Had another late one last night to stay up and have a stab at calibrating the THX mode on my Sharp Quattron, and as suggested by others and expected by me, the results were much superior (both pre and post calibration) than Standard mode.
There were a couple of interesting observations I made along the way... one of the most concerning was that my 10-point scale adjustments did not always match the corresponding grey value, e.g. while the constant readings would visibly adjust when changing RGB levels at, say point 4 (using 40% grey pattern), making adjustments at point 5 with the 50% grey pattern had no effect whatsoever, and it was only once changing the pattern to 60% that changes to point 5 started to register. This virtually replicated across the remaining levels up to 10, as I had to use 70% pattern to adjust point 6, and 90% pattern to adjust point 7, 100% to adjust point 8, and I just winged the last two points as best I could. Very strange, and concerning that clearly my ten-point scale is not actually set to the expected 10% increases as per the included patterns. Any thoughts on this?
Also, when using HCFR via my laptop connected to my TV via mini-Display to HDMI, is there a way I can simply display the automatic test patterns just on the external monitor (TV)? I don't really see how I can use the constant read mode to adjust values when the test patterns obscures the program window where the readings are being presented.
Alas though, my post-calibration adjustments were very satisfying, and a lot better than I have been previously able to achieve (granted that my former calibrations were done on Standard mode). I have included three files in the below attachment, one the pre-calibration greyscale check after setting brightness and contrast (to 40ftL), and the remaining two are post-calibration, with different gamma levels between the two post versions. I had calibrated at Gamma +0 (I had neglected to lower this prior to calibration), but the resultant curve was very well balanced, with the main concern being that it was noticeably higher than the reference curve. Aside from that, I am happy with the other graphs, having an acceptable gamma range, RGB levels that are mostly solid at 100%, and a temp that mostly holds at 6500K. What do you guys think, based on the measurement files?
Following on from that, realizing that the gamma was still at +0, I reduced it down to -2 (out of a max of -3) and re-ran the full measurement checks (attached with Gamma -2 in the filename). This brought my luminance curve a lot closer to the reference curve, but the blue level remained high, the gamma was overall closer to 2.2 but the RGB levels showed a slight oversatiration of blue, and the temp is slightly above 6500K (it should be noted that at pretty much all of the 10 point positions, blue was minimised to its lowest level of -30 and green was increased to its max level of +30)
I guess my question is: based on the measurements between the Gamma +0 and Gamma -2 measurements, what would you recommend I stick with...
- Gamma -2, where the luminance curve the is close to reference but with slightly oversaturated blue and and a slightly-above 6500K temp, OR...
- Gamma +0, where the RGB levels are more balanced, the gamma sits more down at 1.9 and the temp mostly holds at 6500K, but the luminance curve is a fair degree higher than the reference curve?
My biggest struggles were when trying to calibrate the 100% colours to match D65, as I couldn't really get close (as evidenced in both pre and post calibration files) and I just feel they are way off, and no amount of adjusting the hue, saturation, value of each individual colour could bring them closer than what I have managed... but they all seem far more inaccurate when compared to my Standard mode calibrations (but the Standard greyscale is definitely inferior).
Thanks in advance for any advice you're able to provide you're able to provide!
Last edited by Clemery76; 08-12-2015 at 07:58 PM.