, and free update should arrive first week of september... dont see why you will get one, but if it is really cheap why not.
    3_all calibration device are simple to use.. if you can read, you can calibrate ; )
    4_the gamma 2.2, 6500k and 110 is what i use and what i suggest BUT the luminosity (110) can vary due to your need.. some may use 100, some 120 (as suggest by the manufacturer).. my advice is to use the setting you say and see if that work for you.
    Calibration shouldtn be hard; 1 device, press this button, move that cursor.. and voila. no need to be a genius in that department. go with the setting you talk about to start it is a good choice.
    I agree with you, sometime here, people tend to make thing look difficult (me included) but really, the hardess thing to do will be to go out to get the device ; )
    Forget about the Delta E and all other freaking information you can read people talking about.. 1 device, 3 setting, look if that work for you.. and live happy.
    _____
    If the calibration is correctly done the first time, it is normal not to have a major switch with the next one, in fact you shouldtn get much of a difference if you have done them correctly.
    _____
    you dont need to calibrate your camera (or profile it) and you dotn need either to calibrate your printer (profile it) IF you use the correct ink / paper / printer / icc profile
    The problem is that too many people dont know how to print, and then get a special device to calibrate it and go even too far and get lost.. and make there live miserable!.. Get a epson printer, get epson paper and ink.. and use there profile.. most of the time you will get spectacular result. BUT many people even when they do this still have the *my print are too dark* problem.. most of the time because they didtn calibrate there monitor or because they use a too high luminosity setting (110-120-130 etc)
    You dont calibrate your monitor for a particular printer, you calibrate it for a know standard. Theres little point of tweaking the file for a particular commercial printer before having seen a color proof.. and even so, depending of where your image are in the magazine color can shift to accomodate a bigger client or simply because its the way it is when the printer run.
     
  4. digitaldog

    digitaldog Andrew Rodney

    The newest kid on the block with the most impressive new hardware is the X-rite i1Display-Pro. Short of going with a dedicated smart display system (Eizo, NEC SpectraView) that fully controls the entire process, go for the i1D-Pro.
    The older i1Display-2 is end of life, skip it.
    As for getting the display and print to match:
    http://www.luminous-landscape.com/tutorials/why_are_my_prints_too_dark.shtml
     
  5. I agree with patrick, that's why i think calibration it's a bunch of bunk, sales bs, hog wash . . . you need a simple puck and you need to control the light in your workspace. Run the program let it install the results. I dont understand all the details and numbers, I keep it simple. I publish about 500 images a year, from simple editorial to very high end commercial billboards, bus and train wraps in towns like Chicago, and Seattle, ad images in magazines like the new yorker, food network magazine, outdoor photographer, national geographic etc, and have a gallery where i sell my own prints. I calibrate my monitor once a month and dont mess with anything else. . . I dont deal much with fancy profiles, I dont convert to cmyk, I work in pro photo, and leave the rest of the details to the experts, like the art directors, publishers, and printing services . . . i seldom have issues (when i do it's often on the printers side of the equation) but I make conscious choice to never go too deep into the weeds . . . the kiss method . . . if i were going to far off on color i think i'd hear about it . . . the whole calibration thing is simple in theory - in that you calibrate to a standard, but in practice it's a science of it's own that can, if you let it, consume you . . .
     
  6. ...I'm working in a color managed environment but calibrate about once a month and i'm never sure it's doing anything...​
    Even if you don't see a difference between before and after calibration which is often the case when I calibrate, the calibration AND profiling-(mathematically writing a descriptive color characteristic of a display) locks in the meaning of all the possible millions of colors reproducible on the display. No one gets to see all these millions of colors at once to make sure they appear as they should. A hardware calibration provides insurance and reassurance in this respect.
    What this means is maybe you only shoot images that have a set number of overall colors, but one day capture some weird blues, greens and coral hues you've never encountered before. Because the display's color description characteristics have been mapped mathematically, those weird colors will be defined according to where it falls on the display's 3D color gamut model viewed in Apple's Colorsync Utility or a Windows equivalent.
    This allows a more accurate representation during soft proofing on how these millions of colors (including the weird ones) will appear on an outside device like a printer which can't reproduce all colors of a display and some colors a display can't.
    Oh and about conflicting reviews found online, just keep in mind it is possible there are marketing firms that are hired to pay people to write reviews, especially in highly competitive markets. If politicians can do it, why not corporations?
    And sometimes there are those that just like to stir up trouble out of spite or sh*ts and giggles. Check out YouTube responses to see that.
    I'm still using the original i1Display (i1Match 3.6.3 software) in Snow Leopard (Rosetta installed) on my 2010 Mac Mini calibrating an sRGB-ish, IPS panel type Dell 2209WA. I get the same 3D gamut model shape and neutral gray ramp that I got when the Dell was attached and calibrated to my 2004 G5 iMac under Tiger. So I don't see why you should have a problem with any of these calibration packages mentioned above.
     
  7. great post tim, and patrick, thanks for sharing . . . I use the i-1 display two puck . . . it always look
    about the same to me each time as well . . .
     
  8. thank you all so much for the input. You make it seem so much simpler than others., and I never even considered the idea that corps strategically create negative reviews. what a headache for the consumer!
     
  9. I think there is a feature comparison matrix on Xrite somewhere for that. From my read of it, it seems like the CM Display is like the old i1D2 LT product. Same hardware but significantly more limited software feature set and much slower to work (probably just implemented in software as an annoyance). If you're going to blow the cash, get the Pro.
     
  10. digitaldog

    digitaldog Andrew Rodney

    The ColorMunki Display and i1Display Pro are nearly identical hardware. The differences are, the Munki is a different color and its slower. The biggest difference is the software is crippled compared to the Pro but it cost less. I really don’t know why these companies do this. Crippled software is a half solution. But I guess marketing has to target the product at those who think they can and cannot afford a solution. Bottom line, either get the i1Display Pro or the ColorMunki (not ColorMunki display), the former is a Spectrophotometer that also builds paper profiles, the later is the dumbed down Pro Colorimeter.
     
  11. Glenn, to sort out the real reviews from the fake, just look for detail in how the review is written. Just keep in mind it's quite difficult to fabricate a believable lie on something that really didn't happen and especially about something as complex as this subject.
    If the review uses too many simplistic buzz words (positive or negative) that are easily indexed by search engines that will be associated with the product name as search terms, then you know someone's stuffing the ballots.
    Also if possible post specific questions to the reviewer and if they don't answer or don't provide detail that answers your question then you know they're full of it.
     
  12. Glen,
    I am not a shill :) I recently got the ColorMunki. It is a great bit of kit, it is very easy to use and it just does it's job, trouble is it can't match my laptop to my monitor, well it can if you do the laptop first and then choose the option match to existing profiled device, but then both screens are way too yellow. If I profile them as independent devices the profiles are a ways off, even when all set able points are the same, gamma, luminance etc. I haven't set up my printer to them yet though so can't tell what is out of profile.
    Having said that, I'd say the ColorMunki is a good tool. Here is a review of the ColorMunki Display and another review of the i1Display Pro.
     
  13. digitaldog

    digitaldog Andrew Rodney

    trouble is it can't match my laptop to my monitor,​
    Doesn’t surprise me. You have one unit that’s just OK in terms of its performance (gamut, adjustments, white point etc) and one that’s much better. Best you can do is try to dumb down the better display which doesn’t serve you well.
    IF a display is too yellow (warm) adjust the white point to a higher value. Better software products have control over this to a fine degree (5600K is too warm, 5700K too cool, enter 5666K). Products with presets don’t allow this degree of control. Even better, software that allows you to enter xy values so you can adjust the other axis (magenta/green).
     
  14. So what is the point of a calibrator then? My laptop is 100% capable of not being too yellow, it might not be capable of every subtle tone that the monitor is but it is capable of being neutral. So I paid $160 for a calibrator that can't calibrate, it can adjust for flare and ambient light on the fly (who cares?) but it can't get the screen neutral? Crazy.....
    I got around the problem though, I calibrated the monitor and then eyeballed the laptop to it, but is that really the intention of X-Rite?
     
  15. Scott, adjusting the laptop to get rid of the yellow isn't going to affect color under the hood in color managed applications. It's the color temp number the calibrator came up with and associated with that yellow tint that's going to affect hue/saturation.
    If the calibration package measure it to be 6500K give or take 500K, then there should be no hue/saturation errors in CM apps. If it chose 5000K or lower as the number, then you'll probably see darker/richer blues turn a bit light with a bit of magenta and reds and skin tone loose a bit of saturation.
    Not sure why you're getting yellow color temp appearance on the laptop and external display, but I suspect you're on a Windows system and your graphics card can't have two profile LUTs for the two displays occupy the same graphics chip. This is sometimes an issue with Windows depending on video card brand/model and driver. Don't know enough details to know for sure.
    Macs don't have this problem they've been able to have two LUTs (vcgt=video card gamma type) tag for at least 15 years. The Chromix note below (disregard the utility it discusses) explains what happens on Macs which usually has to happen with a loader every time Windows starts up.
    http://www2.chromix.com/ColorSmarts/smartNote.cxsa?snid=1027
     
  16. Hi Tim,
    I only get yellow on the monitor if I calibrate the laptop first and then choose the "Display Match" option, my assumption had been that that would give me two matched displays, it doesn't.
    If I calibrate the monitor first it gets a completely even toned neutral colour, if I try to link the laptop calibration to that, seemingly good, profile it is way off, if I don't try to link them at all, the laptop is way off.
    I am using a 17" MacBook Pro with both graphics cards turned on.
    Incidentally, the monitor has many profiles in it just via the Displays box in Preferences, if I choose Adobe(1998) it is so close to the calibrated profile it amazing. Truthfully, at this point, I'd be better off choosing a ready made profile for the monitor and matching the laptop to it!
     
  17. digitaldog

    digitaldog Andrew Rodney

    So what is the point of a calibrator then?​
    To target a calibration for a match to a print for most. As soon as you want two, very dissimilar displays to match, you have problems. They will most obviously need different calibration targets to get close, but how close you can get may have more to do with the qualities of the panels than anything else. The calibrator CAN calibrate, the question is, what are you asking for in terms of calibration targets (see:http://www.luminous-landscape.com/tutorials/why_are_my_prints_too_dark.shtml)
    The calibrator is simply a measuring device. The resulting profile built is only based on the calibration targets you ask for (white point, TRC gamma, backlight intensity). Your speedometer in your car isn’t going to tell you that you are going too fast and could get a ticket in a zone, it only tells you your speed.
    Eyeball calibration isn’t going to cut it long term for ah so many reasons. Just like you can’t consistently gauge your speed, the reasons we have instruments in the first place. The calibrator is only as useful as the targets the user asks for. That what you have to work on. But assuming two displays will match is another issue all together.
    if I choose Adobe(1998) it is so close to the calibrated profile it amazing​
    That’s not going to fly.
     
  18. "To target a calibration for a match to a print for most. As soon as you want two, very dissimilar displays to match, you have problems. "
    Why? So which of my calibrated screens is the print going to match? I don't want the screens to match per se, I want both screens to match the print, they can't, by accident if the screens both match the print then the screens must be very similar.
    "the question is, what are you asking for in terms of calibration targets"
    I am asking it to match the print, that, I assumed is the whole point. I understand luminance and gamma and white point, and in advanced mode I set them.
    " Your speedometer in your car isn’t going to tell you that you are going too fast and could get a ticket in a zone, it only tells you your speed."
    What is the point of instruments if they don't do their job? I did not just purchase a speedometer, I bought a speedometer and a cruise control, if I set the cruise control to 55 by the speedometer then I expect to be travelling close to 55, but if I do that twice I expect to travel at the same speed both times, maybe a little over 55, or maybe a little under, whatever the speedo error is, but I expect the cruise control to maintain that indicated speed. It seems the ColorMunki can't do that, it is sold as a cruise control and speedo set, but it can't do that consistently, and lets face it, consistency is the one feature we buy a calibrator for.
    This is not complicated, though there are many aspects to it, the puck is supposed to tell the software what it sees, the software is supposed to tell the computer what adjustments to make to the video output to make that output consistent, the software has control, the screen is capable of outputting neutral tone.
    "Eyeball calibration isn’t going to cut it long term for ah so many reasons. "
    That was my assumption and the reason I invested in a calibrator, but my eyes are considerably more accurate than my current laptop profile, my eyes might not cut it for long, but the calibrator sucked on my laptop out of the box.
    I have read your LL article several times, it was one of the reasons I got a calibrator, but if I set the luminance and the white point, or if I ask the software to match an existing screen, the output should be consistent, it is not.
     
  19. scott, the fact that u want a laptop screen to match another external better display is fascinating. following your logic
    any monitor out there should match each other because u calibrate them? lets buy cheap tva monitor then and match
    them to nec monitor... since whe just have to buy a calibrator.

    2 similar monitor should effectively be able to match each other... similar is the key word.

    according to your material, i would trust my external calibrated monitor, and yes put your laptop to adobe rgb and luminosity in the middle of the little sun... it is the close best thing u could do wit it. a profile will only get it warmer.
     
  20. Patrick,
    With the greatest respect to you and Andrew, and you know I respect you, that is hokum.
    One, my laptop is not a cheap anything, two, it is perfectly capable of rendering a neutral tone, three, my expectations are not unrealistic. I am not expecting the two to render exactly the same colours and tonality, but they are not in the same ballpark.
    I am asking a calibrator to calibrate and the accompanying software to make the adjustments it needs to for consistent output. I don't expect my laptop to render colours as well as my monitor, that is why I bought a monitor, I accept there are limitations to it's output, viewing angle etc etc. What I don't expect is for there to be large caste differences between the two when all the options in the advanced settings of the calibration software have been exhausted.
     
  21. digitaldog

    digitaldog Andrew Rodney

    Why? So which of my calibrated screens is the print going to match?​
    IF you could get both to match the print, they’d match each other so I can’t answer that question.
    Think of it this way. Lets say you have one display who’s max luminance can hit 130cd/m2 but the other can only hit 120 cd/m2. You have no choice other than to lower the luminance of the brighter display to match the dimmer display. Lets say one has a wider gamut. You can’t make the narrower gamut match it. And if there are no controls to alter the gamut of the wider unit, they simply can’t match. So do you dumb down the better display just to match it to the lesser?
    Why do the two very dissimilar displays have to match on the same system? And if you work with multiple systems and the goal is to have each match, you really want to be using the same display system and a reference display (Eizo, NEC SpectraView) at that.
    I am asking it to match the print, that, I assumed is the whole point.​
    That requires futzing with the target calibration aim points until the match results IF you have the necessary controls to do so. Display is too green? If you don’t have the control over xy values to alter that position of the white point in your software, you’re going to live with a green cast.
    What is the point of instruments if they don't do their job?​
    They do their job, you have to understand what they tell you and take appropriate action. If the display is too yellow, you have to set the software for a different (cooler) white point. The instrument has NO idea your display is too yellow if it reports the CCT kelvin is 5600K. YOU have to tell the software and the instrument you need a cooler WP and set the value higher. The instrument will then allow the software to do this. But it has no idea what you want.
    Your speedometer has no idea if you are speeding when it reports 55mph. It has no idea you are in a school zone where the top speed limit is 25mph!
    if I set the cruise control to 55 by the speedometer then I expect to be travelling close to 55​
    But you set 55 in a school zone. Its not the speedometers fault.
    my eyes are considerably more accurate than my current laptop profile​
    Your eyes tell you something isn’t matching. That’s good. That doesn’t mean the system can’t produce a match and do so consistently (which your eyes can’t). The issue isn’t the instrument. Its the user.
    if I set the luminance and the white point, or if I ask the software to match an existing screen, the output should be consistent, it is not.​
    Is that the issue? Consistency? It sounds like a mismatch is the issue. And I suspect its consistently not matching. So could it be what you are asking the software and instrument to provide?
     
  22. If I calibrate the monitor first it gets a completely even toned neutral colour, if I try to link(?) the laptop calibration to that, seemingly good, profile it is way off, if I don't try to link them at all, the laptop is way off.​
    Just for clarity I'ld advise against linking. I don't know what you mean anyway.
    Do each of the display's appearance of white match up are look close to each other before calibration? If so, then choose Native WP for each in the calibration software. If both look off (one's pinkish, the other greenish), then try to adjust the OSD of the external (which should be your main monitor in Extended mode) to look correct and choose native WP in the CAL software.
    If the MacBook Pro's WP is pinkish or off color then go into the CAL software and do a custom WP adjustment that allows different hue choices like an X & Y number entry or slider to make the WP change color. Never mind the number it chooses. Don't give the software a standard target to match to like D65 or 6500K. You need to be able to pick a WP hue through the CAL software much like you did with external's OSD only you're having to do it with the CAL software.
    If your CAL software doesn't allow a custom WP hue adjustment, then you're going to have to pick a WP number or preset that gets it the closest and then go back and adjust the external's OSD to match that then recalibrate/profile the external monitor.
    Just to be clear you need to calibrate, name and save a profile for each display separately and don't link them or allow the software to get them to match. AND do all this in Extended Desktop mode, not Mirroring.
     
  23. Andrew,
    I am struggling to reply to your comments. You obviously have not read mine. I don't understand why, if I use the same calibrator, the same white point, the same gamma and the same luminance I end up with two completely different coloured monitors, bearing in mind my test image is well within the gamut and capabilities of both screens. I think most lay people would. What is the point of a calibrator if I have two screens capable of reproducing an image correctly, I set all variables to the same values and yet end up with two completely different screen colours. I didn't put a question mark there, consider it rhetorical, unless you do actually have a useful workaround.
    I fully accept that my laptop does not have the performance of my monitor, but it is not a ten year old crt and we are not talking about the widest 10% of a colour space, we are talking large hue shifts for a simple grey tone.
    Tim,
    Thanks for some actually useful input, something that seems below some who mangae to ask things like "Why do the two very dissimilar displays have to match on the same system?"!
    To be clear, my aim is not to achieve a match between screens per se, it is to get both screens as close as possible to my prints, in so doing the screens should end up being very close to each other. So dumbing one down to match the other just gives me two inaccurate screens, that is of no value to me at all.
    I am in extended, of course, and do rename each profile created. One other stupid thing about the X-Rite software, it overwrites the last profile by default as it only has one save name, how difficult is it to generate a date or number in there? Pathetic.
    I can't go into a profile and edit it. I am going to have to keep re-calibrating and setting different white points until I get them close enough for my printing. I know it can be done, but I assumed it would be close to an automatic five minute process with the ColorMunki, it is far from it.
    By linking I mean, in the calibration software X-Rite give you an option to match a display already profiled, I assume it is a shortcut to gamma white point and luminance. I have appalling results both linking and not linking so it makes no difference.
     
  24. digitaldog

    digitaldog Andrew Rodney

    I don't understand why, if I use the same calibrator, the same white point, the same gamma and the same luminance I end up with two completely different coloured monitors...​
    Because they don’t behave the same, don’t produce the same output and you are using a scale that is a correlated value to produce those numbers. The only D50 possible is emitted from an object 93 million miles away. The only way to produce 5000K is by heating a theoretical object so high it glows, something that would produce a molten pile of plastic that was once your display. The numbers are therefore meaningless with an assumption that two very dissimilar devices should look the same when the numbers are the same. Its why when someone calibrates a display for D50 and uses a so called D50 viewing both, the two don’t match. This nothing new, its a behavior we’ve experienced for nearly two decades.
    See: http://www.ppmag.com/reviews/200512_rodneycm.pdf
    What is the point of a calibrator if I have two screens capable of reproducing an image correctly​
    I told you want the point of a calibrator is. Its to measure something consistently to set it for a visual preference and then build an ICC profile.
    I didn't put a question mark there, consider it rhetorical, unless you do actually have a useful workaround.​
    I told you the work around and provided a URL explaining it in depth. You adjust the two with what will likely be dissimilar values that you consistently use to produce a visual match. You can give that a try or give up, but the facts remain, that will get you closer but nothing guarantees a match due to the vastly dissimilar products you are using (and why they have to match is still a question unanswered).
     
  25. To be clear, my aim is not to achieve a match between screens per se, it is to get both screens as close as possible to my prints...​
    If you think you're seeing differences between two displays, you should see what different lights used to view the prints that's going to make that goal pretty hard to accomplish unless you're willing to settle for close enough. Since we don't know what you consider non-matching color, I'll submit the links below that shows what to expect in the real world matching two displays to a print viewed under various neutral-ish lights.
    Scroll down to my photo I shot of my G5 Imac attached to my Dell 2209WA, both are mid-quality (compared to Pro quality) displays and both calibrated and profiled showing the standard PDI skin tone section color target...
    http://www.photo.net/digital-darkroom-forum/00WAfF
    Note the Dell skin tone is slightly yellower than the iMac's skin tone.
    Now see these prints shot at different distances from viewer to light under various neutral-ish lights...
    http://static.photo.net/attachments/bboard/00R/00RrBB-99245584.jpg
    Scroll down to the photo of the PDI print in this thread...
    http://www.photo.net/digital-darkroom-forum/00UbBI
    Here's a second version made to look after adaptation kicks in viewing back and forth between 6500K display and supposedly 4700K-5000K lighting...
    http://static.photo.net/attachments/bboard/00X/00XOKx-285623584.jpg ...Solux-top/GE Sunshine-bottom
    How inaccurate is your setup? Is it worse than the samples posted?
     
  26. Andrew,
    I am using a measuring device to measure what is output from an unknown input, it then calculates the corrections needed to the input to make that output a know and consistent quantity. For instance, if I say I want luminance to be 110, even if I set the brightness too high, it reads that, and makes corrections to get it down, so, if I ask it to output D65 temp, it should make corrections to do that. If you do that to two monitors next to each other they should be similar. That is the entire point. It can't seem to "measure something consistently to set it for a visual preference and then build an ICC profile."
    What I think most real world people expect from a calibrator is this. Given a screen that is capable of the tones and colours, gamma and temp within a colour space, after calibration ending up with endless matching screens that are accurately calibrated to user input specs of luminance, colour temp and gamma. They then expect their off the shelf icc profiled prints to fairly closely match the screen if they take the time to optimise their viewing experience re luminance and colour temp. Where my, admittedly budget, calibration is failing, despite the fact that I can manually set the three key points, is in getting capable screens to match. I don't understand why this is such a surprise.
    Tim,
    Thanks for your further help. I was going through the process of altering native white points in the Mac software to get the calibration software to accept a slight change to WP but it has crapped out. The Mac calibration now has a conflict in it and I can't save new profiles. I need to work that out now before I can go further.
    My screens, at the moment, are considerably worse than your posted images.
     
  27. digitaldog

    digitaldog Andrew Rodney

    I am using a measuring device to measure what is output from an unknown input, it then calculates the corrections needed to the input to make that output a know and consistent quantity.​
    No, that’s not how it works. There’s no such correction and more importantly, the values it provides are correlated and not absolute. If all displays produced the same results, you wouldn’t need to calibrate nor profile them. But they are all over the map and worse, the values an instrument provides here, for WP (the area you have issues) is a range of colors. Even using the same instrument on the same device with different calibration packages will produce different results. I’m sorry if you don’t like these facts but facts remain. One package that targets CCT 5500K and other package using the same display and instrument will very likely produce a different result. Your Laptop backlight source is most likely vastly different from the display backlight. Even if they were both Fluorescent lights! Ever walk into a large room lit by differing Fluorescent lights? Look the same to you? And one display backlight might be Fluorescent while the other is white LED. The filter matrices in a Colorimeter may have severe issues with one such light source versus the other (not an issue with a Spectrophotometer).
    The bottom line is, you are making assumptions about the technology and process which is interfering with getting to the solution. Using differing values to produce the same visual match (or a closer visual match). As I have tried to explain more than once, the idea that the same settings will produce the same results on the differing displays isn’t going to work and its quite possible you’ll never get an acceptable match. IF you want a match between two displays, get two identical display systems and ideally, those that take display calibration into full account via software and electronics (NEC SpectraView, Eizo, even an old Sony Artisan CRT).
    if I ask it to output D65 temp, it should make corrections to do that​
    At such a point you begin to understand what D65 is, how your display can’t produce it, the better you will find the path to your solution. A URL was provided that should help you understand this.
    What I think most real world people expect from a calibrator is this.​
    Real world people need education when they make assumptions that are based on a miss understanding of the processes! You can continue down this rabbit hole of you can adjust your thinking and get a closer (although not necessarily ideal) match. Your call.
     
  28. My screens, at the moment, are considerably worse than your posted images.​
    If this is the case then you have a serious software conflict either with the calibration software used, OS X and the video driver or all of the above. All this back and forth debating on calibration/profiling theory isn't going to fix Scott's problem.
    I was going through the process of altering native white points in the Mac software to get the calibration software to accept a slight change to WP but it has crapped out.​
    I don't know what you mean by Mac software and what constitutes crapping out. Do you mean an error occurs? or the WP doesn't stick? Are you using separate calibrators, one for the MacBook Pro and the ColorMunki for the external display? Use only one for both.
     
  29. Tim,
    Yet again thanks for the helpful input.
    After looking on the internet last night it seems I do have a software conflict, it seems X-Rite and Wacom might be tripping each other up, thought the crash log only shows Apple lines of code.
    There is a workaround that worked for a couple of tries but that is now ineffective, I am going to do a system reinstall so will be back after that.
    The ColourMunki software only has options for several set WP's, but one of them is native, I was using the Apple Display Calibration to make an unimportant profile but with an adjusted, custom, WP. If I start with this profile the X-Rite should work from that custom WP. It is what Andrew says he does in his seminal LL article that can be summed up by, I can't get a screen and print to match so I go through the motions of calibration, make a print, then eyeball the screen to match it, read my article it is brilliant.
    By crapping out, I mean in Apple's Display Profile app I can't save a new profile I generate. When I get to the Save This Profile, it quits. I am sure I can achieve my goal, despite the doubting of a world renowned colour expert, when I made a new User account on my computer and did the same process it generated two matching screens, nearly, I have mistakenly put the laptop in 1.8 Gamma so that was not correct, but the colours were fine. I'll just have to spend some time working through the software glitch to get what I want. But I am only using the Apple app to generate a customised native WP, I am using the X-Rite to do the calibration of both screens.
     
  30. digitaldog

    digitaldog Andrew Rodney

    It is what Andrew says he does in his seminal LL article that can be summed up by, I can't get a screen and print to match so I go through the motions of calibration, make a print, then eyeball the screen to match it, read my article it is brilliant.​
    You continue to not get it, but that’s fine.
    I'll just have to spend some time working through the software glitch to get what I want.​
    Don’t hold your breath (or maybe you should...)
     
  31. I found the glitch.
    It seems you can't have both the X-Rite app and System Preferences open with the display option open at the same time while creating and/or saving profiles, I'm not sure which exactly as my main user account is compromised and I didn't want to keep creating new user accounts. This is a shame because it is the best way of adjusting brightness and the most accurate way, via Universal Access, to adjust the contrast on a MacBook Pro.
    What happens is the profiles are not created properly, some are not saved and Apple's Display Calibration app, the only way to accurately finely adjust the native white point temp, quits before saving profiles. I have sent crash logs to Apple but not X-Rite.
    The work around, if you are locked out, is to create a new account and profile from that, save with the option allow other users to use this profile and you are good to go.
    It seems calibrators and software do work as I, and real world people, expect them to. Both screens now match. I just need the printer to arrive and then start playing with print viewing criteria.
    As an aside, as a former marine engineer, if an expert arrived to carry out a calibration job on a piece of machinery I looked after and after calibration he decided to "tweak" his findings by eye to get a result he wanted, one he would be laughed off the ship, and second, he wouldn't get paid, amazing what some people get away with and those commissioning them will put up with.
    Here is a picture of my two screens, they match to a very fine degree, though, of course, the laptop screen is critical to viewing angle.
    00ZHl5-395639584.jpg
     
  32. Glen,
    With some reservations with regards the software implementation, I can wholeheartedly recommend the ColorMunki calibration package. It can get even very dissimilar screens to match. If you do not have access to WP and contrast settings on your computer I would suggest the more expensive i1Display Pro, as the software is more capable of fine tuning, especially useful if you are running a dual screen setup.
     
  33. Now both Scott and Andrew can kiss and makeup.
    There's something about you two's drawn out back and forth hashing out of this subject with Scott, an engineer and a PN member since 2003 with over 4000 postings and assuming a photographer at least that long who's obviously had enough time to learn how calibration and computers work finally fixing it on his own, that just doesn't sit well with me.
    Wow! Scott! You didn't know you can't retain a custom WP adjusting Display Preferences Eyeball Calibrator (which creates a profile that's tied to the custom WP) and switching to another calibration package while still keeping Display Options open? How long have you been working with computers as a photographer?
    I'm going to try and salvage what's left of my labor day weekend.
     
  34. Tim,
    Many thanks, enjoy your weekend :)
     
  35. digitaldog

    digitaldog Andrew Rodney

    if an expert arrived to carry out a calibration job on a piece of machinery I looked after and after calibration he decided to "tweak" his findings by eye to get a result he wanted, one he would be laughed off the ship.​
    What Scott fails to understand is the process of calibration of a device like a display is to set it to some desired output condition. What condition? Well as I’ve said over and over again, for most, its to match a print. Or for two displays to match each other and a print.
    Why does the calibration software have so many calibration options? Lets look at only setting the white point. Good calibrate software provides not only presets for CCT Kelvin or Standard Illuminates but exacting input of xy values. If the idea is, calibration is a simple, standard definition of targets, we would hardly see such options and yet, the higher end packages all provide them. That no display can produce a Standard Illuminant or that any CCT kelvin value you ask for is a large range of colors seems to be missing from Scott’s thinking about the numbers here. Or that two differing software packages will produce differing results when asked to produce the same calibration. Take the math up with the software engineers.
    How does the calibration system know what targets to use? Can you measure the display WP and the booth and get an automatic match? Nope. How does the end user know they have produced the desired calibration? They VIEW the print and display together (correctly I’ll add) and see a visual match. The idea that taking two vastly different devices, one emissive, the other reflective and produce a match by using some predefined targets (when the differences in just one, let alone both products here in terms of what is used for an illuminant is huge), illustrates that simple thinking by Scott in terms of calibration will not work. It hasn’t worked for decades. As I pointed out, its been well known for decades that asking to calibrate a display at 5000K will absolutely not guarantee a visual match to a “5000K” viewing booth. It almost guarantees a mismatch!
    Calibrating a display to match a print (or another dissimilar display) is not calibration of a boat’s machinery, speed, weight or whatever you think you are calibrating because there are absolute measurements here that are attempting to come up with one result (what does this boat weigh, how long is it, how much fuel did it use to go from here to there). There is no correlation of vastly dissimilar objects here to deal with to produce a perceptual condition we call a visual match. If all this were that easy, every display could be calibrated at the factory. They would all match each other rather closely. An LED white sRGB backlit $100 laptop panel and a CCFL $2000 high bit wide gamut panel would produce the same visual match when calibrated using the same targets or out of the box. They don’t.
    Scott, take up the issues above, issues that are known and expected with whoever you wish, the facts are, YMMV when calibrating a device and that you will need to visually adjust to produce a match (if and when you get that printer setup). You’ll have to soft proof and now deal with the possible disconnect between the two profile tables that may not sync up (one table affects your output, one affects the soft proof). You can laugh all you want that they don’t match when you use the same settings, facts are, depending on why and what built just the print profile, you can end up with a mismatch. Likewise, you may find the Perceptual table doesn’t match well but the RelCol does. Or that your lighting and possible OBA’s in your paper will hose the appearance of neutrals or the WP of the print. You can place your calibrator and ask for a specific target till the cows come home, you will end up with a mismatch.
    The analogy of calibrating machinery on a ship and calibrating a complex system of dissimilar devices with the goal of a match shows that you have to do a lot more reading and thinking on this subject. I’ve tried to steer you in the right direction. I can’t do much more.
    I would suggest the more expensive i1Display Pro, as the software is more capable of fine tuning...
    Ah, yes, by visually altering the values to produce a match. That did produce a laugh.
     
  36. digitaldog

    digitaldog Andrew Rodney

    Two screens, different capabilities and technology but they match
    That you have two differing zoom ratio’s is yet another area you need to investigate!
    Yes, they appear closer, that’s good. But your methodology is sloppy. Dissimilar sized but identical images are not the way to view a color match (or mismatch). Having the two in full screen mode such that only elements that are color managed in a color managed app is also recommended. Having the laptop keyboard backlight off would also be a good idea...
     
  37. Your assumption is that the illustrative image was an accurate representation of my methodology. It wasn't, it was just a quick shot to show where I was, like Tim's image in a thread he linked to earlier.
    But I really have no interest in furthering a painfully drawn out thread. I'd love to take the time to talk with you more, but on this occasion I have work, and a life, to share my time with.
    Take care all and thanks for the input, Scott.
     

Share This Page