- This topic has 83 replies, 4 voices, and was last updated 6 months ago by
.
-
Posts
-
I honestly have no idea what causes this error. The many markings in the ICC profile mean little to me. However, the only thing I noticed is that the ICC profiles generated with PMU take up a lot of space compared to others – 12.6MB. It is strange why so much, it can cause an error?
Still, we can’t complain – the software calibrates very well. But for idealists like me, such partial clipping of animations is annoying and contradicts the assumptions of macOS.
I honestly have no idea what causes this error.
Which error? PMU calibration errors on white point? It is likely to depend on WP variation across brightness scale (QC). Ask benq to make more steps in the WP+luminanbce target aproximation, then measure the 18 calibration patches for the LUT. Or apply an xy offset.
But by statistical chance, yours was not affected by this PMU issue, so you meant macOS animation issue.
The many markings in the ICC profile mean little to me. However, the only thing I noticed is that the ICC profiles generated with PMU take up a lot of space compared to others – 12.6MB. It is strange why so much, it can cause an error?
Qcal tag, report to BenQ both profiles and your issues.
Still, we can’t complain – the software calibrates very well.
You should because it does not… actually mievaan2 should complain to them because due to these QC issues his PME does not (fail to aim to target and PMU acknowledges it, by stored WP in profile), and you should complain about profile compatibility, ask for an option on PMU to create vanilla, simple, over idealized matrix profile.
But for idealists like me, such partial clipping of animations is annoying and contradicts the assumptions of macOS.
macOS assumptions is “buy our HW and out of the box it wil behave close to what it should” … and this assumption does not apply with a benq (or an eizo, there where issues aboyt wrong prixel format on first macOS for M1 and it was apple’s fault)
I honestly have no idea what causes this error.
Which error? PMU calibration errors on white point? It is likely to depend on WP variation across brightness scale (QC). Ask benq to make more steps in the WP+luminanbce target aproximation, then measure the 18 calibration patches for the LUT. Or apply an xy offset
I keep talking about the animation bug…
Of course, I’m going to write to BenQ Support.You should because it does not… actually mievaan2 should complain to them because due to these QC issues his PME does not (fail to aim to target and PMU acknowledges it, by stored WP in profile), and you should complain about profile compatibility, ask for an option on PMU to create vanilla, simple, over idealized matrix profile.
Honestly I asked a few other BenQ users. Models SW240 and others. None of the 8 people encountered this problem. Each model gets the perfect WP when calibrated with the PMU. Mievaan2 seems to have encountered a single problem with a faulty monitor. Or he just didn’t wait for the monitor to stabilize after turning it on – I can’t see any other explanation here…
So I calibrated again using a small offset as suggested by @Vincent. Results look better now.
It seems good! The gray balance is fine. Let Vincent take another look at it.
And I have a personal request – can you check the uniformity of your panel on as many control fields as possible in DisplayCAL?. I would appreciate if you do the first test in factory AdobeRGB mode and the next test in hardware calibration mode. Thank you in advance.
Hmm, unfortunately, my euphoria with the PMU program was too early. I am very sorry for potentially misleading other people. This program causes larger deviations than the factory calibration of the monitor…
I am attaching my calibration report below. The gamma graph is average and the RGB Gray Balance too. The PMU program unbalances the R, G, B components. The second report shows the factory condition and is perfect. The large Delta on the color temperature results from different information in the factory profile. However, the monitor is set perfectly to 6500K
What’s more, when I placed the BenQ monitor next to a better monitor, the BenQ after PMU calibration has green tints on gray gradients. This does not occur with factory calibration. Vincent, is it a big sin to use a factory calibrated monitor? However, as you can see, the factory calibration is much more accurate and better. I’m saddened because the hardware calibration on BenQ monitors is a scam! It’s good that at least the factory calibration done with Konica-Minolta is very good…
Did you use slow mode in PMU (enhanced grey calibration)?
PMU report is fine for what it is, a Benq SW, grey range <1.1 fine compared to some reports in reviews BUT I am seeing just a bunch of patches, I not seeing a lagom lcd test gradient PNG opened in a non color managed app like MS paint. That PNG test NON COLOR MANAGED (MS PAINT) will give you a more detailed info of grey.Factory calibration is fine excluding whitepoint. I assume that you have no RGB gain control in AdobeRGB preset. If white looks white use it (<4dE) but since native primaries are ~green AdobeRGB and ~P3 red its a waste not using it at full gamut in color managed apps and be limited to AdobeRGB
Does user/custom mode allow native gamut? if yes, set RGB gains to D65 and test if grey range is bad. You can apply GPU calibration to correct it.
Also any grey error from PMU or user/custom OSD mode can be corrected with DWMLUT. Simulate an idealized version of display after a slow speed 400+ patch XYZLUT calibration & profile. As long as uniformity is good, DWMLUT uses dither so it’s visually equivalent to HW calibration and these Benqs, dells, LGs and Asus can be corrected this way. OS display profile will be matrix+1TRC synth profile idealization that DWMLUT is simulating, these way all color managed apps will color manage upon this idealization, so you minimize the errors from undithered ouput in some apps.
Try it. As long as uniformity is good, DWMLUT can solve any issue from poor vendor HW cal solutions.You know what, PMU has a lot of shortcomings. In better gamma mode, it causes more greenness… and in this mode, the lowest contrast it sets is about 800. If better gamma is turned off, the maximum contrast is 550-600, a strange phenomenon in a monitor without DUE. In this mode, dark gray is less green, because gray is simply much lighter, so it seems better. Additionally, the program calibrates differently each time, zero repeatability – it can’t even set the gamma to 2.2. Remember what I wrote about animation stuttering in macOS with ICC PMU profiles? After updating to Sonoma in Safari, when viewing photos, RAM usage is 13 GB, without the PMU profile, the computer uses 6 GB RAM. These are some miracles… I’m giving up on this program, it’s crap. What’s more, the monitor looks worse after calibration and I can’t use the computer pleasantly.
AdobeRGB is the maximum gamut possible in the OSD, Native only via PMU, but AdobeRGB is enough for me.
How do you explain that the maximum DeltaE on a particular PMU gray is over 1, and in the case of factory calibration only about 0.20. Please explain to me why there is such a large Delta on whitepoint? After all, the monitor achieves a good 6500K. Is this due to the difference entered in the factory profile, or is each white different and I currently have something incorrectly set? I would like to avoid playing with profiling again, but I need to know if it can stay the way it is now.
You know what, PMU has a lot of shortcomings. In better gamma mode, it causes more greenness… and in this mode, the lowest contrast it sets is about 800. If better gamma is turned off, the maximum contrast is 550-600, a strange phenomenon in a monitor without DUE. In this mode, dark gray is less green, because gray is simply much lighter, so it seems better. Additionally, the program calibrates differently each time, zero repeatability – it can’t even set the gamma to 2.2. Remember what I wrote about animation stuttering in macOS with ICC PMU profiles? After updating to Sonoma in Safari, when viewing photos, RAM usage is 13 GB, without the PMU profile, the computer uses 6 GB RAM. These are some miracles… I’m giving up on this program, it’s crap. What’s more, the monitor looks worse after calibration and I can’t use the computer pleasantly.
AdobeRGB is the maximum gamut possible in the OSD, Native only via PMU, but AdobeRGB is enough for me.
I forgot that your use macOS, no DWMLUT in that OS. But… apple GPUs (AMDs or Apple Mx) should provide dither so if you ever need to calibrate using GPU, do not hesitate, it shoudl work and be bandless.
How do you explain that the maximum DeltaE on a particular PMU gray is over 1, and in the case of factory calibration only about 0.20. Please explain to me why there is such a large Delta on whitepoint? After all, the monitor achieves a good 6500K. Is this due to the difference entered in the factory profile, or is each white different and I currently have something incorrectly set? I would like to avoid playing with profiling again, but I need to know if it can stay the way it is now.
dE = error in color and error in gamma, just one scalar for 2 values. I didn’t bother evaluating dE in your report. Values were overall low and grey range was OK => OK
Bu since you are in macOS i’m afraid that IDNK how can you test lagom’s lcd gradient NOT color managed.
I use Photoshop without color management to evaluate gradients. After calibrating the PMU, the gradient is overall too green and has a magenta band in one place. Unfortunately it’s hard for me to take a photo of this. In the case of factory calibration, everything is perfect, the same as on the NEC calibrated SVII, so I have a good comparison. PMU software decalibrates the monitor instead of calibrating it. Additionally, I cannot freely use the computer with the faulty ICC file from this software enabled. I’m writing to BenQ right now, I won’t give it up to these motherfuckers. I feel cheated about the HW calibration of BenQ monitors.
Ah, yes! I forgot that, set PS color settings to no color management. I’ll try it on a mac soon. Thanks.
No problem! Maybe I’ll play around with profiling in DisplayCAL and see what results I get. DisplayCAL is wonderful. The only problem is that this program works very poorly on Apple M. Will there ever be any more updates (the last one was in 2019, i.e. before the release of Apple M)?
Vincent, but in the case of factory calibration, the green bar is less full. However, after hardware calibration they are nicely balanced.
Maybe that’s why the gray color in the factory calibration seems less green? And his correct hue are only after HW calibration and the factory calibration is deceiving my eyesight?
And should I worry about RGB Gray Balance, because after calibration it is not perfect like in the case of e.g. Eizo? Does this really affect the image?
“VDT”, you may be not measuring what you think. Just measure “uncalibrated report” on console and check dE to assumed WP.
Also instead checking RGB balance it’s easier (for me) to check measured a*b* fro greys and range… but if you wish calman like reports use HCFR.
