Calibration

T

tiler65

Suspended / Banned
Messages
10,544
Name
Tom (I think)
Edit My Images
Yes
This thread has thrown up a puzzle.

http://www.talkphotography.co.uk/forums/showthread.php?t=465602

Is there a definitive way in which a monitor can be calibrated or are all monitors and calibrating devices variable that much.

Is is possible that one of the reasons we get so much confusing feedback regarding colours/shadows/highlights because there isn't a standard?

If anyone can shed any light (pun intended) about how this can be achieved, then I am all ears.
 
Last edited:
I'm really interested in the answer to this too, it seems people with supposedly calibrated monitors are seeing dramatically different images in my thread Tom has linked to above.
 
Maybe nobody cares if it is right or wrong.
 
i can just about see the tats on my calibrated screens, what settings are you guys using when calibrating?
i've only got my luminance set to 80, white point of d65 and gamma set at 2.2
 
ljphil78 said:
i can just about see the tats on my calibrated screens, what settings are you guys using when calibrating?
i've only got my luminance set to 80, white point of d65 and gamma set at 2.2
Click to expand...

For what it's worth, this pretty much describes the settings on my calibrated Spectraview 2690 and I can make out the tattoos although they are not immediately obvious.

Anthony.
 
Provided you use a decent calibration device you should be able to match multiple monitors. The maximum I've ever had to match is 8 and it wasn't that difficult provided you are using the same aim points, such as Gamma 2.3 and D65 as you white point.

One variable can the the screen brightness, as this will vary as to the environment. If you are producing hard copy such as prints then it's essential that the screen brightness represents the final viewing conditions. Too bright a screen is usually the main problem.



My iMac is calibrated to D65 and gamma 2.2, but my brightness level is set to 120 cd/m2. This may be too bright for some people, but prints and magazine work matches the screen.
 
Chappers said:
Provided you use a decent calibration device you should be able to match multiple monitors. The maximum I've ever had to match is 8 and it wasn't that difficult provided you are using the same aim points, such as Gamma 2.3 and D65 as you white point.

One variable can the the screen brightness, as this will vary as to the environment. If you are producing hard copy such as prints then it's essential that the screen brightness represents the final viewing conditions. Too bright a screen is usually the main problem.



My iMac is calibrated to D65 and gamma 2.2, but my brightness level is set to 120 cd/m2. This may be too bright for some people, but prints and magazine work matches the screen.
Click to expand...
OK John but with the link provided to Keith's thread, can you see the tattoos on photo #2.

Keith says that he wants the tatts in shadow and not visible but some viewers are able to see them but have a calibrated monitor. Keith also has a calibrated monitor.

Which are right or is that none are right? Surely the inconsistency is a headache for viewing critique?
 
My ageing NEC is calibrated to D65, gamma 2.2 and brightness 120 cd/m2 - I can make out the tatts, though they are not obvious.

As for which are right, I think that depends on intended purpose - I wanted to get monitor and prints somewhere close, the above does that for me. When I was researching monitor settings it seemed as though the ones I am using were the most common - it doesn't mean they are right though.
 
My monitor has been calibrated with a Colormunki display calibrator and I cannot see the tats
 
When I originally saw the post I can see the tats.
Yet I am looking at the post on a relatively old Acer laptop, which has never been calibrated, other than by, setting it up so it "looks right".
Interestingly, when I prepare something for printing, using the printers own profiles which I downloaded, the prints come back exactly as I see them in Lightroom.
Maybe I'm lucky? Would like to compare it to a "properly calibrated" monitor though, to see what difference it would show.
 
Calibration is dependant on where the monitor is placed and what - if any ambient light there is.

There will always be differences throughout the day looking at the same image on the same monitor because of ambient light (be it daylight or artificial)

You have to calibrate in conditions that are most likely the same ones you use for the bulk of your screen time. Also when complete you must check that it actually does match up against your lab prints (that you have had done WITHOUT any correction their end)
 
Tom I looked at the photos at lunchtime on my works' (uncalibrated Dell screen) and I didn't notice the tattoos until I read Philip's post and then relooked and I could see the 'skull' on her neck and something on her back but it was more of an outline.

I've just checked on my calibrated screen at home and no tattoos !!
 
Gamma should be 2.2

There are no standards for white point, but D65 (6500K) seems to be the most popular choice across the industry.

Luminance can vary depending on your working conditions, but assuming relatively subdued lighting, 120cd/m2 seems to be a default setting for most calibration software.

Assuming all monitors were calibrated to those settings, then we'd all see the same things, yes. It does presume the equipment is being used correctly by everyone though, and that everyone's screens are in good condition and capable of profiling well.

In that thread, I can see the tattoos. The neck tattoo is clearly visible, and the back tattoo is just visible, but is very dark.

That's on a Eizo ColorEdge GC303W with only 200 hours use from new, and hardware profiled using Eizo Color Navigator and a X-Rite i1-Display3 colorimeter. Calibrated to gamma 2.2, D65 and 120cd/m2.

tiler65 said:
Maybe nobody cares if it is right or wrong.
Click to expand...

Maybe not... but it's nice to know that it's "right" and as you want it before you send it anywhere. You can't control how it will be viewed on other screens, no, and you shouldn't lose any sleep over that as that's their fault, not yours, but I'd hate to be creating stuff on equipment that's not calibrated.
 
Last edited:
Pookeyhead said:
Gamma should be 2.2

There are no standards for white point, but D65 (6500K) seems to be the most popular choice across the industry.

Luminance can vary depending on your working conditions, but assuming relatively subdued lighting, 120cd/m2 seems to be a default setting for most calibration software.

Assuming all monitors were calibrated to those settings, then we'd all see the same things, yes. It does presume the equipment is being used correctly by everyone though, and that everyone's screens are in good condition and capable of profiling well.

In that thread, I can see the tattoos. The neck tattoo is clearly visible, and the back tattoo is visible, but is very dark.

That's on a Eizo ColorEdge GC303W with only 200 hours use from new, and hardware profiled using Eizo Color Navigator and a X-Rite X1-Display3 colorimeter. Calibrated to gamma 2.2, D65 and 120cd/m2.
Click to expand...

So, which is correct?

The shooter shot it as to have the tats in shadow, various calibrated devices show it different, a non calibrated device like a tablet/phone doesn't show the tats.

If there were to be screenshots of various histograms, of those that can see the tats and not, would the histograms show a difference?
 
tiler65 said:
So, which is correct?

The shooter shot it as to have the tats in shadow, various calibrated devices show it different, a non calibrated device like a tablet/phone doesn't show the tats.

If there were to be screenshots of various histograms, of those that can see the tats and not, would the histograms show a difference?
Click to expand...

No.. the histograms would be the same.... it's the devices' displays/profiles/colorspaces that are varying.
 
Last edited:
Pookeyhead said:
No.. the histograms would be the same.... it's the devices' displays/profiles/colorspaces that are varying.
Click to expand...

So would an AdobeRGB colourspace histogram display different than an sRGB one? (I only have one device with PS on, so I can't compare much)
 
That would make no difference, as the monitor, and or display profile you are using is interpreting the image at the GPU level just for the display.. it's not making any changes to the image itself.

You shouldn't be using Adobe RGB as a display profile though!

If you are referring to the profile embedded in the image, then yes, that will change the histogram a little.. depending on what profile you use.
 
Last edited:
Pookeyhead said:
That would make no difference, as the monitor, and or display profile you are using is interpreting the image at the GPU level just for the display.. it's not making any changes to the image itself.

You shouldn't be using Adobe RGB as a display profile though!

If you are referring to the profile embedded in the image, then yes, that will change the histogram.
Click to expand...

OK, so I understand that lighting situations will vary for what your calibration device is reading, how then, is a device like a tablet calibrated at the factory - surely there must be a 'standard' for them to work to?

The reason for my questions are that when a photo is posted for critique, the varying levels of, folk saying, to bright/dark etc must be off putting for the photographer.
 
I'm sure it is very off putting. If someone comments that they feel an image of mine is too dark I usually ask if they are using a profiled monitor or not. If not, I take their comment with a pinch of salt. Even if they are, they may just "feel" the image is too dark even if it's displayed correctly at their end. If my histogram looks healthy and I'm happy with it, I pay little attention if I like the way it looks. I'm fairly confident that what I have here is "correct" technically. That's what calibration is for... to give confidence that what you are seeing is correct to an established standard. Some cheaper calibrators are fussy though, and some software allows too much user intervention and allows for variance. Someone can well have calibrated their screen, but to a different standard, or maybe just not calibrated it properly (user error) or their screen may just be crap :)

Some people's images are clearly under, or over exposed though, as the histogram reflects this.. that's a different matter.


Most tablets are probably not calibrated at all, not to any standard anyway... maybe just for gamma... most monitors aren't either. Some Dell's are from teh factory, but it's pointless, as calibration drifts as the display ages. The only way to ensure accuracy is to calibrate regularly.



[edit]

Don't get too worked up about it is my advice. Even those on here that are seeing different things, it's only because the back tattoos are right on the edge of being visible, so any small variances will make the difference between can see it, and can't see it. For most images, most people will see a fairly accurate rendition unless their screen is miles off.
 
Last edited:
You must log in or register to reply here.
Back
Top