Pages: [1]   Go Down

Author Topic: lcd calibration and profile, luminance  (Read 10845 times)

sgwrx

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 310
lcd calibration and profile, luminance
« on: March 25, 2006, 02:04:53 pm »

i ended up getting a slightly used sony p234 lcd. i've also invested in an eye-one display2.

i find that a target luminance of 120 cd/m2 is quite bright, gamma 2.3.  according to the i1, my ambient light is 4400k and about 8lux, 12lux depending on where it's pointed.

is it wrong for me to have that low level of ambient light? would it be wrong to readjust to a target of 80 cd/m2 so my eyes aren't hurting?
Logged

61Dynamic

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1442
    • http://
lcd calibration and profile, luminance
« Reply #1 on: March 25, 2006, 02:14:37 pm »

120cd/m2 is the proper luminance for an LCD. I wouldn't go below that since both whites and blacks are adjusted by the display brightness (where-areas, a CRT had individual controls for each). 120 provides the best performance for both whites and blacks with current LCD tech.

ISO standard 12646 states ambient light can be as high as 36 lux. I'd recommend upping the brightness to near that so your eyes aren't strained.

Your gamma should be 2.2 and not 2.3.
Logged

digitaldog

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 20630
  • Andrew Rodney
    • http://www.digitaldog.net/
lcd calibration and profile, luminance
« Reply #2 on: March 25, 2006, 06:12:53 pm »

Quote
i ended up getting a slightly used sony p234 lcd. i've also invested in an eye-one display2.

i find that a target luminance of 120 cd/m2 is quite bright, gamma 2.3.  according to the i1, my ambient light is 4400k and about 8lux, 12lux depending on where it's pointed.

is it wrong for me to have that low level of ambient light? would it be wrong to readjust to a target of 80 cd/m2 so my eyes aren't hurting?
[a href=\"index.php?act=findpost&pid=60999\"][{POST_SNAPBACK}][/a]

The luminance target is basically tied to the ambient light in the environment. You want the darkest and brightest item you view to be on the display. Unfortunately there are no spec's for LCDs as the very old ones for CRTs (which basically recommended a lumanice of 90-95 cd/m2 in an environment around 16-25 lux. That's REAL dim! LCDs can easily go higher for long periods (so could CRTs but not for all that long).

Eye-One Match 2.6 (new as of last week) has a native gamma setting and I'd use that. They force you to also use a native white point to be able to use Native gamma but that's really ideal for LCDs anyway.

Now the question is, how bright is the light booth you're viewing your prints?
Logged
http://www.digitaldog.net/
Author "Color Management for Photographers".

sgwrx

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 310
lcd calibration and profile, luminance
« Reply #3 on: March 25, 2006, 07:14:56 pm »

i see, so the reason behind the 120 luminance is to help balance the dynamic range of black to white then?

for the light-booth, which i do not own yet, i was hoping to be able to get a lower luminance one to help match my monitor and surrounding environment.

doing that, however, it would seem that i'd be trying to create my own standard as opposed using _the_ standard.

thanks both.

off to recalibrate.  ps, i made a hood out of matt-black covered foamcore it's about 8.75" deep and supprisingly for $10 worth of materials and velcro it seems darn nice! it makes the luminance that much more intense though LOL.

on a side note relating back to the 120 cd/m2 level, i did that test in photoshop where you marquee a sqaure in the middle of a black (or white) field, hide everything and bring up the curves dialog to step up from 0 (black) or down from 255 (white). i noticed that before this bright calibration and profile i could not see a change in black until 13 and now i see a change in black at 5 (4 if i strain and flip back/forth). it holds at the same level for two or three steps over a few further adjustments but this is obviously telling me that i've got a much better black level.  as for white, i can see changes all the way from 254 to 230. with maybe two or three spots with the two-three step plateau issue.

curious, i did upgrade to the latest release of the match software and don't see a difference in the before/after calibration. but maybe that's because i'm coming from a few previous calibrations.
Logged

sgwrx

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 310
lcd calibration and profile, luminance
« Reply #4 on: March 25, 2006, 07:16:36 pm »

oh shoot, i talk to much and forgot to ask:

with regard to the native whitepoint and gamma, so in the software in the first step i select native white point and native gamma?  i have to make sure my monitor settings (in the menu) are gamma 2.2 and 6500k whitepoint?

thanks
-steve
Logged

digitaldog

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 20630
  • Andrew Rodney
    • http://www.digitaldog.net/
lcd calibration and profile, luminance
« Reply #5 on: March 25, 2006, 07:24:55 pm »

-->i see, so the reason behind the 120 luminance is to help balance the dynamic range of black to white then?

No,that's the brightness of the white. IF you have control over that and black level, you can control the dynamic range. In Match, that's not an option.

Using Advance Mode in Match, you can select Native White Point and then Native Gamma (in version 3.6). You can't select Native Gamma unless you first select Native White Point (something I hope they fix in the next release). Forget any monitor settings as there's nothing "physically" going on in an LCD other than the intensity of the backlight (the luminance which you can adjust and will be asked to adjust in Match in a later screen).
Logged
http://www.digitaldog.net/
Author "Color Management for Photographers".

sgwrx

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 310
lcd calibration and profile, luminance
« Reply #6 on: March 25, 2006, 08:33:39 pm »

Quote
No,that's the brightness of the white. IF you have control over that and black level, you can control the dynamic range. In Match, that's not an option.

Using Advance Mode in Match, you can select Native White Point and then Native Gamma (in version 3.6). You can't select Native Gamma unless you first select Native White Point (something I hope they fix in the next release). Forget any monitor settings as there's nothing "physically" going on in an LCD other than the intensity of the backlight (the luminance which you can adjust and will be asked to adjust in Match in a later screen).
[a href=\"index.php?act=findpost&pid=61010\"][{POST_SNAPBACK}][/a]

i see.

well, thanks again. i'll post a follow up results soon.
Logged

sgwrx

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 310
lcd calibration and profile, luminance
« Reply #7 on: March 25, 2006, 09:15:45 pm »

edit: image removed

so how's this look?
before starting i put brightness/contrast/backlight back to normal (default). i know you said it didn't matter but did anyway.

the ambient lighting was just about smack dab in the middle of the green bar range when measured. the higher ambient makes it less harsh on the eyes.

so this isn't really telling me what the gamma is huh? or do i know that b/c it's a windows XP machine - 2.2.

edit: i also set the general options in eye-one match for "According to ICC spec v4" and monitor profile size "large (LUT)".
« Last Edit: March 26, 2006, 11:26:04 am by sgwrx »
Logged

digitaldog

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 20630
  • Andrew Rodney
    • http://www.digitaldog.net/
lcd calibration and profile, luminance
« Reply #8 on: March 25, 2006, 09:34:24 pm »

Quote


so how's this look?
before starting i put brightness/contrast/backlight back to normal (default). i know you said it didn't matter but did anyway.

the ambient lighting was just about smack dab in the middle of the green bar range when measured. the higher ambient makes it less harsh on the eyes.

so this isn't really telling me what the gamma is huh? or do i know that b/c it's a windows XP machine - 2.2.
[a href=\"index.php?act=findpost&pid=61018\"][{POST_SNAPBACK}][/a]


The green indicator for Ambient light is only mildly useful. It's based on real old ISO spec's (CRT; 20th century) and if at least its in the middle, you're not way off (too bright). But in the future, it would be nice if the software could measure the ambient light and recommend a luminance level.
So don't put too much faith in the green bar correlating to the best ambient light around the display and having any basis on the best target luminance levels.

The software doesn't report the target gamma but that's not really important although it would be nice to know (this entire screen is pretty much eye candy anyway) What you've done is tell the software NOT to adjust the gamma to some arbitrary value like 2.2 but instead to leave the gamma value as is. Your display might be 2.2 or 2.1 or 2.3. Nearly all displays are in that neighborhood. The native gamma is set so that you don't make an adjustment to the graphic card but rather simply describe it inside the ICC profile where that profile will produce a preview in ICC aware applications based on naive gamma. The result is less banding (aliasing) because you haven't applied some adjustment on an 8-bit graphic card system. So say your native gamma is 2.1 but you select 2.2.the result of the adjustment is produced on the 8-bt graphic card. That buys you nothing. The profile is describing the actual (native) gamma. Outside of the Sony Artisan, this is the only product I know of that does this. IOW, if the native gamma of your display is 2.1, it's better to describe this in the profile than adjust in the graphic card to 2.2 just because 2.2 is about where most displays produce a Tone Response Curve.

The only downside to a native gamma is outside of ICC aware applications but who cares?

The only beef I have with Match is it asks you to calibrate to a correlated color temperature (6500K) which isn't an exact color of white (it's a range of colors). I'm trying to convince them to use D65 which is an exact color. When you use a correlated color temperature (Kelvin), you really don't know what color of white is being used. D65 is an exact color. 6500K isn't.

As for contrast ratio, it's easy to figure this out if you have both luminance value and black point value. As an example, with a white point luminance target of 100cd/m2 and a black point of .25 cd/m2, a ratio of 400:1 (100/.25) is created. Using the graph you've shown, the contrast ratio is 300:1 (120/.4)
Logged
http://www.digitaldog.net/
Author "Color Management for Photographers".

sgwrx

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 310
lcd calibration and profile, luminance
« Reply #9 on: March 25, 2006, 10:03:10 pm »

Quote
The green indicator for Ambient light is only mildly useful. It's based on real old ISO spec's (CRT; 20th century) and if at least its in the middle, you're not way off (too bright). But in the future, it would be nice if the software could measure the ambient light and recommend a luminance level.
So don't put too much faith in the green bar correlating to the best ambient light around the display and having any basis on the best target luminance levels.

The software doesn't report the target gamma but that's not really important although it would be nice to know (this entire screen is pretty much eye candy anyway) What you've done is tell the software NOT to adjust the gamma to some arbitrary value like 2.2 but instead to leave the gamma value as is. Your display might be 2.2 or 2.1 or 2.3. Nearly all displays are in that neighborhood. The native gamma is set so that you don't make an adjustment to the graphic card but rather simply describe it inside the ICC profile where that profile will produce a preview in ICC aware applications based on naive gamma. The result is less banding (aliasing) because you haven't applied some adjustment on an 8-bit graphic card system. So say your native gamma is 2.1 but you select 2.2.the result of the adjustment is produced on the 8-bt graphic card. That buys you nothing. The profile is describing the actual (native) gamma. Outside of the Sony Artisan, this is the only product I know of that does this. IOW, if the native gamma of your display is 2.1, it's better to describe this in the profile than adjust in the graphic card to 2.2 just because 2.2 is about where most displays produce a Tone Response Curve.

The only downside to a native gamma is outside of ICC aware applications but who cares?

thanks. sure and that's the point of profiling - describing something. somehow this makes more sense now than when i read it all in "color mgmt for photographers".  eye-candy is a good thing when you drop a couple hundred

Quote
The only beef I have with Match is it asks you to calibrate to a correlated color temperature (6500K) which isn't an exact color of white (it's a range of colors). I'm trying to convince them to use D65 which is an exact color. When you use a correlated color temperature (Kelvin), you really don't know what color of white is being used. D65 is an exact color. 6500K isn't.

As for contrast ratio, it's easy to figure this out if you have both luminance value and black point value. As an example, with a white point luminance target of 100cd/m2 and a black point of .25 cd/m2, a ratio of 400:1 (100/.25) is created. Using the graph you've shown, the contrast ratio is 300:1 (120/.4)
[a href=\"index.php?act=findpost&pid=61020\"][{POST_SNAPBACK}][/a]

to site the book again, AH! that's what it was talking about when it was describing contrast ratio. i had the numbers in front of me, but didn't correlate them to the formula in the book.  and yeah, after all the reading i've done about actual colors D50, D65 vs. the K numbers, it is a little irksome that no more is being done to standardize. though, maybe it's going to get more and more popular in OS's and such since digital photography and home printing is becoming increasingly more popular.
Logged

digitaldog

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 20630
  • Andrew Rodney
    • http://www.digitaldog.net/
lcd calibration and profile, luminance
« Reply #10 on: March 26, 2006, 09:55:51 am »

Its important to note that just because you ask the software for a D65 white point, that's not what you're getting (there's only one output device that can produce D65 and it's 93 million miles from your display). But at least we know what the software is trying to correlate to where with 6500K, we don't.
Logged
http://www.digitaldog.net/
Author "Color Management for Photographers".

32BT

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 3095
    • Pictures
lcd calibration and profile, luminance
« Reply #11 on: March 27, 2006, 07:13:53 am »

Why do you believe that 6500K is not a well defined color? I personally believe that 6500K is more precisely defined than D65 is. Have you ever tried to derive the D50 XYZ values recommended by the CIE for the profile illuminant?


Quote
Its important to note that just because you ask the software for a D65 white point, that's not what you're getting (there's only one output device that can produce D65 and it's 93 million miles from your display). But at least we know what the software is trying to correlate to where with 6500K, we don't.
Logged
Regards,
~ O ~
If you can stomach it: pictures

digitaldog

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 20630
  • Andrew Rodney
    • http://www.digitaldog.net/
lcd calibration and profile, luminance
« Reply #12 on: March 27, 2006, 08:40:46 am »

Quote
Why do you believe that 6500K is not a well defined color? I personally believe that 6500K is more precisely defined than D65 is.
[a href=\"index.php?act=findpost&pid=61117\"][{POST_SNAPBACK}][/a]

 It's somewhat dangerous to use color temperature to define what you want because the reality is if all light sources were true blackbodies a particular color temperature would produce the same color of light. Because natural materials are not theoretical blackbodies, heating them to a specific temperature creates deviates from the theoretical color from magenta to green. It's really much safer to use the term correlated color temperature (CCT) because many colors of white may correlate to the same blackbody color temperature. Different illuminants can have the same correlated color temperature.

 This is one reason why the CIE defined the Standard Illuminants.These illuminants are defined spectrally meaning a certain amount of energy at each wavelength across the spectrum. This is an exact and non ambiguous description of color. D65 is an exact color, it is not a range of colors. If you have a color meter that reports color temperature of a light source many light sources that appear different could read the same, that's kind of a problem!

a) D65 is a spectral power distribution (a certain amount of energy at each wavelength across the visible spectrum).
 D65 is a tristimulus value; the D65 spectrum, when viewed by the CIE standard observer, produces an XYZ triplet (or xyY if you prefer).
c) 6500K blackbody radiator is a spectral power distribution.
d) 6500K is blackbody tristimulus value; the 6500K blackbody spectrum, when viewed by the CIE standard observer, produces an XYZ triplet—similar to, but slightly different from, the one found in (.
e) Correlated color temperature takes a color's chromaticity coordinate (x,y) and finds the particular blackbody temperature whose chromaticity coordinate (d) is closest to it. Note that there are many different colors that have the same correlated color temperature. So a spectrum is very precise and unique. Its xyY is less precise and unique. Its CCT is even less precise and unique.
D65 is a unique SPD (there exists only one). A color whose CCT is 6500K is not unique (there are infinitely many different xyY and SPDs that share it).
Logged
http://www.digitaldog.net/
Author "Color Management for Photographers".

digitaldog

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 20630
  • Andrew Rodney
    • http://www.digitaldog.net/
lcd calibration and profile, luminance
« Reply #13 on: March 27, 2006, 09:21:35 am »

figure 1-6 shows the lines of correlated color temperature. The one that shows 6500K (for example); any color on the line running perpendicular to the blackbody curve could be 6500K.
Logged
http://www.digitaldog.net/
Author "Color Management for Photographers".

32BT

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 3095
    • Pictures
lcd calibration and profile, luminance
« Reply #14 on: March 27, 2006, 10:10:56 am »

Oh, okay, that explains the mix-up.

Your image shows what in literature are usually called the "isotemperature" lines. The points where these lines cross with the "Planckian locus" are then called a "correlated color temperature".

I presume you mean to say:
If you use a measurement device to measure the color of some reference white and the software converts this to a correlated color temperature, then the resulting white may deviate from the measured white.

Because:

CCTs are on the Planckian locus and most white lights are not,

In addition:

The Planckian locus deviates slightly from true daylight spectra, so if you want to emulate daylight, it is best to use the daylight spectra.


Quote
figure 1-6 shows the lines of correlated color temperature. The one that shows 6500K (for example); any color on the line running perpendicular to the blackbody curve could be 6500K.
Logged
Regards,
~ O ~
If you can stomach it: pictures

digitaldog

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 20630
  • Andrew Rodney
    • http://www.digitaldog.net/
lcd calibration and profile, luminance
« Reply #15 on: March 27, 2006, 10:18:31 am »

-->Your image shows what in literature are usually called the "isotemperature" lines. The points where these lines cross with the "Planckian locus" are then called a "correlated color temperature".


I think/hope I was clear that I was talking about correleated color temp and that when a software product asks you to select such a value, it's ambiguous whereby when asked for a standard illuminant, it isn't. I think/hope I also made it clear that in both cases, that's not what you're getting. Unless you put CCT in front of a value followed by Kelvin, you're not clearly defining the value as a range rather than a specific.
Logged
http://www.digitaldog.net/
Author "Color Management for Photographers".

32BT

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 3095
    • Pictures
lcd calibration and profile, luminance
« Reply #16 on: March 27, 2006, 11:17:14 am »

CCTs are not ambiguous. They are clearly defined and have a mathematical SPD.

When a software asks for a temperature, then it is quite clear what color will be calculated, and that color will be close to a daylight equivalent, usually close enough.

Only in the case where a software allows you to measure a temperature of some reference white and use that temperature as an input value, only then chances are you end up with a different white. You rightfully warned us for this case.

Using daylight equivalents on the other hand requires conversions from discrete SPDs which, for some odd reason only known to the gods at CIE, result in a D50 XYZ value different from the recommended profile illuminant.

Just look at the whitepoint defined in the ColorMatchRGB profile that comes with Adobe Photoshop for example...

Just giving you a hard time to keep you sharp. :-)

Quote
I think/hope I was clear that I was talking about correleated color temp and that when a software product asks you to select such a value, it's ambiguous
Logged
Regards,
~ O ~
If you can stomach it: pictures

digitaldog

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 20630
  • Andrew Rodney
    • http://www.digitaldog.net/
lcd calibration and profile, luminance
« Reply #17 on: March 27, 2006, 11:42:16 am »

-->CCTs are not ambiguous. They are clearly defined and have a mathematical SPD.

Yes, when given an actual SPD. But without, no. What is the SPD of 6500K? If you define it on the CIE chromaticity diagram, give an XY value then yes I agree with you. But just saying "6500K" doesn't. That's my point.

In Match 3.6 you can select "6500K" or you can actually entire an XY chromaticity value. The later I have no problem with (although I think its far too geeky for the intended audience). Or you can measure a light source to get those XY values. But are you suggesting that method and selecting "6500K" from the popup are the same?

-->When a software asks for a temperature, then it is quite clear what color will be calculated, and that color will be close to a daylight equivalent, usually close enough.

You know the old saying about close <g>

-->Just look at the whitepoint defined in the ColorMatchRGB profile that comes with Adobe Photoshop for example...

What about it?
Logged
http://www.digitaldog.net/
Author "Color Management for Photographers".
Pages: [1]   Go Up