On display calibration instruments

Started Nov 18, 2009 | Discussions
Iliah Borg Forum Pro • Posts: 29,053
On display calibration instruments

http://lists.apple.com/archives/colorsync-users/2009/Nov/msg00175.html

http://lists.apple.com/archives/colorsync-users/2009/Nov/msg00204.html

For the purpose of printing, setting 0.45 to 0.5 cd/m^2 as a black point is quite correct.

-- hide signature --
lovEU Veteran Member • Posts: 3,176
Re: On display calibration instruments

Iliah Borg wrote:

http://lists.apple.com/archives/colorsync-users/2009/Nov/msg00175.html

http://lists.apple.com/archives/colorsync-users/2009/Nov/msg00204.html

For the purpose of printing, setting 0.45 to 0.5 cd/m^2 as a black point is quite correct.

Having read both the posts I'm not sure what you're hinting to -- maybe that such black point makes sense to avoid problems the devices can cause at lower black point values? If I didn't get it wrong, would that suggest to set 'black luminance" in a software like Display 4 manually to 0.45 cd/m2?
--
regards, eric

OP Iliah Borg Forum Pro • Posts: 29,053
Re: On display calibration instruments

Having read both the posts I'm not sure what you're hinting to

There is an urban legend that colorimeters are better suited to calibrate monitors. This legend is disproved yet again in the referenced posts.

Many people complain that their monitors calibrated with different colorimeters do not match. That is especially the case when Spyders and i1 Display devices are used.

Any device, be it a camera, a colorimeter, or a spectrophotometer, adds noise in shadows. Setting black point below 0.5 cd/m^2 is impractical in most cases, because not only deepest shadows are read wrong (noise); but dark tones too are getting distorted while smoothing tone curves in a profiling software. Error in readings from the deepest shadows propagates up to Zone III.

It is extremely important to calibrate in complete darkness, otherwise the light leak through the panel distorts the calibration.

It is extremely important not to press the device hard against an LCD screen, it must just touch it.

It is important to calibrate to the same gamma as the gamma in the working space - 1.8 for ProPhoto and like, 2.2 for Adobe RGB/sRGB, L* for ECI RGB v.2 and like. If most of the work is done in Lab, L8 is the preferred tone curve. Thus screen posterization caused by gamma adaptation is avoided.

It is important to verify calibration and profiling; and for verification it is better to shift the device from the point where the calibration took place. Results of verification should be in the green zone. Some yellow is acceptable, but no orange or red is OK.

Monitor must be warmed up before the calibration, that is left to stand for one hour with the desktop filled with middle grey (L about 50, or RGB about 118..121).

-- hide signature --
lovEU Veteran Member • Posts: 3,176
Re: On display calibration instruments

Iliah Borg wrote:

Any device, be it a camera, a colorimeter, or a spectrophotometer, adds noise in shadows. Setting black point below 0.5 cd/m^2 is impractical in most cases, because not only deepest shadows are read wrong (noise);

While I understand the wrong-reading problem …

but dark tones too are getting distorted while smoothing tone curves in a profiling software. Error in readings from the deepest shadows propagates up to Zone III.

I don’t understand how a device adds noise. How does it work?

It is extremely important to calibrate in complete darkness, otherwise the light leak through the panel distorts the calibration.

Would be interesting to get to know how many people are doing so -- not me, until now.

It is extremely important not to press the device hard against an LCD screen, it must just touch it.

I’ve done that (pressing) once in my first calibration attempt when the Huey came out -- the result still can be marveled on my old monitor

It is important to calibrate to the same gamma as the gamma in the working space - 1.8 for ProPhoto and like, 2.2 for Adobe RGB/sRGB, L* for ECI RGB v.2 and like. If most of the work is done in Lab, L8 is the preferred tone curve. Thus screen posterization caused by gamma adaptation is avoided.

Accidentally, I’ve done right using aRGB mainly and having gamma at 2.2.

OT: I’m not sure if it is a good idea to use aRGB when handling raw files saved as Tiff’s in Photoshop or LR. So, better change to ProPhoto and set monitor to 1.8 then?

It is important to verify calibration and profiling; and for verification it is better to shift the device from the point where the calibration took place. Results of verification should be in the green zone. Some yellow is acceptable, but no orange or red is OK.

Great advice as well, such easy but I had never thought about it.

Monitor must be warmed up before the calibration, that is left to stand for one hour with the desktop filled with middle grey (L about 50, or RGB about 118..121).

It's new to me to let a monitor warm up with middle grey, makes a lot of sense.

Thanks for sharing, Iliah. And btw, a pity I couldn't I read such advice from i1d2 "manual"... (but maybe my fault and I didn't find it at their website)
--
regards, eric

OP Iliah Borg Forum Pro • Posts: 29,053
Re: On display calibration instruments

dark tones too are getting distorted while smoothing tone curves in a profiling software. Error in readings from the deepest shadows propagates up to Zone III.

I don’t understand how a device adds noise. How does it work?

It is the same problem as with the sensors in cameras - readings from the shadows are unreliable. There are some other critically important problems that must be taken into account - fading of the filters in colorimeters and temperature drift of both colorimeters and spectrophotometers. I actually let the device sit on the monitor during that hour it takes a monitor to warm up.

It is extremely important to calibrate in complete darkness, otherwise the light leak through the panel distorts the calibration.

Would be interesting to get to know how many people are doing so

I'm afraid very very few.

OT: I’m not sure if it is a good idea to use aRGB when handling raw files saved as Tiff’s in Photoshop or LR.

Why not? I actually use ECI RGB v.2 or modified with L* BetaRGB most of the times.

So, better change to ProPhoto and set monitor to 1.8 then?

No monitor can display ProPhotoRGB even remotely close. If you are not editing by numbers, I see no real reason to use ProPhoto RGB. But if you are going to use it, it is better to have one of the monitor profiles set with gamma 1.8. Creating a group of 3 profiles for each 3 common gamma settings, 1.8, 2.2, and L* is a good way to address the problem. I prefer to use v.4 profiles.

Generally, it is easier to sell something that appears as a push-button solution, at least on the surface. That is why detailed calibration procedures are rare and not encouraged by manufactures.

-- hide signature --
lovEU Veteran Member • Posts: 3,176
Fading of the filters?

Iliah Borg wrote:

It is the same problem as with the sensors in cameras - readings from the shadows are unreliable. There are some other critically important problems that must be taken into account - fading of the filters in colorimeters and temperature drift of both colorimeters and spectrophotometers.

What is the reason for the fading? I guess the filters are made of plastics, so is it a problem with evaporating softeners? Is it caused by the intended use of colorimeters and maybe increased because filters get warm when placed on the surface of the monitor?

And, could the fading slowed down somehow?
--
regards, eric

Keyboard shortcuts:
FForum MMy threads