• Please use real names.

    Greetings to all who have registered to OPF and those guests taking a look around. Please use real names. Registrations with fictitious names will not be processed. REAL NAMES ONLY will be processed

    Firstname Lastname

    Register

    We are a courteous and supportive community. No need to hide behind an alia. If you have a genuine need for privacy/secrecy then let me know!
  • Welcome to the new site. Here's a thread about the update where you can post your feedback, ask questions or spot those nasty bugs!

The importance of Colour temperature, Gamma and luminance.

T

Tim Davey

New member
Hi All,

I am setting up my first ever truely colour calibrated workflow using a Spectraview 1990, Spectraview software (version 4.1.3) EyeOne Display 2 puck and was wondering what settings I should use for temperature, Gamma, white luminance and black luminance. Nec Support have said I should go with the default settings D50, L*, Max White luminance and Min Neutral Black luminance.

But I've been reading lots of threads that suggest a variety of settings so can some one give me some pointers for profiling using the above setup. I'm also an amateur enthusiast so am working out of my bedroom (with curtains shut or should they be open?) and my PC is running VISTA (I can hear the groans!! but I have not had any conflict issues).

When playing around with the settings I have created several test profiles (varying the 4 factors above) and all the profiles created seem to pass the visual test at http://www.simpelfilter.de/en/farbmanagement/monitorcalibration.html.

So finally can any one tell me how you know you have got the profilling right.

Thanks Tim..
 
T

Tim Davey

New member
Thanks for the reply, Do you have any thoughts on what I should set the black luminance to as in the spectraview profiler I can set this value (or set the contrast value).

Thanks Tim..
 
Andrew Rodney

Andrew Rodney

New member
Native Gamma and White Point, luminance is based on viewing conditions (ambient light around display, ambient light of item you are proofing). There's no standard for luminance. And there's nothing to really adjust on an LCD but the intensity of the backlight so you might as well leave this all alone, stick with native settings (record, don't 'adjust' since there's nothing to really adjust on an LCD).
 
Asher Kelman

Asher Kelman

OPF Owner/Editor-in-Chief
Hi Andrew,

Thanks for your post. People need reminding about the fixed state of the LCD. By "record" you mean, record the profile so that a picture's colors can be perceptually remapped to the color space of the
LCD monitor.

Asher
 
T

Tim Davey

New member
What are everyones feelings on lighting a room for correct viewing, I am a bit in the dark (pun intended!!) on this. Should I setup a room thats dark with curtains shut or is it better to have light from a window (sideways on to monitor for me) or is it better to use one of those daylight bulbs. I'm starting to think/realise that the answer might be one of personal preference but it would be great to hear everyones take on this.
 
Andrew Rodney

Andrew Rodney

New member
Tim Davey said:
On reading Andrews post again could you clarify what you mean by native settings.

Thanks Tim..
Click to expand...

The actual conditions of the device. Better software products allow you to set Native White Point. Instead of trying to adjust the display through a LUT to some arbitrary value like 6500K, it simply records (characterizes) the white point the display is actually producing. Same with Tone Response Curve (gamma).
 
T

Tim Davey

New member
Hope I'm not being a dim wit but is it good to select Native Gamma for profiling an LCD rather than say D50 or D65? my Spectraview user guide states the following, which makes me unsure .

Monitor’s native
Each monitor has it’s own, typical white point, that is displayed
when all 3 channels, R, G and B (Gains) are set to max. With this setting the luminance and gamut of the monitor are at their maximum.

"Monitor’s native“ is not recommended for color critical work in an ICC workflow. This setting may be useful for laptop monitors if their luminance is inadequate after calibrating
them to D50.

Also as I'm getting my head round this it appears that my monitor can use L* for a tonal response curve.
 
T

Tim Davey

New member
Hope I'm not being a dim wit but is it good to select Native Gamma for profiling an LCD rather than say D50 or D65? my Spectraview user guide states the following, which makes me unsure .

Monitor’s native
Each monitor has it’s own, typical white point, that is displayed
when all 3 channels, R, G and B (Gains) are set to max. With this setting the luminance and gamut of the monitor are at their maximum.

"Monitor’s native“ is not recommended for color critical work in an ICC workflow. This setting may be useful for laptop monitors if their luminance is inadequate after calibrating
them to D50.

Also as I'm getting my head round this it appears that my monitor can use L* for a tonal response curve.
 
Andrew Rodney

Andrew Rodney

New member
Tim Davey said:
Hope I'm not being a dim wit but is it good to select Native Gamma for profiling an LCD rather than say D50 or D65? my Spectraview user guide states the following, which makes me unsure .
Click to expand...

You're mixing up gamma (Tone Response Curve) for the color of white (D50). Two different target aim points. And again, you can't adjust either on an LCD using CCFL (Fluorescent) backlighting. They are what they are.
 
T

Tim Armes

New member
Andrew Rodney said:
You're mixing up gamma (Tone Response Curve) for the color of white (D50). Two different target aim points. And again, you can't adjust either on an LCD using CCFL (Fluorescent) backlighting. They are what they are.
Click to expand...

Hi Andrew,

Could you please expand on this a little. There's the colour of white with the RGB values at their maximum, which is what it is.

But surely profiling with an aim point of D65 will have an effect. It will ensure that the RGB values chosen for white by the CMS are such that you'll get close to D65, as opposed to the native white point with all the RGB values at their maximum.

Is there a reason that you wouldn't want to do this?

TIm
 
Andrew Rodney

Andrew Rodney

New member
There's the TRC of the display and there's the color of the whitest white, they are totally different specifications for calibration of a display.

They don't have to match the same names used to define the white point of a working space (or TRC). Just as your working space and output space don't have to match the same values.

As for the white point of a CCFL LCD, it is what it is, you can't change it without applying alterations to the LUT in the graphic card which is 8-bits. Better to just measure it (native white point) and specify that in the display for the Display Using Monitor Compensation that happens in ICC aware applications. That's happening at more than 8-bits per color, at least in all Adobe applications.
 
B

Bart_van_der_Wolf

pro member
Tim Davey said:
Hope I'm not being a dim wit but is it good to select Native Gamma for profiling an LCD rather than say D50 or D65?
Click to expand...

That should probably be "Native Gamut". That is basically the brightest white the LCD can produce, and often corresponds to something like 9500 Kelvin color temperature. Profiling that to a lower Kelvin, e.g. 6500 K, means that the result will have a lower emission and a narrower gamut than at 9500 K because the RGB channels (or their lookup tables, LUTs) will be tuned down to achieve the new white point.

my Spectraview user guide states the following, which makes me unsure .

Monitor’s native
Each monitor has it’s own, typical white point, that is displayed
when all 3 channels, R, G and B (Gains) are set to max. With this setting the luminance and gamut of the monitor are at their maximum.
Click to expand...

They speak about "gains" but that is just relevant to CRT types of monitors, where the individual channels are amplified to change output luminance. The maximum emision from most LCDs on the other hand is determined by its unobstructed backlight, and can only be changed by partially obstructing/filtering/reducing the output.

"Monitor’s native“ is not recommended for color critical work in an ICC workflow. This setting may be useful for laptop monitors if their luminance is inadequate after calibrating
them to D50.
Click to expand...

This is basically an observation that the reduced LCD channel output will lead to a lower emission, maybe too low for some environments.

Also as I'm getting my head round this it appears that my monitor can use L* for a tonal response curve.
Click to expand...

Could you explain which monitor type/model you are referring to, that might clear up some of the confusion?

Bart
 
T

Tim Armes

New member
Andrew Rodney said:
As for the white point of a CCFL LCD, it is what it is, you can't change it without applying alterations to the LUT in the graphic card which is 8-bits. Better to just measure it (native white point) and specify that in the display for the Display Using Monitor Compensation that happens in ICC aware applications. That's happening at more than 8-bits per color, at least in all Adobe applications.
Click to expand...

Ah, of course.

So the LCD is profiled and the graphics application is responsible for displaying the "correct" white (the white defined by the image's colour space) by using the LCD profile to do the translation.

With a CRT we can calibrate the white which is even better.

Crystal clear.

Thanks,

Tim
 
Andrew Rodney

Andrew Rodney

New member
Bart_van_der_Wolf said:
That should probably be "Native Gamut". That is basically the brightest white the LCD can produce, and often corresponds to something like 9500 Kelvin color temperature.
Click to expand...

Nope, its a tone response curve that is often (incorrectly) referred to as gamma (which on a CRT is correct):

http://www.ppmag.com/reviews/200612_rodneycm.pdf

Brightness of white is defined using Candelas per metere squared.

Color of white is defined by either correlated color temperature (not so great) or using a standard illuminant (D50, D65), an exact color of white.

Gamma or TRC is the tone curve that represents input versus output of a device as described in the above PDF.
 
Cem_Usakligil

Cem_Usakligil

Well-known member
Andrew Rodney said:
...As for the white point of a CCFL LCD, it is what it is, you can't change it without applying alterations to the LUT in the graphic card which is 8-bits. ...
Click to expand...
Hi Andrew,

Although I agree with you (after all, who am I to disagree with an expert like yourself <smile>) you've got me confused a bit with the above statement.

If I'm not mistaken, many high end LCD monitors have hardware controls for adjusting the the RGB values (the mine do). Does changing the RGB values not affect the "native" white point of the LCD then? In my case, I am able to partially calibrate my monitors using those controls to a near D65 value, prior to actually profiling them.

Please excuse me if this is a very stupid question :).

Cheers,
 
Andrew Rodney

Andrew Rodney

New member
Again, you can't physically alter the white point of the display with CCFL (you could with three color LED and of course CRTs).

These high end displays are still 8-bit in, 8-bit out. They use high bit internally which helps but its not a total nor satisfactory solution. We will not see high bit path from the entire video system until both the OS and applications are updated (something we'll probably see sooner under Mac OS than later, but we still need Adobe to update their applications to use this data path).

The controls you refer to are again, only altering the LUTs internally. They are not adjusting the color of the Fluorescent backlight itself (they can't). Changing the values doesn't provide the same functionality as leaving the LUTs alone and recording the actual native white point of the display and letting the ICC profile handle the adjustments from there.
 
Cem_Usakligil

Cem_Usakligil

Well-known member
Andrew Rodney said:
...The controls you refer to are again, only altering the LUTs internally. They are not adjusting the color of the Fluorescent backlight itself (they can't). Changing the values doesn't provide the same functionality as leaving the LUTs alone and recording the actual native white point of the display and letting the ICC profile handle the adjustments from there.
Click to expand...
OK, I'm not confused anymore. Thanks for your patience and the clear explanation :).

Regards,
 
B

Bart_van_der_Wolf

pro member
Andrew Rodney said:
Nope, its a tone response curve that is often (incorrectly) referred to as gamma (which on a CRT is correct):

http://www.ppmag.com/reviews/200612_rodneycm.pdf[
Click to expand...

You are right, but I didn't express myself as I intended. The initial misunderstanding was about "native gamma" and "native gamut", two different things altogether.

The second part of my comment was about the reduced gamut on LCDs, because the volume of the gamut varies with max. luminance level. The output Luminance level, as you said, is the result of the TRCs for R, G, and B.

Bart
 
Last edited:
T

Tim Davey

New member
Just got bak to my PC Ray, will need to sit down tonight to review all the posts over the last day. But its great to get so much discussion on this point its really helping my understanding. Will post thought later on.

Regards

Tim..
 
T

Tim Davey

New member
Well I have now sat through my first reading of the last 11 posts. I'm on a steep learing curve already and some of the terms and detail are still beyond my understanding at the moment. I am going to read the thread again in a minute and I think this is a thread I will be returnung to on a regular basis over the next few weeks.
 
You must log in or register to reply here.
Top