[argyllcms] Re: ColorMunki Create on CRT
- From: Mike Windsor <puddytat1234@xxxxxxxxxxxxx>
- To: <argyllcms@xxxxxxxxxxxxx>
- Date: Wed, 30 Mar 2011 11:06:25 +0000
Hi Derek,
Based on what you told me yesterday, I calibrated both screens with a white
point target close to the coordinates of one screen's native white point from
dispcal -R. The screens are now a lot more consistent with each other and
nothing is pink. I'll try again later once I've done some much needed reading
up on the different profile types I can choose from, how to select a white
point and what the dispcal -R results all mean.
I did mean the monitor's presets when I said it was set to 6500K. I didn't
realise I could change the individual channel gains... but fiddling with the
controls and the manual, I can't work out how to select "user mode"! Can you
remember how you did it?
With your help, I'm satisfied that nothing is broken and / or incompatible. I
expect I'll be able to improve my displays and hopefully have them consistent
with one another, so you've definitely helped me out a lot. Graeme's also been
kind enough to solve my other problem. I'm no longer worrying that the
colorimeter was a waste of money, or that I'd have to buy a couple of new
monitors in order to edit my photos more accurately!
I do have another question though: the target I set for white was (0.25, 0.35),
which was close to readings I got from an uncalibrated screen at its 7500K
preset. On the CIE 1976 diagram this looks to be purple: miles away from the
black body curve (about the same distance as between 3,000K and 20,000K). Is
this actually correct (it actually looks like a pretty good white to me,
certainly no visible purple tinge), or are absolute values from my entry-level
colorimeter just not all that accurate?
Regards,
Mike
----------------------------------------
> From: sigma3@xxxxxxxxxxxxx
> To: argyllcms@xxxxxxxxxxxxx
> Subject: [argyllcms] Re: ColorMunki Create on CRT
> Date: Tue, 29 Mar 2011 21:29:22 +0100
>
> Mike,
>
> Ok, first thing to say is that the monitors native / set whitepoint is very
> blue: this is good, since it means that it should be possible to adjust the
> monitor to 6500K, although you probably will lose some brightness along the
> way. You'll also probably gain some of that lost brightness back where the
> videoLUTs aren't doing extreme things to compensate for the colour
> temperature mismatch.
>
> Second thing to say: these are not new monitors, so you may find that you
> can't reach your desired brightness.
>
> I'm assuming that by "The monitor itself is set to 6500K." you mean that the
> monitor is set to the "6500" preset: I always found those to be rather
> optimistic in terms of the actual colour temp.
>
> I'd start by going into the monitor's colour setup menus and switching the
> monitor to "user colour", which allows you to adjust the Red and Blue
> channel gains.
>
> Now run dispcal, and in the display adjustment menu, use option 2, and
> adjust the channel gains to get a good match to your required colour
> temperature. You'll end up reducing the gain of the blue channel a large
> amount, I suspect.
>
> Check your monitor's maximum brightness now, and see if you can get an
> acceptable figure for that.
>
> I hope this helps.
>
> Best,
>
> Derek
>
> > -----Original Message-----
> > From: argyllcms-bounce@xxxxxxxxxxxxx
> > [mailto:argyllcms-bounce@xxxxxxxxxxxxx] On Behalf Of Mike Windsor
> > Sent: 29 March 2011 18:25
> > To: argyllcms@xxxxxxxxxxxxx
> > Subject: [argyllcms] Re: ColorMunki Create on CRT
> >
> >
> > Hi Derek,
> >
> > Thanks for responding.
> >
> > I tried to set the black and white points so they were within
> > the range of both screens, although, as you can see, I've
> > clearly failed as the uncalibrated white level for the one
> > I've examined below isn't up to the 90 cd/m^2 I've requested.
> > Is that likely to be a problem?
> >
> > The response from one monitor is as follows (the other one is
> > very similar in it's values, other than a black level of .16
> > and a correspondingly lower contrast ratio):
> >
> > Uncalibrated response:
> > Black level = 0.08 cd/m^2
> > White level = 85.83 cd/m^2
> > Aprox. gamma = 2.61
> > Contrast ratio = 1022:1
> > White chromaticity coordinates 0.2618, 0.3677
> > White Correlated Color Temperature = 8684K, DE 2K to locus = 25.2
> > White Correlated Daylight Temperature = 8636K, DE 2K to locus = 24.8
> > White Visual Color Temperature = 6330K, DE 2K to locus = 24.9
> > White Visual Daylight Temperature = 6511K, DE 2K to locus = 24.6
> > Effective LUT entry depth seems to be 10 bits
> >
> > I don't understand some of the data in the lines starting
> > "white..." or know whether the chromaticity coords are
> > reasonable (although they seem a long way from what
> > dispcalGUI seems to be suggesting correspond to 6500K
> > daylight). The monitor itself is set to 6500K. Are you saying
> > that for both my screens, 6500K daylight is simply going to
> > be unobtainable? If so, how do I choose something suitable to aim for?
> >
> > Thanks,
> > Mike
Other related posts:
