Pages: 1 [2] 3   Go Down

Author Topic: Profiling Monitor  (Read 27892 times)

Ethan_Hansen

  • Full Member
  • ***
  • Offline Offline
  • Posts: 114
    • Dry Creek Photo
Profiling Monitor
« Reply #20 on: May 20, 2010, 02:20:59 am »

Any monitor and calibration package that supports DDC (automated control over the monitor settings via software) allows you to hide either the monitor or yourself under a blanket.

There are three primary ISO specs that cover print viewing and display luminance. ISO 3664:2009 specifies the conditions for evaluating a print or an image on-screen, but not necessarily both at once. ISO 12646 lists requirements for comparing soft and hard copies, while ISO TC130 (included in ISO 12646) describes soft-proofing procedures. The quick summary along with some editorial comments:

Print viewing (ISO 3664):
  • Critical comparison (P1 appraisal level): 2000 lux. This is definitely not for soft-proofing, as the illumination corresponds to a monitor luminance level of 570cd/m2. The reason the light is so bright is to maximize shadow resolution. Recent research shows that illuminance levels above 600 lux reduce the ability to perceive color differences in saturated colors except for yellow, so this number may well change in the future.
  • Practical comparison (P2 appraisal level): 500 lux. This viewing condition corresponds to a monitor white luminance of 160 cd/m2, so there is some hope of accurate comparisons. Shadow details may appear compressed vs. the P1 level, but color variances are at least as visible.
  • S Appraisal level (soft proofing only on a monitor): At least 80 cd/m2 with a D65 whitepoint and ambient light at or below 32 lux. The assumption is that one is editing for the web.

Soft proofing (ISO 12646 with TC130 procedures):
  • Soft-proof only: D50 display whitepoint, minimum 80 cd/m2 luminance, maximum ambient light of 20 lux.
  • Soft-proof with comparison to hard copy in a viewing booth: Hard copy illuminanted at P2 appraisal level, display at least 100 cd/m2, with 160 recommended, display whitepoint matches viewing light color temperature, maximum ambient light of 20 lux.
  • Soft proof with the display integrated into the viewing booth: As above, but max of 64 lux around display.

What does this heap o' numbers tell us? First. the ambient light level is seriously low. 20 lux approximates a fireworks display. 8 seconds at f/8 for 100 ISO. There are multiple reasons for this cave-like value. The first is that the display whitepoint is set to D50. For most people, this reads visually with a yellowish tint unless the display luminance is very high and the ambient lighting low. The second is that the only consideration is maximum correlation between screen and print. Darker ambients work better. The specs also still hark back to the CRT luminance levels of 80cd/m2. Finally, these specs are geared toward typical commercial printing on groundwood stocks, which have a more yellow base than most artsy inkjet or photo lab papers. Moving the display whitepoint to the D60-D65 range helps here.

In a more realistic work environment, one where you need to see keyboard, mouse, and tablet, increasing the ambient light will not make much of a difference in the quality of comparison. Matching the viewing booth illumination level to the display is, however, a good idea. Higher quality booths support the 500 lux/P2 illumination level, or a display luminance of 160 cd/m2. Depending on how dark a room you wish to work in, and whether your viewer has a dimmer, lowering the display output is certainly possible.

Pat Herold

  • Full Member
  • ***
  • Offline Offline
  • Posts: 161
Profiling Monitor
« Reply #21 on: May 20, 2010, 03:59:04 pm »

Also, the specific purpose for the dark room or cloth while calibrating is to get an accurate black reading.  If you think about it, this makes sense.  When the black patches come onto your screen during the calibration, your puck is basically asked to report what color / brightness it is seeing.  If it's black then there is hardly anything there at all for it to measure.  So any stray light that sneaks around the foam padding on the face of the instrument is going to enter into the measurement of those black patches.  This will basically increase the noise in what the instrument is reading.  Black won't be the blackest your monitor can produce, but will be a little bit lighter.  Therefore the profile that is made  - and through which all the colors running through that profile to your display will be adjusted - reflect that grayish black.  Even if you don't sit in a cave when you do your normal viewing, you don't want to generally cripple your monitor to a higher black point (if you can help it!)
Logged
-Patrick Herold
  Tech Support,  chromix.com

John R Smith

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1357
  • Still crazy, after all these years
Profiling Monitor
« Reply #22 on: May 24, 2010, 03:45:11 am »

Well, I popped into our local Staples at the weekend and got a pack of black A3 foamboard. Spent a fun evening last night measuring and cutting, lashed up the Smith Mk I monitor hood and blu-tacked it into position. It seems to work really well, actually, the cheap LCD screen that I have looks much more posh. Cost was ten quid and I have two sheets left over.

John
Logged
Hasselblad 500 C/M, SWC and CFV-39 DB
an

Ethan_Hansen

  • Full Member
  • ***
  • Offline Offline
  • Posts: 114
    • Dry Creek Photo
Profiling Monitor
« Reply #23 on: May 24, 2010, 02:15:50 pm »

-> Cost was ten quid and I have two sheets left over.

Now you're all set to get a second monitor!

BobD

  • Newbie
  • *
  • Offline Offline
  • Posts: 42
Profiling Monitor
« Reply #24 on: May 27, 2010, 06:14:50 pm »

Quote from: digitaldog
Doesn’t have to be a dark room however, darker is better because any ambient light that strikes the display affects its black appearance. Ambient light can be too bright, it can’t be too low.
It has been my experience that...   If the brightness of your light is…
- right for your monitor, then it is too dark to evaluate your prints
- right for evaluating your prints, then it is too bright for your monitor.


Andrew (Digital Dog) do you want to weigh in on these premises?

Bob D
Logged

digitaldog

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 20630
  • Andrew Rodney
    • http://www.digitaldog.net/
Profiling Monitor
« Reply #25 on: May 27, 2010, 06:49:32 pm »

Quote from: BobD
It has been my experience that...   If the brightness of your light is…
- right for your monitor, then it is too dark to evaluate your prints
- right for evaluating your prints, then it is too bright for your monitor.

I don’t understand. The light I use for my prints is either a GTI booth I can dim to match the luminance I set on the display or a Solux task lamp I can move closer or farther away from a print to result in a match. There are no other lights used when viewing print to display matching.
Logged
http://www.digitaldog.net/
Author "Color Management for Photographers".

BobD

  • Newbie
  • *
  • Offline Offline
  • Posts: 42
Profiling Monitor
« Reply #26 on: May 28, 2010, 04:49:11 pm »

Quote from: digitaldog
I don’t understand. The light I use for my prints is either a GTI booth I can dim to match the luminance I set on the display or a Solux task lamp I can move closer or farther away from a print to result in a match. There are no other lights used when viewing print to display matching.
Andrew,
I find most people try make monitor adjustments and print evaluation in the same room under the same light brightness.

What I am saying is: if the brightness of your light is properly “dim” for monitor evaluation then it is too dark to evaluate your prints…  AND conversely, if the light brightness is sufficiently bright to evaluate prints then it is too bright for monitor evaluation. (I think this agrees with your prior post: “… (a darker room light) is better because any ambient light that strikes the display affects its black appearance.”)

Changing Brightness of the Print Viewing Light to Match… but match what?
You say “The light I use for my prints is either a GTI booth I can dim to match the luminance I set on the display or a Solux task lamp”.
Are you saying the you change your “print evaluation light” to match the brightness of your display… and not the brightness of the light where the print will be hung?

I also use the Solux task lamp to evaluate prints and move it closer or farther away to change the brightness (measuring lux with the ColorMunki). However, I try to match the brightness of the light in which the prints are to be hung.  If I don’t know the brightness of the “hanging light” then I try to attain a brightness of 250–300 lux (brighter than the average room but a standard brightness that most museum seem to use.)

Are you saying that you try matching the print viewing light to the brightness of the monitor?  Am I right in thinking that there is no conversion from cd/m2 to lux because one is the light source measurement and the other is a reflected measurement?

Solux Bulb?
Which Solux bulb do you use in your task lamp:
... 50 watt / 3500K / 36 degree? or
... 50 watt / 4700K  / 36 degree? or
... or?

Thanks in advance,
Bob DiNatale
Adobe Certified Expert Photoshop Lightroom
Xrite Coloratti
Logged

digitaldog

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 20630
  • Andrew Rodney
    • http://www.digitaldog.net/
Profiling Monitor
« Reply #27 on: May 28, 2010, 04:55:13 pm »

Quote from: BobD
Are you saying that you try matching the print viewing light to the brightness of the monitor?  Am I right in thinking that there is no conversion from cd/m2 to lux because one is the light source measurement and the other is a reflected measurement?

There is no direct conversion I’m aware if and if there were, it probably wouldn’t work. Just like a D50 illuminant and a D50 calibrated display rarely match. All the target values for white point will differ from user to user. I suspect that is one reason one can place an x/y chromaticity value into the better software products, rather than having to force a value like D50, or 6500 etc.

I adjust the display to the lower limits it can comfortably hit as lower means it will last longer. In the case of my NEC’s, that’s about 150cd/m2. New, out of the box, its hard to get them much lower. Then I adjust the viewing booth, which in the case of the GTI has a digital dimmer. The value that produced a visual match is 50% but that value is meaningless. The number means nothing. The match is what counts. Same with the Solux. I have to move it closer or farther away to result in a match. The distance value is a meaningless metric. What is meaningful is a match:

Logged
http://www.digitaldog.net/
Author "Color Management for Photographers".

BobD

  • Newbie
  • *
  • Offline Offline
  • Posts: 42
Profiling Monitor
« Reply #28 on: May 28, 2010, 05:12:03 pm »

Quote from: digitaldog
I adjust the display to the lower limits it can comfortably hit...
Andrew,

My modest NEC P221w display can be set to 40 cd/m2 using the SpectraView II software and my ColorMunki... however, I find 100cd/m2 is giving me a good match. Do you choose 150 cd/m2 or is that as low as your NEC's will go?

Also, what Solux bulb do you use for print evaluation, 3500k or 4700 k?

Bob DiNatale
Adobe Certified Expert Photoshop Lightroom
Xrite Coloratti
Logged

digitaldog

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 20630
  • Andrew Rodney
    • http://www.digitaldog.net/
Profiling Monitor
« Reply #29 on: May 28, 2010, 05:48:58 pm »

Quote from: BobD
My modest NEC P221w display can be set to 40 cd/m2 using the SpectraView II software and my ColorMunki... however, I find 100cd/m2 is giving me a good match. Do you choose 150 cd/m2 or is that as low as your NEC's will go?

Natively its about around there.

I’m using 3500K Solux.
Logged
http://www.digitaldog.net/
Author "Color Management for Photographers".

Mark Paulson

  • Jr. Member
  • **
  • Offline Offline
  • Posts: 89
Profiling Monitor
« Reply #30 on: May 30, 2010, 08:54:27 am »

Quote from: digitaldog
Natively its about around there.

I’m using 3500K Solux.

This has always confused me. If you are shooting for D50 why use lights that are 3500K? It seems to me you would use 5000K Solux, but I know that hardly anybody does. I just ordered some of the 120v Solux 3500K for my home office. I also noticed the task light on the GTI. I though the purpose of the GTI was for viewing and matching. Does the Solux not corrupt that display or do you turn the GTI off when you have the task light on?
« Last Edit: May 30, 2010, 08:57:37 am by MarkPaulson »
Logged

digitaldog

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 20630
  • Andrew Rodney
    • http://www.digitaldog.net/
Profiling Monitor
« Reply #31 on: May 30, 2010, 01:53:17 pm »

Quote from: MarkPaulson
This has always confused me. If you are shooting for D50 why use lights that are 3500K? It seems to me you would use 5000K Solux, but I know that hardly anybody does. I just ordered some of the 120v Solux 3500K for my home office. I also noticed the task light on the GTI. I though the purpose of the GTI was for viewing and matching. Does the Solux not corrupt that display or do you turn the GTI off when you have the task light on?

Because the numbers are often meaningless, we’re dealing with correlated color temp values, because 3500K looks better and matches better.

Keep in mind the important bit about any kelvin value is its correlated to a theoretical object called a black body radiator.
Logged
http://www.digitaldog.net/
Author "Color Management for Photographers".

John R Smith

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1357
  • Still crazy, after all these years
Profiling Monitor
« Reply #32 on: May 30, 2010, 03:10:31 pm »

Andrew

What the hell is "a theoretical object called a black body radiator"? It sounds really good.

John
Logged
Hasselblad 500 C/M, SWC and CFV-39 DB
an

digitaldog

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 20630
  • Andrew Rodney
    • http://www.digitaldog.net/
Profiling Monitor
« Reply #33 on: May 30, 2010, 03:22:16 pm »

Quote from: John R Smith
What the hell is "a theoretical object called a black body radiator"? It sounds really good.

See: http://www.ppmag.com/reviews/200512_rodneycm.pdf
Logged
http://www.digitaldog.net/
Author "Color Management for Photographers".

John R Smith

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1357
  • Still crazy, after all these years
Profiling Monitor
« Reply #34 on: May 30, 2010, 05:13:48 pm »

Thank you, Rodney. As always, the truth was rather less exciting han I had imagined it might be (possibly a sort of black scaly thing with tusks and fins, radiating malice) but I am now better informed than I was.

John
Logged
Hasselblad 500 C/M, SWC and CFV-39 DB
an

Eric Myrvaagnes

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 22814
  • http://myrvaagnes.com
    • http://myrvaagnes.com
Profiling Monitor
« Reply #35 on: May 30, 2010, 07:25:22 pm »

Quote from: John R Smith
Thank you, Rodney. As always, the truth was rather less exciting han I had imagined it might be (possibly a sort of black scaly thing with tusks and fins, radiating malice) but I am now better informed than I was.

John

I think you weren't far off with the "radiating malice" part. I often think I hear a black body radiator giving an evil cackle of glee whenever somebody tries to explain in simple terms the connection between, say, D65 and 6500K. As for what you need in practice, I find it suffices to try to do whatever Digitaldog suggests, without going too deeply into the physics of it.

My own understanding of Color Management is similar to the father of a friend many years ago explaining how an automobile works: "You pour gasoline in the back and water in the front and they mix in the middle and make the car go."


Eric

Logged
-Eric Myrvaagnes (visit my website: http://myrvaagnes.com)

digitaldog

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 20630
  • Andrew Rodney
    • http://www.digitaldog.net/
Profiling Monitor
« Reply #36 on: May 30, 2010, 08:00:56 pm »

The bottom line is, don’t put pure faith into the numbers per say. What correlates to a visual match may numerically be way off. Doesn’t matter. A visual match is what counts. Color management is far from a perfect numerical science.
Logged
http://www.digitaldog.net/
Author "Color Management for Photographers".

Mark Paulson

  • Jr. Member
  • **
  • Offline Offline
  • Posts: 89
Profiling Monitor
« Reply #37 on: June 01, 2010, 08:32:57 am »

Quote from: digitaldog
See: http://www.ppmag.com/reviews/200512_rodneycm.pdf

Very informative. Thanks for the link. I have show my ignorance and ask how the what the number on the outside of the gamut plot relate to the X-Y scale and what does the X-Y scale represent? TIA.
Logged

digitaldog

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 20630
  • Andrew Rodney
    • http://www.digitaldog.net/
Profiling Monitor
« Reply #38 on: June 01, 2010, 09:08:20 am »

Quote from: MarkPaulson
I have show my ignorance and ask how the what the number on the outside of the gamut plot relate to the X-Y scale and what does the X-Y scale represent? TIA.

Visible colors based on the “standard observer”,  theoretical human vision. X-Y is just a means to plot 2 dimensionally, some color within human vision.
Logged
http://www.digitaldog.net/
Author "Color Management for Photographers".

Mark Paulson

  • Jr. Member
  • **
  • Offline Offline
  • Posts: 89
Profiling Monitor
« Reply #39 on: June 01, 2010, 12:19:28 pm »

Quote from: digitaldog
Visible colors based on the “standard observer”,  theoretical human vision. X-Y is just a means to plot 2 dimensionally, some color within human vision.

OK I hate to be a PITA, but I am a land surveyor and very anal retentive when it comes to numbers. You chart was very informative on the different tints of the the same color temp, but I cannot see the correlation to the numbers on the x-y axis nor the numbers on the outside. I trust your chart and I am not arguing with you results, I'm just trying to understand the chart.
Logged
Pages: 1 [2] 3   Go Up