BenQ 27 calibration for gray neutrality

[deanwork]

I recently bought the BenQ Sw2700PT pro display to do critical color work with.

I have been very happy with it. It has very consistent illumination across the plane and I have no complaints for working with photo printmaking.

This was calibrated and tested in various ways, but primarily with my I1 Pro 2 spectro and I ended up using the X-Rite Software and tweaking the result with the BenQ softare.

I use the Atkinson printer evaluation target to compare the image to print and the on screen image to my other display, the Nec PA271 calibrated with the Nec Spectravision software.

Both of these displays show the same hues of the target, same contrast and the same saturation of colors. What is differient is the neutrality of the gray images. I notice this on the large neutral black and white steel ball image in that target. The BenQ is a cooler gray, tending very slightly toward magenta, where the Nec is apparently neutral.  My NEC is 7 years old though. This is probably not going to effect my printing since I don't soft proof the hue of black and white prints, and the color work is fine. It just kinda of bugs me.

Does anyone know a way to tweak the gray neutrality of this display without adversely effecting the rgb color content? Someone told me on another list that you can not tweak the X-Rite calibration results in the Ben Q software afterwards and get good results. I was tweaking the black point as my gray image on the target was slightly darker than my Nec as well as slightly off neutral. Other colors right on.


I like this display enough to order the new 4K 32 inch version. The price is definitely right.

[keithcooper]

Have you tried the latest BenQ software with a larger target size? One reason I used their software with the SW2700 and recently when testing the SW320, was that it can work with the hardware calibration of the monitors.

There are several profile creation options worth experimenting with in the software - I'd prefer to use just the one package. Multiple software packages just suggests a lot of chances for unintended interactions?

Writing this though has reminded me that the SW320 is due a calibration and that BenQ has updated the software - this needs a restart for my Mac, so time for some more testing...

[keithcooper]

Well, just updated the BenQ monitor software and here's what it's saying (de2000) after running it with an i1 Display Pro (large target, 16 bit LUT profile  6500/2.2/100 )

Minor irritations are that the 'writing to LUT' stage of the software is slow and that I can't use the software with a 2nd monitor, so have to switch to another

[Doug Gray]

Well, just updated the BenQ monitor software and here's what it's saying (de2000) after running it with an i1 Display Pro (large target, 16 bit LUT profile  6500/2.2/100 )

Minor irritations are that the 'writing to LUT' stage of the software is slow and that I can't use the software with a 2nd monitor, so have to switch to another

Interesting. The pure green is outside ICCLAB's gamut.

The Lab values are adapted to D50 from D65, xyY colors are, of course, correct (white point=D65). The green RGB (0,255,0) is clipped since ICC Lab space is limited to -128 in the a* and b*. The actual pure green CIELab values (adapted for D50) are  (83.8, -143.3, 85.1) from Bruce Lindbloom's calculator.

This is an intrinsic limitation of ICCLAB and use of ICC profile LUTs which requires LAB PCS. Matrix profiles do not have that but you can't get as good color accuracy with them. Not an issue with printable colors of course.

Unlike ICC profiles, LUTs that are inside a monitor are not limited to ICCLAB.

That's quite a bit of clipping. Curious as to what unclipped, measured values are.

[keithcooper]

Interesting - I did look to see what i1Profiler would report, but as you might expect it doesn't like the profile produced by the BenQ software.

I've still got the SW320 here for a bit, so any suggestions for better comparing the various profiling options (it's on a Mac) are welcome...

[Doug Gray]

Interesting - I did look to see what i1Profiler would report, but as you might expect it doesn't like the profile produced by the BenQ software.

My EIZO, native D50, reports the green (0,255,0) target as Lab(85.84, -140.74, 93.98) and measured as (85.47, -141.80, 93.37) with deltaE2k of .35

But the EIZO's LUTs are in the monitor so it's not limited by the ICCLAB gamut. The created profiles operate in XYZ space. So long as colors are displayed only though ICC profile (XYZ) space those high saturation green colors are valid. But when they are run through ICC profile LUTs in PCS LAB space then they will be clipped. What clipping is used, I don't know. Most likely it is just truncating at +127/-128.

The MacAdam gamut exceeds (slightly) the ICCLAB limits so, theoretically, a printer could run into the ICCLAB limit. I'm not aware of any that do.