Originally Posted by Dave Harper
I’m not sure you’re reading what I wrote completely. On the LK970 graphs (not the graphs I posted, which are NOT from the LK970. They’re only there to show how the sweeps curve when they run out of that color).
On the LK970 gamut that tnaik4 posted, you can clearly see the measured green point is HIGHER (in amplitude) than the rec709 green point. On the graphs you just posted and are using for your stated position, it is LOWER. That isn’t the case with the LK970.
Until someone shows a sweep (and not just a final 2D saturation/phase point, you don’t know if the ACTUAL green at the rec709 point isn’t actually there with some calibration adjustments. What tnaik4 is showing is just to show how BenQ set it up to map when getting a rec709 green image pattern, which is wrong (for that mode). This is why many projectors need calibration, to bring them in line. Which is what I am saying is possible because the green point is HIGHER than the rec709 point, most likely due to it running out of gas shortly after achieving the rec709 point, thus veering to the right of the gamut chart ABOVE the rec709 point. This is a phase/tint issue, which can/should be able to be calibrated out.
I am reading it. What do you mean by "amplitude"? You mean because the green point is seemingly outside the REC709 triangle?
It is only outside the triangle in one direction really which means it isn't a given this can be fixed by calibration. If it were outside the triangle in both directions (eg, in the same vertical position but a bit further to the left) it could definitely be fixed at least by 3DLUT if not by internal CMS controls.
If this were a pure RGB mixing display showing 0,255,0 at its native gamut it definitely couldn't be fixed as there isn't any way to remove yellow / red once you are already at 0,255,0. On my JVC the only way you can achieve this is by using the cinema filter to remove the yellow light, which pushes the green primary further out (well, stops it being pulled in I guess might be a better way of describing it)
However I understand this display is not pure RGB, that it has a yellow component to the light being projected also which BENQ do for brightness. If the CMS allows the ratio of that yellow light to be controlled (directly or indirectly) then maybe it can get pulled back.
I'm just a bit surprised that in this thread which is now 40-odd pages long I couldn't seem to find much of anything being measured in that way achieving REC709 coverage.
If the projector is achieving it then great, that is good news, but the charts which were being used to make these points weren't making these points!
. Would be good to see some more results of these calibrated to understand what they can achieve. It is all very wooly at the moment. I'm sure it makes sense in what you've seen in your own measurements but there isn't anything I've seen published which agrees with "very high P3 coverage" or even "100% REC709 coverage". There is a dearth of quality calibrated data for these.