[argyllcms] Dispcal on WinXPSP2+ i1pro

  • From: Roger Breton <graxx@xxxxxxxxxxxx>
  • To: argyllcms@xxxxxxxxxxxxx
  • Date: Sun, 28 Oct 2007 21:32:49 -0400

Hello Graeme,

I am not sure I successfully completed the calibration of my LCD display
with V0.70 dispcal since, towards patch 60 of 100, in the verification pass,
I got a nice "dispcal.exe has encountered a problem and needs to close"
run-time error? The error window popped smack in the center of the screen
where the i1 was measuring. I promptly moved it out of the way as I saw that
dispcal was still running in its DOS window. Dispcal terminated with "The
instrument can be removed from the screen" message. So, I am tempted to
conclude that the error message had no effect on the outcome of the
calibration.

On to some comments.

There are many, many things going on with dispcal, Graeme. Wow! Many things
for me to digest. I used the following command line:

> dispcal -v -yl -w 0.3457,0.3585 -b90 -gl -e1 -K -H 245bw

And got:

> Setting up the instrument
> Instrument Type:   Eye-One Pro
> Serial Number:     102242
> Firmware version:  101
> CPLD version:      101
> Date manufactured: 30-4-2002
> U.V. filter ?:     No
> Measure Ambient ?: No
> Tot. Measurement Count:          35761
> Remission Spot Count:             6378
> Remission Scan Count:             1020
> Date of last Remission spot cal:  Tue Mar 27 19:21:52 2007
> Remission Spot Count at last cal: 6367
> Total lamp usage:                 7557.952148
> 
> Place the instrument on its reflective white reference Serial no. 102242,
>  and then hit any key to continue,
>  or hit Esc, ^C or Q to abort:
>
> Calibration complete

So far so good. Now, what I don't understand is why did the program went
through a second round of instrument calibration below?

> Setting up the instrument
> Instrument Type:   Eye-One Pro
> Serial Number:     102242
> Firmware version:  101
> CPLD version:      101
> Date manufactured: 30-4-2002
> U.V. filter ?:     No
> Measure Ambient ?: No
> Tot. Measurement Count:           35761
> Remission Spot Count:             6378
> Remission Scan Count:             1020
> Date of last Remission spot cal:  Tue Mar 27 19:21:52 2007
> Remission Spot Count at last cal: 6367
> Total lamp usage:                 7557.952148
> 
> Place the instrument on its reflective white reference Serial no. 102242,
>  and then hit any key to continue,
>  or hit Esc, ^C or Q to abort:
> 
> Calibration complete

Was it the result of some of the arguments I used on my command-line?

This is what follows:

> Place instrument on test window.
> Hit Esc, ^C or Q to give up, any other key to continue:
> Display type is LCD
> Target white = xy 0.345700 0.358500
> Target brightness = 90.000000 cd/m^2
> Target gamma = L* curve
> 
> Display adjustment menu:
> 
> Press 1 .. 7
> 1) Black level (CRT: Brightness)
> 3) White level (CRT: Contrast, LCD: Brightness)
> 4) Black point (R,G,B, Offset)
> 5) Check all
> 6) Continue on to calibration
> 7) Exit
> 
> Doing check measurements

I pressed "5" and got :

> Target Brightness = 90.00, Current = 138.50, error =  53.9%
> Target 50% Level  = 25.70, Current = 29.53, error =  4.3%
> Target Near Black =  1.38, Current =  1.04, error = -0.4%
> Target W = x 0.3457, y 0.3585, Current = x 0.3478, y 0.3603, error = 1.21 DE
> Target K = x 0.3457, y 0.3585, Current = x 0.3432, y 0.3432, error =  9.09 DE

Here, I admit I didn't quite knew what I was doing when I pressed "5". What
confused me was the text in bracket for each choice. For example, for Black
level, I wasn't sure what this test would do for an LCD display, since it
showed CRT: Brightness in bracket. I guess what you mean to say, in bracket,
is "for a CRT monitor this relates to the Brightness knob". When I take the
time to read through your reported measurements things start to make sense
however. Maybe a better term for brightness would be Luminance? Because
that's the units into which dispcal is reporting brightness?

On to more interesting stuff. I got the following:

> Commencing device calibration
> patch 6 of 6
> Black = XYZ   0.38   0.37   0.47
> Red   = XYZ  71.50  36.67   2.92
> Green = XYZ  44.52  92.98  18.48
> Blue  = XYZ  16.37   7.00  88.62
> White = XYZ 133.61 138.28 111.51
> patch 64 of 64
> Target white value is XYZ 86.786611 90.000000 74.259414
> Adjusted target black XYZ 0.40 0.39 0.44, Lab 3.89 1.31 -2.52
> Target black after min adjust: XYZ 0.40 0.39 0.44, Lab 3.89 1.31 -2.52
> Current gamma curve offset = 0.129556, Gamma curve power = 2.669110
> Total Iteration 3, Final Samples = 64 Final Repeat threshold = 0.800000
> 
> Creating initial calibration curves...

I like how you make things explicit. No black box approach here ;-)
 
Then the rest of the sessions read as follows:

> Doing iteration 1 with 16 sample points and repeat threshold of 3.200000 DE
> patch 16 of 16
> Brightness error = 0.542557 cd/m^2
> White point error = 0.888139 deltaE
> Maximum neutral error (@ 0.000000) = 2.396818 deltaE
> Average neutral error = 1.241365 deltaE
> Computing update to calibration curves...
--
> Doing iteration 2 with 32 sample points and repeat threshold of 1.600000 DE
> patch 32 of 32
> Brightness error = 0.042274 cd/m^2
> White point error = 0.602503 deltaE
> Maximum neutral error (@ 0.000000) = 1.940052 deltaE
> Average neutral error = 0.824863 deltaE
> Computing update to calibration curves...
--
> Doing iteration 3 with 64 sample points and repeat threshold of 0.800000 DE
> patch 64 of 64
> Brightness error = -0.240370 cd/m^2
> White point error = 0.330876 deltaE
> Maximum neutral error (@ 0.000000) = 5.996416 deltaE
> Average neutral error = 0.805294 deltaE
> Computing update to calibration curves...
--
> Doing verify pass with 100 sample points
> patch 100 of 100
> Verification results:
> Brightness error = -1.507341 cd/m^2
> White point error = 0.241024 deltaE
> Maximum neutral error (@ 0.314318) = 66.765037 deltaE
> Average neutral error = 5.135653 deltaE
> 
> The instrument can be removed from the screen.

I think the next thing is to apply this calibration inside an argyll monitor
profile. That's my next task.

Roger Breton 



Other related posts: