Re: Monitor calibration software/hardware
Re: Monitor calibration software/hardware
- Subject: Re: Monitor calibration software/hardware
- From: Derrick Brown <email@hidden>
- Date: Fri, 21 Sep 2007 21:17:53 -0400
Hi Mark,
There is one point here in particular that interests me - a purely
technical one. Derrick, you may recall that some time back when you
and Jack made a presentation on colour management here in Toronto
there was discussion of the appropriate white point to select for
monitor calibration. This was in the context of "WYSIWYG" between
the soft-proof on the display and the print, all due allowances
made for the difference between transmitted and reflected light
etc. I distinctly remember the gist of the discussion being that
while one could logically think a D50 white point would correlate
properly with a D50 light source for viewing the prints, given how
we humans see, such is not the case; rather we get a better display
to print match with the display at D65 and the print viewing light
source at D50. It had something to do with D50 on a display looking
too yellow and dull relative to D50 lighting. Indeed, I find this
to be the case in my set-up, where the room is dim unless I turn on
the D50 lighting for viewing results. So my questions here: has the
passage of time and new information revised this wisdom?
Some time ago, especially as CRTs were aged, their respective
brightnesses were on the decline. As a result maintaining the D50
calibration with sufficient luminance became impossible. There was
just no longer enough blue to maintain both color temperature and at
the same time enough brightness. And you were likely using all the
signal the red gun had.
In cases like this especially, it was a much better idea to calibrate
to D65 because at least you now had some extra luminance.
I would also agree that the D65 whitepoint may yield a better
calibration on a display with out DDC controls since its likely
closer to the native white of the LCD. In the case of your 321,
depending on the 10 or 12 bit version I think your native white was
closer to 5700k.
Now, let me ask two supplementaries on the cyan/green issue Derrick
mentions below. When I run ColorEyes Validation, the software
generates 14 extra patches, of which according to the read-out
produced, 11 are shades of grey and the other three are R, G and B
at level 255 each. Usually, my validation shows insignificant Delta-
E for those three primaries - most of the DE occurs in the grey
readings. So with this kind of result, two questions: (1) where
does Cyan come into the picture (OK it's green and blue but not
measured as "Cyan", and (2) when I'm getting DE of 0.07 for these
primaries, what possible difference would it make whether I
calibrated my display (in this case a LaCie 321) at D50 or D65?
if you look deeper into the validation tool, there is an item called
"evaluate display profile". Inside this you will see a set of
reference files.
note the GMB macbeth. select it.
sun a validation. note the errors.
make two different calibrations. one d50, and one d65. i mention
this because the macbeth was developed to be viewed at d50.
so once you run these to profiles then evaluate, you will notice less
errors in the rich colors thru the d50 profile than that of the d65
profile. Now this also depends on the nature of your display, if its
a relatively new ddc LCD (yours is) I would expect the D50 to yield
you a closer match, if your booth is 5000k. What luminance are you
running? is that possibly influencing the display match to the viewer?
thanks!
Derrick Brown
Integrated Color Corporation
81 Rogers Street
N. Billerica, MA 01862
tel: 978-670-1416
email@hidden
_______________________________________________
Do not post admin requests to the list. They will be ignored.
Colorsync-users mailing list (email@hidden)
Help/Unsubscribe/Update your Subscription:
This email sent to email@hidden