Monitor ICC profile - confused

[enaiman]

Yesterday I've got hold of a i1Profiler and I thought I'll try to get a correct profile for my monitor.

After the job was done, I tried to see what would be the difference between what I see on a non profiled monotor and the newly profiled one.
I was surprised to see such a difference - have a look for yourself:



There is a saying: "When you have only one clock, you always know what the time is. When you have two, you never know."
You can see why this saying applies to my case.
I have always seen the picture as the monitor on the right shows it (the not-profiled one) and not only on this monitor. I have seen it the same on my camera's LCD screen, on my home monitor, on my Android tablet and on my IPad. On all these devices, the picture looked like the one on the right.

Now, the left monitor (profiled) shows the colours more vibrant, a lot less brightness, more contrast and the colours are changed.

Now, I don't know which version is the "true one" - that's why I said I am confused.
- is my picture "really looking" like on the left side and because of default profiles I see it washed out?
- is my picture looking like the left side and the new colour profile is not "true"?

Because on so many devices it looked like on the right - I tend to believe the 2nd version is right (right side = correct).

I don't know - I hope somebody can help me with this.

Thanks,

[Simon Garrett]

It looks as though you're using Windows Photo Viewer to display the images in both cases; is that right?

In Windows 7, the Windows Photo Viewer is colour-managed, so should be displaying the images correctly. Provided the monitor is correctly calibrated and profiled then the image should be correctly displayed by colour-managed programs. It should be more accurate than on uncalibrated/unprofiled monitors, even if several look the same!

I guess you've tried calibrating and profiling more than once, in case something went wrong. If you know someone else with a calibrated/profiled monitor then you might see how it displays on their monitor.

PS - edited to add: if you're not knowledgeable about colour management, see the tutorial info here on this site at https://www.cambridgeincolour.com/co...t-printing.htm

[Manfred M]

The problem with out equipment is that we get rather used to seeing things a certain way and become comfortable with it. I had a similar result as you did because my monitor / video card combination had a blue cast that was not noticable until after I calibrated and profiled the system.

One thing to remember is that your application has to be colour managed, i.e. it needs to use the icc profile you created to give correct results, so a non-profiled application cannot be guaranteed to display properly. Your camera's display is definitely not profiled, and your Android device is not either, so they cannot be used to judge the colours correctly.

If you are concerned that the profile is incorrect (errors can happen), review the manual that came with your profiler and ensure that you have set up your monitor correctly (contrast and brightness settings, etc.) as per the instructions and reprofile again.

[bambleweeney]

I've profiled mine several times, each time with a slightly different result What I do after profiling is to use a black and white target image to check that at least I'm getting the maximum tonal range possible with no strange colour shifts anywhere in the tonal range. Have a look at the northlight site (Google it), full of good advice and from there you can download several target images, all monochrome but at least you will know if your losing shadow or highlight tonality. If anybody knows of any colour target images that are as effective - now's the time to chip in ...

[Mike Buckley]

I don't know anything about how profilers work, but I have heard that software profilers are not as effective as hardware profilers. I don't know if that is true but it makes me wonder if that is your issue.

I have used Spyder 2 and Spyder 4 on my monitor for years. Both systems display a before-and-after effect. The differences are slight and they are consistently the same each time that I run it every two weeks. When using the Spyder 2 on my Windows 2000 system, running the profile slightly increased the contrast. When using the Spyder 4 on my Windows 7 system (Spyder 2 isn't compatible with that OS), a slight green cast is eliminated. That's using the same monitor on both computer systems. The differences in the before-and-after displays are nothing compared to the difference you're experiencing.

[Andrew1]

You don't say which flower looks like the original. Also, if these photos are indicative how how far off the reds were/are then the skin tones in your people shots would be just as bad. By the way I too get varying colours depending on when I calibrate my monitor. My monitor is in a small east facing room with taupe walls. The combination of light creates a problem with the ambient light measurement. Because of that I only profile in the evenings in a dark room. I have a relatively good screen and have not noted any degradation so I only calibrate it about once every four months. Before and after shots show negligible difference. I've never used one or know if it's appropriate for your problem but do a quick check on the Colorchecker Passport. It might help.

[Arlen]

Color management is fairly complicated and lots of things can affect it. Reading some of the tutorials suggested would be a good start. Also, I think you will get advice that is better targeted if you give us as many details about your system as you can. For example, your computer's operating system (Windows 7?), your monitor model, the applications you are using (or could use) to view the image (do you have Photoshop?), how you took those screen shots, etc.

[xpatUSA]

The screen views looks to me as if the image on the right is that of either an Adobe or a ProPhoto RGB image being rendered as if it were sRGB. Surely it cannot be that simple?

[edit]
just remembered that I've read somewhere that, if two monitors are connected to the same computer, only one of them gets profiled "properly"?
[/edit]

[HMW12]

I read that as well and its because the profile gets loaded on the graphics card. I used to use a monitor with my laptop and until I read that I could never figure out why the displays would not match .

[enaiman]

Thank you all for your input.

Because I am checking the replies quickly before work I cannot answer them properly. I'll give a more detailed answer in a couple hours.

Edit: what I can do quickly is to post the image here SOOC; hopefully, somebody with a calibrated monitor can see how it looks like (like the left part of my first pic in the thread or like the right part).

[Mike Buckley]

Your image on my calibrated monitor isn't similar to either monitor shown at the beginning of the thread. It's in between the two, though closer to the monitor on the right.

[Arlen]

Same thing here: intermediate between the two versions in the original post, on my calibrated and profiled monitor.

[Manfred M]

just remembered that I've read somewhere that, if two monitors are connected to the same computer, only one of them gets profiled "properly"?

This is not correct; I have two different monitors and both have been profiled. There were some issues with early versions of Windows. I know this was an issue with Windows 2000. This is certainly not an issue with Windows 7, as this is what I am running and I don't know when this was fixed. but I've been running dual monitors starting with XP.

If you do not profile the second monitor, it will use the default profile.

[enaiman]

Thanks Mike and Arlen for your help.

On my home monitor (not calibrated) it looked as you said - between the two of them. Now I am pretty confident that my monitor is good enough to not let me make big mistakes in PP

The first image was taken with my mobile phone, the window moved between the screens to show half and half.

[Arlen]

Glad to help. Correct profiling is most important when you are printing. Everything in your workflow needs to be properly profiled and color managed, or you will drive yourself crazy trying to get prints that match what you see on your monitor. But unless your monitor is way off, it's not as big an issue for web images, since everyone else is likely to see it a little differently on their (usually uncalibrated) monitors anyway.

[Manfred M]

Glad to help. Correct profiling is most important when you are printing. Everything in your workflow needs to be properly profiled and color managed, or you will drive yourself crazy trying to get prints that match what you see on your monitor. But unless your monitor is way off, it's not as big an issue for web images, since everyone else is likely to see it a little differently on their (usually uncalibrated) monitors anyway.

I'm not sure I quite agree with your statement, but I do understand where you are coming from. Profiling your monitor means that you have a consistent base that is "correct", so regardless of what other people are viewing your image on, your work will be consistent. If viewed on another profiled screen, the viewer should be seeing exactly what you meant to show.

From a printing standpoint, I would agree that a profiled screen gets you close; but I find that I end up using Photoshop's soft proofing functionality to emulate what the printed image will look like and use this information to tweak my work before printing. Remember that your screen is a RBG (additive) / transmitted light device, while your printer uses a CMYK (subtactive) / reflective process, with your whites coming from the colour of the paper stock you are printing on.

[Arlen]

If viewed on another profiled screen, the viewer should be seeing exactly what you meant to show.

Agreed. But in the great big world of the interwebs, only a tiny percent of monitors are properly profiled. I would be willing to wager that even here on a photography site, a large fraction of monitors are not profiled.

From a printing standpoint, I would agree that a profiled screen gets you close; but I find that I end up using Photoshop's soft proofing functionality to emulate what the printed image will look like and use this information to tweak my work before printing.

Yes, same here. And now Lightroom 4 has an even better implementation of softproofing than PS.

[Manfred M]

Agreed. But in the great big world of the interwebs, only a tiny percent of monitors are properly profiled. I would be willing to wager that even here on a photography site, a large fraction of monitors are not profiled.

Agreed. On the other hand it feels good when someone says "that image has a blue cast" and I can say, no it doesn't, it must be your screen", knowing that I am right.

It is a shame that the weakest link in colour management is the tool that everyone uses.

[Markvetnz]

What amazes me is the number of people who still question whether using a profiled monitor makes a difference. In our camera club we have about 40 members. Buying a Spyder 4 would cost about $10 each, but the idea has gone nowhere. Every time I show people the difference between a profiled and non profiled monitor they are gob smacked. In the land of the blind the one eyed man is king.

[Colin Southern]

What amazes me is the number of people who still question whether using a profiled monitor makes a difference. In our camera club we have about 40 members. Buying a Spyder 4 would cost about $10 each, but the idea has gone nowhere. Every time I show people the difference between a profiled and non profiled monitor they are gob smacked. In the land of the blind the one eyed man is king.

Hi Mark,

Just be a little careful with the "camera club" approach -- I'm pretty sure it would violate the license agreement (plus could be difficult because it's an activated product).

In reality, calibration & profiling can make very little difference or it can make a huge difference -- or something inbetween. In my experience - usually - colours are reasonably close in most monitors (being basically standardised devices), but I've seen variations in black and white points of over 2 stops, which translates into HUGE differences in the photos. The big problem appears to be though that (a) many folks "put it in the two hard basket" because they don't understand it, and (b) most don't realise how far out their images are because they only ever see them on their own screens and of course in that scenario they're adjusting them to compensate without realising what's happening.

So I'm with you 100% on this one -- if they're not working from a calibrated and profiled monitor then they're playing Russian Roulette with their images without ever realising it.