Confusion on Profiles, both Printer and Monitor

Discussion in 'General Technical Discussion' started by Kevin Scott, Jul 19, 2005.

  1. I'm hoping someone might help me to better understand the relationships between monitor and printer profiles.

    I recently purchased a monitor calibration device (EyeOne mentioned in this forum). I ran the software and created a profile for my monitor which the software automatically assigns. Previously I had done the simple Adobe Gamma(?) calibration and this new one makes a difference. (Btw, there's a huge difference now between viewing the photos at home on the calibrated monitor and my laptop at work--4 yrs old and uncalibrated!! I was shocked at the difference in viewing my own photo postings here!)

    Anyway, so I just started to print out some of my photos. I worked up a photo in Photoshop(v7), used the embedded profile (sRGB) as opposed to converting to my recently calibrated profile, and then printed via Photoshop. In the dialog box, I decided to select the sRGB profile as I figured if the printer used that profile it would match my screen. Wrong! It wasn't terrible but just much darker. I brightened up my image some and reprinted using sRGB profile for the printer and it was satisfactory.

    In the future, as I become more serious about printing, I'd like to understand the relationship between the monitor profile, sRGB/AdobeRGB of the image from the camera and the printer profile. How does the printer profile relate? I'm a bit confused because I don't see the relationship between the profile of a printer and what I see on my monitor. Or does each profile calibrate to a standard of some sorts so they should all match?

    I'm also not sure when I open an image from my camera, should I be converting to my monitor profile or leave it as shot?

    If this subject is too complex for discussion here, any links to material on the web would be helpful.

    With many thanks,
     
  2. Gale

    Gale

    978
    Jan 26, 2005
    Viera Fl
    Join the club.. thats why I have Joe at access photo graphics print all my stuff. PS guru and great prices and great delivery .. No problems, no profiles, no ink... Home free no worriew.

    Have fun.

    Ya should come over some time and I will introduce you to him. He is also a PS guru and one hell of a phogographer
     
  3. heiko

    heiko

    May 15, 2005
    Israel
    Hi Kevin,

    I had my own tough time getting good results when printing on my Epson 2100 (2200 in the US). And it was all about the profiles.

    I found the following link to be helpful:

    http://www.earthboundlight.com/phototips/printing-without-pain-part1.html

    One thing to watch out when printing is to NOT have two programs do the color management. You could, for example, set Photoshop to do the color management (which you may usually do) and also specify an ICM profile in the printer driver. This would apply the color corrections twice, leading to some more or less unpleasant result.

    Regarding sRGB and AdobeRGB, these are color spaces. They define, or better limit the tonal range of your picture. The sRGB color space was defined to meet the capabilities of the most ordinary monitor, and it's the default color space for all websites (just for the reason so that anyone can see the different colors and tonal differences) - it certainly isn't the best to use for printing. AdobeRGB is an enhanced and enlarged color space, providing a wider color gamut. Pictures taken in AdobeRGB usually have less pop, but show a much wider tonal range. Most photography guides will advise you to shoot in AdobeRGB, unless you only need the pictures for web display.

    AdobeRGB pictures need to be converted into sRGB before posting them on websites - otherwise they show up dark and sort of washed out. In PS you need to use the Convert to colorspace (or so) menu item (don't assign color space, unless the picture was shot in the color space you want to assign).

    When you open a picture which was shot in, say, AdobeRGB, make sure that you are also using this color space within PS. Also, don't convert from AdobeRGB to sRGB and back - this will cause a lot of tonal detail to be lost. You could use AdobeRGB all the way through post-processing up until the very last step when you may want to produce a jpg and convert it to sRGB for web display.

    Also, don't try to bring / send pictures in AdobeRGB to your print shop - at least here they can't process this color space and the photos will look quite aweful.

    Finally, color space (sRGB, AdobeRGB, etc.) has nothing to do with printer or monitor profiles, although it sometimes may look like. For example, if you try to view a picture in a program that is not color space aware, AdobeRGB will look weird and sRGB will look good. This has nothing to do with the monitor calibration - you only need to open it in PS and set the correct color space (if it's not taken automatically from the embedded color space profile). The monitor profile adjusts the monitor to display the correct colors that it gets from the application (like PS). The printer profile, set for a specific paper type, will adjust the ink output.

    Well, there are entire books written about it, and I myself have just started to get a little idea of what's going on - a lot thru trial and error. Hope this will help you a little.

    Regards,
     
  4. Hi Kevin,
    Like others have said, this is a topic on which many books have been written. There is quite a bit of information on the web as well. This is one such link, I found the diagram under the section "Color managment basics" to be simple and to the point. You might find it helpful. http://www.normankoren.com/color_management.html. As your curiosity grows, you might want to check out a more detailed diagram on the same page :D
     
  5. Thanks, Heiko, for your response. It helps to clarify some things for me. I checked the link you provided and that answers the question I had regarding how to properly print from within Photoshop. I only read the first page but have bookmarked it to continue reading later.

    Appreciate your assistance!
     
  6. Hi Vinod! Thanks for the link on color management. That looks like it will help tremendously. At least now I understand more precisely what a profile is. I'll be visiting that site frequently to continue to learn color management.

    Thanks for taking the time to respond!
     
  7. Iliah

    Iliah

    Jan 29, 2005
    nowhere
    Colour spaces define not only limits of reproduction of colours. Most important, they define meaning of the colour. We can construct a colour space where R will actually mean blue, or R=255 will have zero lightness, to give some extreme examples.

    RGB colour is characterized with 3 numbers, but those numbers are not universal. If you take some (solid, to make it easier to observe) colour and convert it to another colour space you will see the numbers change (Info palette in Photoshop shows that change)

    It is the same as saying "The distance is 40". 40 what? Miles, leagues, parsecs, light years, kilometers, Ångströms, some non-linear non-Euclidean measure in artificially constructed N-dimensional space? To understand the meaning we need to know some scale, some measure function the space is built around.

    We have different paints labeled as "red". They often have different colours in reality. So, to convert from one red to another one red, we need some rule. sRGB red may well be very different from your monitor red, and from your printer red. What colour management is doing - it converts colour numbers given in one colour space to perceptually same colour in another colour space, calculating new appropriate numbers for that matching colour.

    Usually we have four profiles to deal with:
    - input profile
    - working profile
    - display profile
    - output profile

    All except for working profile are non-standard, they need to be determined through the process of profiling. That's why we calibrate camera (input profile), monitor (display profile), and printer (output profile).

    So, our colour management looks like this: we open an image in post-processing application and convert it from input profile to working profile (sRGB, Adobe RGB, or some other). Good image processing application shows the image through monitor profile, converting it on the fly from working space to monitor profile. We do not convert the image to monitor profile ourselves, image processing application takes care of that just for accurate viewing purpose. We do not edit images in monitor colour space, as it is not a neutral one. That means in monitor colour space same values of R, G, and B numbers usually do not give neutral gray. Also, monitor colour space is pretty limited. These two limitations have numerous implications, and starting with Photoshop 6.0 editing in monitor colour space is something only most stubborn printing houses do. We edit our image in working colour space enjoying WYSIWYG (more or less, depending on the quality of the monitor and its profile; as well as videocard and cables, proper lighting, or better - dimming :) of the computer room and the colour walls are painted to, magnetic fields present around, and amount of coffee we had) due to the job colour management is doing behind the scenes. Next, when it is time to print, we convert our image to printer colour space; or - if the image is meant for web, to sRGB colour space.

    Shortcuts: in many cases we do not have a possibility to profile camera, and trust manufacturer in that the camera generates images in some pre-determined, generic profile, like sRGB, or Adobe RGB. If we do not have custom-made output profile, we trust printer manufacturer in supplying acceptable quality generic printer profiles. If the image from camera is coming in sRGB, and our working colour space is sRGB, we do not need to convert at input. The only conversion we need to take care of is conversion to printer colour space.
     
  8. Thank you Iliah for the very detailed response. This, along with the web links is helping me to make sense of the profiles and understand what's going on.

    I was never really clear on what exactly a profile did but it seems as though it is a mapping mechanism. This seems to be the "rule" you're referring to above.

    What you wrote also helps me to understand how the monitor calibration profile enters in the equation. I wasn't sure if I was supposed to be using that profile when I open the photo in PS. Now I see there's no need to do that.

    Currently, I'm working in sRGB mode but plan to move to AdobeRGB soon. I wanted to make sure I understood how these all relate before making that move and attempting to accurately print a file once processed.

    I appreciate everyone for offering feedback on this. I'm further along on my way to understanding! :) Thanks again!
     
  9. heiko

    heiko

    May 15, 2005
    Israel
    Kevin, allow me to jump on your bandwaggon with some further questions on this subject.

    Iliah,

    You mentioned an aspect I haven't considered yet: In-camera conversion to color space. I took it for granted that the RAW data has been recorded according to some color space, for example sRGB or AdobeRGB. If I read you correctly, the RAW data is the same no matter what color space I select on the camera, and that the post-processing (NC or PS) does the conversion into a color space. In other words, I could convert the data into a different color space, for example CMYk or other.

    In other words, on a D70, if I shoot in AdobeRGB, would the RAW data be essentially the same as when shooting in sRGB?

    Perhaps I misunderstood this point.

    Another aspect for selecting color space is the way some post-processing software works with regard to image manipulation. It looks to me that NeatImage, for example, is best used in YCrCb space to do the noise filtering. Same goes for PS when doing some noise reduction.

    Is there a general recommendation as to which color space to use throughout post-processing?

    Regards,
     
  10. Iliah

    Iliah

    Jan 29, 2005
    nowhere
    I hope not. It was about conversion in post processing, as we are discussing RAW I think.

    There is no such thing as RAW colour space, at least it does not confirm to CIE standard. Sensors do not have gamuts.


    Yes you are right on both accounts

    yes, it will be absolutely the same. Colour spaces and modes are just tags in NEF files, like instructions what Nikon software is supposed to do with the NEF when converting it to tiff, jpeg, or transferring to another application. Other then that, they do not influence the captured data.

    Noise reduction in the space which separates luminosity from colour is beneficial if the algorithm is properly coded. YCbCr is one of such spaces, Lab is another one. In RAWMagick Lite we also use this approach for sharpening and noise reduction.

    No. Depending on the task, we use RGB, CMYK, and Lab colour spaces at different stages of processing, and even blend channels from different colour spaces. All colour spaces are one, as Dan Margulis put it. If the question is about what RGB colour space I use, most of the times it is BetaRGB - you can download it from http://www.brucelindbloom.com[/quote]
     
  11. heiko

    heiko

    May 15, 2005
    Israel
    Thanks Iliah,

    This has been very helpful. By the way, when is your RAWMagick Lite being released?
     
Loading...
Similar Threads Forum Date
Some confusion about shutter speed General Technical Discussion Nov 3, 2014
Honestly, VR concept still confusing General Technical Discussion Mar 30, 2011
PhotoMechanic color profile, anyone? General Technical Discussion Feb 18, 2011
Bracketing Confusion General Technical Discussion Oct 29, 2010
DVD questions/confusions General Technical Discussion May 10, 2008