- This topic has 6 replies, 3 voices, and was last updated 4 years, 12 months ago by
.
-
Posts
-
Dear Florian! (didn’t find the right topic, so I write here)
We have a problem with usb ports on the new macbook! I mean, it has only usb-c, as all we know, but the colorimeter (i1 Display Pro) is with “normal/older” usb connector. Now, we attached it through a quality adapter (we tried two devices), but the measuring was total catastrophic, and we didn’t find out the reason. At last, I tried to attach the colorimeter to the monitor, it has usb ports (of course it connects to the macbook via usb cabel and works as a hub) and like this, the measurment was much better, at least green results, but it stays far behind the usual reviews’ results. The monitor is a BENQ SW2700PT, I’m sure, you know, this is one of the bests on the market now for the price.
Because the same reason we couldn’t make a correct calibrating for the macbook’s own screen!
So dou you have any experience calibrating new macbooks with only usb-c?
Calibrite Display Pro HL on Amazon
Disclosure: As an Amazon Associate I earn from qualifying purchases.At last, I tried to attach the colorimeter to the monitor, it has usb ports (of course it connects to the macbook via usb cabel and works as a hub) and like this, the measurment was much better, at least green results, but it stays far behind the usual reviews’ results.
AFAIK there just one objective & reliable review of that monitor showed “not bad” results in individual tests (prad.de) and they did not test HW calibration. All others showed poor color uniformity or severe bugs (still present) in Benq’s HW calibration software.
I mean… you may have read “wrong” reviews (or may not, IDNK which reviews you name)
Because the same reason we couldn’t make a correct calibrating for the macbook’s own screen!
So dou you have any experience calibrating new macbooks with only usb-c?
It’s a QLED screen. Xrite or Benq lack of spectral corrections for that kind of backlight… hence you cannot measure it properly with i1Profiler or Benq Palette Master Elements (PME).
Instead of using a proper correction Benq uses RGBLED spectral correction for verification. Some reviews testing “with the same rules” how good or bad PME performs use the same correction for verification.
You should use it too in your verifications (you may have done it but IDNK from the information in your message)Also if you attach your HTML results Florian or other people in this forum could point which configuration issues you have… or if your results show poor performance from monitor(+PME) like bad range a*b* an grey neutrality issues.
Make sure that you verify without “simulation profiles”, since you just want to know how it performs against profile made after calibration (unless you want to test factory calibration in sRGB/AdobeRGB modes)Also PME does not seem to store CHAD information when making ICC v2 profiles but they use PCS D50 white for white point… so unless you calibrated to D50 white, “display profile white vs measured white point white” will not match. This is PME’s fault, not monitor’s fault or DisplayCAL’s fault. Check “assumed white point vs measured whitepoint” instead, they should be OK.
PS:
DisplayCAL colorimeter correction database had some actual spectral measurements of a SW2700PT (or related screens with same QLED backlight) but unfortunately most if not all of them were measured in emulated gamuts smaller than native, like a sRGB or AdobeRGB HW calibration. Because of this, those CCSS files are not as good as they should be to use in DisplayCAL but it is not difficult to edit and “clean” them although it may be out of scope of this thread.Example, this is a sRGB-like HW calibration of a SW2700PT QLED:
https://colorimetercorrections.displaycal.net/hash/ac0f7037779be6ad43be8d344844f8a2/BenQ%20SW2700%20%28ColorMunki%29.ccss
Here you can see that emulated sRGB-like green and sRGB-like red are a combination of native gamut channels (3 RGB spikes)… clean the addition of unwanted native primaries for each channel.
Also this is NOT equal to a RGB LED spectral sample (specially in red) so if your i1DisplayPro sensivity curves differ from standard observer in wavelengths where RGBLED and QLED backlights differ from each other, this may cause a visual noticiable difference in a* axis.@Vincent, as far as I know, the SW2700 should have a RB (red+blue) LED with green phosphor backlight. The spectra are distinctly a bit different from (e.g.) Samsung QLED TVs (see Q9FN measurements from colorimeter corrections database).
They look the same to me with a little difference in green peak wavelength, extra blue seems to be just white point.
SW2700PT + HW cal sRGB emulation:
CCT = 6201.160190, VCT = 6003.811156
CDT = 6202.191530, VDT = 6148.608214
R= emulated sRGB red
G= emulated sRGB green
AFAIK panel manufacturer is AUO:
http://www.panelook.com/M270DAN02.1_AUO_27.0_LCM_overview_23626.html
But of course it may be RB+g, spectrally they look “equivalent”.Q9FN:
CCT = 6998.256510, VCT = 7163.769720
CDT = 7009.162553, VDT = 7436.297014
Anyway, @tarna if want to check PME with its own rules, trusting its measurements, use RGBLED bundled with DisplayCAL.
If reports are OK in assumed WP but you experience a small green-magenta cast, maybe you want to tweak it with DisplayCAL and some of available corrections for that QLED(needs cleaning) / RB+g backlights.Thank you for the answer, but I don’t really understand you. The graphs tell me nothing, sorry, I’m not an expert with charts, colorscience, etc… 🙂 Just calibrating my and my fellow’s monitors… 😉
I haven’t ever talked about Palette Master software, nor HW calibration, don’t understand, why you mention it at all… I calibrated the 2018 macbook’s screen, and the attached Benq SW2700PT monitor with i1 Display Pro and DisplayCal 3.7.1. And yes, I searched a lot reviews about the monitor, I’ve read everywhere really good results, even at prad.de (no not bad, it’s a 4 star Gut.. )
And again, the main problem was the usb-c port, with the adapter, we got bad results (don’t have the html chart sorry), but with my own 2015 macbook, which has usb 3 ports, no problem at all. That was my original question, did you guys experienced any issues with the new usb-c ports, but there was no answer, only HW cal, QLED and PME and charts… 🙂
And don’t really understand the colorimeter correction though. When I calibrating my monitor (Asus Pro Art PA279Q, DisplayCal, i1 DP), and I choose the Tools/Corrections/Import…from another.. or from the net, with that correction the software offers the gamut is only 90% of AdobeRGB, with no correction it is 98%, which is factory data… Interesting…
So you said, for the Benq SW2700PT, I have to choose RGBLED correction from the ArgyllCMS own database? (Tools/Correction/Choose colorimeter corr.)
It is an RB LED monitor, not an RGBLED.
Do I have to choose it also for my Asus? It’s a GB-r LED monitor.
Thank you
Thank you for the answer, but I don’t really understand you. The graphs tell me nothing, sorry, I’m not an expert with charts, colorscience, etc…
Just calibrating my and my fellow’s monitors… 
I haven’t ever talked about Palette Master software, nor HW calibration, don’t understand, why you mention it at all… I calibrated the 2018 macbook’s screen, and the attached Benq SW2700PT monitor with i1 Display Pro and DisplayCal 3.7.1. And yes, I searched a lot reviews about the monitor, I’ve read everywhere really good results, even at prad.de (no not bad, it’s a 4 star Gut.. )
Most objective test shows that SW series have very bad color uniformity (CCT test are not a valid test of color uniformity, it’s explained in several threads). Uniformity reports from owners (some of them former owners) made with DisplayCAL are not close to good.
Prad.de individual test is the only one with good uniformity. Prad.de individual test are VERY interesting but final score is useless because very often it says things which are in contradiction with their individual (& objective) test… so most people do not care about how many stars prad.de put in a monitor, just its individual test.To have a monitor with HW calibration and do not even try to use it… it’s a nonsense to me. Well, Benq’s PME is not a reliable software but at least I would give it a try. You can gain a lot of things (or do not suffer some issues) if it works properly.
Laptops are very prone to suffer banding issues caused by GPU calibration (the kind of calibration DisplayCAL or i1Profiler does). It usually is not application’s fault but caused by: iGPU LUTs, LUT truncation due to standby … so HW cal would help you in your particular situation (IDNK which GPU combination uses your macbook, you did not write it).
If you use PME you can use DisplayCAL to verify if PME did its job properly (no color cast in grey for example)And again, the main problem was the usb-c port, with the adapter, we got bad results (don’t have the html chart sorry), but with my own 2015 macbook, which has usb 3 ports, no problem at all. That was my original question, did you guys experienced any issues with the new usb-c ports, but there was no answer, only HW cal, QLED and PME and charts…
I would quote your 1st message and you’ll see why I wrote that:
Now, we attached it through a quality adapter (we tried two devices), but the measuring was total catastrophic, and we didn’t find out the reason. At last, I tried to attach the colorimeter to the monitor, it has usb ports (of course it connects to the macbook via usb cabel and works as a hub) and like this, the measurment was much better, at least green results, but it stays far behind the usual reviews’ results.
And don’t really understand the colorimeter correction though.
So it seems that you did both measurements in a wrong way so those measurement values cannot be trusted. You need to redo them in the right way (explained bellow and in previous messages).
Also “total catastophic” lacks of objective meaning: bad contrast? bad grey ramp neutrality to white point? smaller gamut than it should be? wrong white point?
You need to provide such information to solve that issue.Without colorimeter corrections your i1DisplayPro readings are not accurate, perhaps more than 4dE for white point (it depends on your colorimeter behavior & firmware data).
That is why i1d3 family was designed to use spectral corrections (CCSS files -text- in DisplayCAL or EDR -binary- in Xrite related software)When I calibrating my monitor (Asus Pro Art PA279Q, DisplayCal, i1 DP), and I choose the Tools/Corrections/Import…from another.. or from the net, with that correction the software offers the gamut is only 90% of AdobeRGB, with no correction it is 98%, which is factory data… Interesting…
It depends on what you did playing with PA279Q OSD controls, which OSD mode you use, what calibration curves looked like…etc. The information you provide is too small for a diagnosis.
Anyway, no correction means measurement is not very close to actual value unless by chance your i1DisplayPro is exactly the same as CIE 1931 2º observer. Xrite’s i1d3 are close to that observer but they need a little help to be on spot.So you said, for the Benq SW2700PT, I have to choose RGBLED correction from the ArgyllCMS own database? (Tools/Correction/Choose colorimeter corr.)
It is an RB LED monitor, not an RGBLED.
No, I’ve not said exactly that. Details are important. I wrote:
-Xrite or Benq software lacks of a proper correction for SW2700PT backlight.
-PME uses an innacurate correction for Benq SW series (they did not bother to get a proper one), but if you wish to validate PME results (using its rules) with DisplayCAL then use RGBLED correction. If results are OK with RGBLED correction the ony problem could be white point which could be close or not to actual value, depeding on some stuff stored in colorimeter firmware. If it looks “white” to your eyes then don’t care, it worked “properly“.
-DisplayCAL community provided you some QLED samples to use as spectral correction for SW2700PT. It is the only source of an accurate spectral unless you get an spectrophotometer and made one by yourself.So if you choose to do not use HW calibration at all, I’ll use a QLED/RBg spectral sample as correction (previous messages).
The one linked by Florian was measured in native gamut. It should be easier to use than the others since you do not need to “clean” spectral data like in the one I plotted. Using the other community samples needs some spreadsheet work on you side (“cleaning” explained previously).If you find some troubles or are not able to import that community corrections for some reason then use RGBLED correction. It not as accurate as it should be but it is the next closest spectral sample.
Do I have to choose it also for my Asus? It’s a GB-r LED monitor.
DisplayCAL 3.8, use bundled GB-LED U2413 correction. Same backlight as yours.
In previous versions use bundled “RGPhosphor” correction (its a GB-LED sample plus other samples)Summary:
-Best to use HW calibration (or at least try it), then verify it al it’s OK with DisplayCAL
-If you do not wish to use HW cal (why?) or if HW cal results are not very good when you verify it with DisplayCAL , you can do a GPU calibration using DisplayCAL. This is what you wanted to do.
-Whatever calibration or validation you do with DisplayCAL, you need to use a correction for your i1d3 colorimeter. Which one is better for each display as been explained above.
