Join Date: Nov 2006
Location: San Leandro, CA, USA
Mentioned: 13 Post(s)
Tagged: 0 Thread(s)
Quoted: 431 Post(s)
About the HDR metadata stripping.
You have to consider what HDR10 is: Bt.2020 colorimetry and a PQ curve, representing peak luminocities of 1000 or 4000 nits.
Now, if you simply strip the HDR metadata (basically a flag stating the signal is Bt.2020, PQ, and the peak nit value), and replace it with a simple signaling of just Bt.2020, then the display will interpret the pixels as Gamma encoded with a peak of 100 nit.
The problem is that the pixels themself are still encoded with a PQ curve, and much higher intended luminocities. To do this right, the Oppo would have to a complex color volume modeling, in order to map the higher luminocity and PQ curve to a 100 nit Gamma curve. This will involve a lot of color volume clipping.
So, simply removing the HDR signaling, and repacing it with Bt.2020 signaling will not provide the intended results. It is possible that somehow the pixel value range of the PQ encoded pixel looks somewhat acceptable when interpreted as a Gamma encoded pixel, but that would be pure luck
I do not know what the Panasonic does differently, but I'm pretty sure that the HDRFury does not touch the pixel values to do any real color conversion.
Overall, if you have issues with feeding a valid HDR signal to your projectors, then you should bring that to the attention of the projector manufacturer, so that they can implement the correct handling of HDR10.