Hi Chris,
I’m having some issues with the linearity of my monitor. In the past, I’ve externally calibrated the monitors (using an XRite) and then used those files to set the linearity in the display control. Now with the latest version of Mworks it doesn’t seem like that’s an option and I’m trying to use the internal linearization. It seems to be working fine on the mirror window (on the computer I’m using to control Mworks), but not on the actual monitor I’m using for stimulus presentation. It’s relatively linear up to ~70% luminance, but then saturates. This is not an issue with saturating the monitor’s pixel range since 50% luminance is 50 cd/mm2 and it’s saturating around 70 cd/mm2, well below the max brightness of the monitor. It makes me think that I’m missing some step for integrating the second monitor into Mworks. Is there something specific that I need to do?
Thanks,
Lindsey.
Hi Lindsey,
If you want to keep using your previous calibration method, you can just turn off MWorks’ color management: Go to MWServer → Preferences → Display → Advanced, and uncheck “Use color management”.
If you want to use MWorks’ color management, then you need to create an ICC profile for your display and assign it to the display in System Preferences. For my own setup, I’ve done this using X-Rite’s i1Display Pro and i1Profiler. If you can tell me more about how you’ve attempted to configure your main display, I should be able to help figure out what steps (if any) that you missed.
Cheers,
Chris
Ok- I hadn’t realized that the MWorks color management still used the
internal .icc from the calibration.
This now works fine.
Thanks,
Lindsey.
Chris,
While the luminance now seems linear with the calibrated .icc, the max
luminance is low compared to the calibration file (I calibrated it to 100
cd/mm2 and the max is only 80). It also has a greenish/yellow tinge. This
only odd color only appears after the first time I start- the background is
a more neutral gray when first loading the experiments. I tried turning
the “Use color management” off, but this didn’t change either the max
luminance or the color tint.
Lindsey.
Hi Lindsey,
While the luminance now seems linear with the calibrated .icc, the max luminance is low compared to the calibration file (I calibrated it to 100 cd/mm2 and the max is only 80). It also has a greenish/yellow tinge.
When you create the ICC profile, are you telling i1Profiler that you want a linear gamma? This may be what you’ve done in the past, but it’s not what you should do if you want to use MWorks’ color-management support.
Instead, you need to use i1Profiler strictly to profile the display. That is, you’re trying to create an ICC profile that accurately characterizes the display’s native color gamut, white point, response curves (gamma), etc. The only input you should provide to i1Profiler is the desired maximum luminance of the display (which is achieved primarily by adjusting the display’s brightness control, as directed by i1Profiler). Once you’ve created this ICC profile, either let i1Profiler assign it to the display, or assign it manually (in System Preferences → Displays → Color).
Once this is done, MWorks will take care of color matching from its internal color space (“linearized” sRGB) to the color space of the display (as described by the ICC profile you created). For a more detailed description of this process, see this comment.
This only odd color only appears after the first time I start- the background is a more neutral gray when first loading the experiments.
Not sure I understand this. Do you mean that, before you load an experiment, the desktop background looks OK, but once the experiment loads, the stimulus display background looks funny? If so, does the color look OK again once you close the experiment?
I tried turning the “Use color management” off, but this didn’t change either the max luminance or the color tint.
Changes to that setting don’t take effect until the next experiment is loaded. To see the effect for a currently-loaded experiment, you’ll need to close and reload it.
If it’d be easier to talk this through on the phone or over Skype, I should have time this afternoon.
Cheers,
Chris
Hi Chris,
I don’t see an option for choosing a “linear” gamma in the iProfiler. I am
currently using the standard setting for gamma- the only other options in
the drop down menu are Custom and sRGB. But perhaps I’m looking in the
wrong place- I’ve attached a screenshot of the Display Settings (this is
the first screen that follows the selection for Profile Display). I
typically do the calibration in Basic mode- does it need to be Advanced?
When I first load the experiment, the background is set to the mean
luminance. However, before pressing start, it is a neutral gray (at least
to my eye). It is only after pressing start that it takes on the
greenish/yellow hue. Then, if I stop, it stays the same hue.
I did try restarting Mworks and reloading the experiment after I toggled
“use color management”, but it still doesn’t seem to matter.
Thanks,
Lindsey
Attachment: Screen_Shot_2020-06-05_at_12.25.40_PM.png (32.1 KB)
Hi Chris,
Quick update:
I switched from the thunderbolt-DVI to a HDMI-HDMI cable and that seemed to
help with the hue issue. It also made it so that toggling the “use color
management” box now updates the screen brightness- and improves the
linearity.
However, the overall luminance is still lower than the 100 cd/mm2 that I
calibrated. I can’t find any option to change the version of the ICC
saved, and I worry that earlier versions of iProfiler won’t be compatible
with Catalina. I wrote to X-Rite to find out if they had solutions for
saving 2.x versions. I’ll let you know what they say.
In the meantime, I can manually increase the brightness of the screen
(increase it from where the iProfiler had me set it), and it maintains the
linearity, so that could be a solution.
Thanks for your help,
Lindsey.
Hi Lindsey,
I switched from the thunderbolt-DVI to a HDMI-HDMI cable and that seemed to help with the hue issue. It also made it so that toggling the “use color management” box now updates the screen brightness- and improves the linearity.
OK, great. I imagine this was the sole source of the problem, so I wouldn’t worry about the ICC profile version.
However, the overall luminance is still lower than the 100 cd/mm2 that I calibrated.
Now that you’ve eliminated the faulty adapter, maybe you should try calibrating again? It’s possible that the adapter was messing with the display brightness, too.
Cheers,
Chris
Hi Chris,
The low luminance measures were after recalibrating.
Apparently I can choose the icc version in the advanced mode. Will try
that.
Lindsey.