Spyder 3 Pro - Color space question

Err I don't think you understand how things work on the whole. The resident program is just a quick management tool and all calibrators come with them. What it does is ensure the correct correction curves are loaded into the video card / monitor LUT assuming it supports hardware lookup tables. Unless you're doing something very strange with your calibration (like trying to force a monitor to be sRGB when it's not which is definitely not the right way to go about things), then things run like this:

- The resident program checks to ensure the correct curve is loaded into the videocard / monitor LUT, as well as ensure the correct profile is set in windows.
- The video card's output will ensure consistent colour across the tonal range and set your white point to the one you selected. If you have a monitor with hardware lookup table then this is handled by the monitor and the video card output remains as per normal. This only provides consistency and temperature nothing more.
- The software you are using is still required to recognise the colour profile to know just how red that value of red is and what it actually should be. What the software will assume is that your monitor colour is consistent (something the video card LUT is ensuring).

Short answer: All calibrators come with a little program like that. All software needs to have calibration turned on. This is far more important if your monitor doesn't have an sRGB gamut.
 
ok, thanks a lot - it clarifies. I naively thought all adjustements could be made in the video card, hence did not understood the need for these two layers of adjustement. I made some research over the last few days and you're the only one who came with a good explanation !

Something interesting though: I just calibrated two different DELL monitors (laptop Latitude E6410 and monitor SynchMast 225BW) under the same dim light and the pictures are significantly different in saturation. Both were set to factory default. By significant I meant changing the mood of the pictures - one punchy the other desaturated, the equivalent of a +20 in PS. But still, calibrating brought the two much closer.

I repeated the experiment twice, same results. I will have to print to see which one is the best... and let's hope the shop has calibrated printers ;)

- The resident program checks to ensure the correct curve is loaded into the videocard / monitor LUT, as well as ensure the correct profile is set in windows.
- The software you are using is still required to recognise the colour profile to know just how red that value of red is and what it actually should be. What the software will assume is that your monitor colour is consistent (something the video card LUT is ensuring).
 
Just a quick guess was the Latitude E6410 the desaturated one? I use a Dell Latitude E6400 at work and it has to have the worst screen I have ever used.

One thing to note is that a profile will only define the limits of the colour gamut for the display. Say you have a wide gamut display and a normal display and you are opening a standard sRGB image in an application like photoshop that understands colour profiles. The image should look identical on both monitors. However if you have a very saturated and colourful AdobeRGB image then it will look far more saturated and punchier on the wide gamut monitor which is capable of displaying the full colour range.

The same could happen if you have an sRGB monitor and a monitor that isn't even capable of displaying that gamut and open a sRGB image, or surf the web on a colour managed browser. The actual sRGB monitor will look punchier and be the accurate one of the two.

I'm not sure what software you're using to calibrate, but often it'll draw the gamut triangle for you. Can you see if there's any appreciable difference between the gamuts in the profiles using that software?

By the way you're not comparing the Syncmaster connected to the Dell are you? Because try as I might I've never been able to properly get colour management software working on a laptop video card. One of the screens will always drop out or both screens end up with the same curves or something stupid. Mind you I've never tried very hard.
 
Actually no :) The dell laptop E6410 is the most saturated. I don't feel it is as hazy as my previous dell, 3 years old. Maybe they made some progress.

But still - I am able to saturate more the picture on the Samsung SynchMaster 225BW; do you think it can still be a gamut limitation of the device then ?

I use SpyderExpress 3 for calibration. The monitors have different computers. Out of curiosity, I could try to link the synchmaster to the laptop - just can't find the cable ')
 
I'd finger a problem with the calibration profile not being loaded then. Using photoshop on both?

Open any image and click edit -> Colour settings, and then click the down arrow next to RGB working profile. Take note of (BUT DON'T SELECT) what Monitor RGB is. It's name will be what monitor profile was loaded. If it's sRGB then something is wrong.

Basically a good way to test the absolute saturation is to create an image with the sRGB working space and just fill it with a rainbow gradient.
 
thanks for helping troubleshoot.

I am positive it is loaded... because in picasa I can turn color management off and I do see the difference; I also do see the difference if I turn the resident program on or off. Can also see "Monitor - <profile>" in photoshop.
 
ok... I solved my problem :)

I disabled the "ambient light sensor" on the laptop, then re-calibrated after that. It was the source of the problem.

I made prints at adorama.com, they were too much on the yellowish side. But then tried my local shop and it now almost matches my screen !!

Thanks a lot for all the help !
 
:) Good to hear. Technology is getting way too clever for our own good. For its own good actually :)
 
Hi Padang,

can you post your .icm profile that you calibrated for your e6410?

thanks!
 
Sorry, I don't agree with the Garbz's point. Maybe I'm wrong. But I really don't see any benefit using the spyderpro like this way.

After the calibration, new ICC profile did apply the change to all non-colorprofile aware programs like I.E. 6 or FF2.x. or Window Picture & Fax viewer. However, PS & ACDSEE , color-profile aware program, have the options not to use customized profile, but it doesn't mean it's the correct way. (e.g. still set working color space to sRGB)

After the calibration, windows color profile has been changed to customized ICC file.
Then I did a test. There is a sample jpg in C:\Program Files\ColorVision\Test Image\PDI Test Image.jpg. (Spyderpro 2.35) Everyone can remember the last step during the calibration. Spyerder pro shows a picture comparison between before/after calibration. That it is. If I open this picture in a non-colorprofile aware program. (IE, irfan-viewer). which will show the same color as the after-calibration image in sypder pro. which means spyder-pro did help non colorprofile aware program show more accurate color.

Then I use Photoshop to open that picture using sGRB working colorspace, the color is washed and dimmed, which is not acceptable. It means Spyder's ICC file is not in use then. The difference is very apparent which can be identified by many users here. However, to change the working profile to spyder's ICC profile. PS show the same color as non colorprofle aware program does, which is also the same pic as spyderpro calibration program shown to us after the calibration. So to set PS's color working-space to monitor customized profile can help editor adjust the picture in calibrated environment, produce more accurate colored picture (Sample pic proof) and same result viewed in different programs. The final goal is to let us get the same (or closest) color display of the same picture, whatever picture viewer programs you and your friend use, and on whatever calibrated monitors and pcs.
 
Last edited:
You are confusing two issues, that of the colour profile, and that of the correction curve.

When you calibrate your monitor two things are done.

1. The output of the video card (or the lookup table of the monitor) is modified to make grey tones read the exact selected colour temperature (often 6500k) and tonecurve (nearly always gamma 2.2). This ensures accurate color tracking thoughout the range of the monitor. This is applied by programs such as AdobeGamma Loader or any number of custom programs that will apply these curves when the operating system starts. This correction is available to ALL programs by virtue of the fact that the video card lookup table or monitor lookup table is directly modified. But this only gets you half of the way there.

2. The other thing that is done is a colour profile for your device is created. This colour profile has information about the absolute colour properties of the device. Just how red is RGB(255,0,0)? Just how green is RGB(0,255,0). This colour profile is set as default in windows, however don't confuse this with it actually having any effect. The next part of the equation relies on software itself to call functions to the Windows Colour System (WCS). WCS has APIs that provide programs with information about the device's colour system, as well as APIs that will handle colour conversion IF they are used.

Now in many cases if you have a monitor that conforms to the sRGB colour space then I would agree step 2 is redundant and all the benefits of calibration happen as soon as that before and after shot is seen. However if your monitor does NOT conform then the after shot will be wrong depending if the program recognizes how to correct for it or not (the BasICColor for instance doesn't apply the the colour profile to it's preview, just the initial correction).

So accurate colour is a two step process. The Monitor needs to display the correct colour for every input value (step 1), and the program needs to know what colour the monitor displays for every value (step 2).


Now your biggest issue. The WORKING Profile has absolutely NOTHING to do with colour calibration. I can have 100% accurate colour regardless if my working profile is sRGB, AdobeRGB, my Monitor profile, any number of CMYK profiles, or even a greyscale profile. The working profile defines ONLY the limits of what can be displayed in a given image, and how the data is stored in the image.

Now here's an explanation for your likely results which make you think this way: When converting an image between two working profiles you should only ever see a difference if you're going from a larger to a smaller one. Also if Photoshop did not correctly load your colour space then it will assume your monitor is sRGB and simply output the numbers with no correction what so ever.

If you monitor is different from sRGB and Photoshop is not reading the monitor profile from windows (read my posts above to find out how that is) then the picture will look wrong if the working profile is sRGB. Now if you set (set, not convert, just set) the working profile to the same as your monitor profile, things will look just great... On your computer and on your computer alone.


So in summary:
- Colour profiles is a two step process involving a correction curve and a colour space.
- Working profiles have nothing at all to do with displaying a correct colour providing your colour system is working correctly.
- Working profiles should at all time be standard, either IEC sRGB, AdobeRGB, or ProPhoto, or if you work for a magazine / newspaper one of the CMYK spaces. There is never a good reason to set your working space to the profile created by your Spyder.
- Applications must be colour aware for things to work properly.
- If this doesn't work, DON'T fudge the settings through the working profile, it may look good on your screen, but your colour system is screwed somewhere and will put you in a world of hurt when you send your pictures somewhere else.

- Once should NEVER have to set a different working space in Photoshop. If this ever has to be done then something has gone wrong along the way (someone sent you a non-standard picture with no embedded profile, scanner software setup incorrectly, etc). Colour spaces should only every be converted, and a colour space conversion should nearly always look identical to the original save for obvious things like the target colour space being much smaller than the original and the monitor colour space.
 
OK, old post but I have the same issue. Garbz thanks for your expertise.

But Photoshop photos look washed out unless I View>Proof using the Monitor Profile. The same photo in Firefox is much more vibrant.

Is that washed out Photoshop color what most people will see or do I have something setup wrong? I don't understand if Firefox is handling color correctly why it doesn't look the same in Photoshop.

Thanks.
 
OK, old post but I have the same issue. Garbz thanks for your expertise.

But Photoshop photos look washed out unless I View>Proof using the Monitor Profile. The same photo in Firefox is much more vibrant.

Is that washed out Photoshop color what most people will see or do I have something setup wrong? I don't understand if Firefox is handling color correctly why it doesn't look the same in Photoshop.

Thanks.

Sounds like you have something set up wrong. The same photo viewed simultaneously in Photoshop and Firefox should appear identical with Firefox set to clr_manage value 1.

Photoshop should be using the display profile by default and if you go to View > Proof and set Photoshop to use the display profile then what you're seeing shouldn't change in the slightest since Photoshop should be using the display profile to begin with.

Here's a screenshot of PS and Firefox (value 1) on my system displaying the same photo.

clr_manage.jpg


May be a conflict with dual monitors?

Joe
 
Hey Joe,

If I have Windows Color Management setup to use the Spyder profile(s) and I open a photo in Photoshop when I change proofing between srgb and Monitor Display there is a huge saturation difference.

Look at this attachment. Toggling the proofing back and forth (Monitor Profile looks like Firefox) and sRGB looks like the Photoshop or Windows Photo Viewer.

Compare.jpg
 
Hey Joe,

If I have Windows Color Management setup to use the Spyder profile(s) and I open a photo in Photoshop when I change proofing between srgb and Monitor Display there is a huge saturation difference.

Look at this attachment. Toggling the proofing back and forth (Monitor Profile looks like Firefox) and sRGB looks like the Photoshop or Windows Photo Viewer.

View attachment 134934

There we go -- you should not proof to sRGB. You should not proof to a monitor profile either. Soft proofing is for output devices.

Joe

Edit: When you remove the soft proof from Photoshop you should see your photo the same as in Firefox. That's your calibrated display and that's correct.

Windows Photo Viewer will display your photo incorrectly because it does not properly use the Spyder display profile -- that you can't do anything about.
 
Last edited:

Most reactions

Back
Top