Lenny93 said:
The human eye can not percieve extra detail past a certain resolution (8K, 90 degrees field of view) so higher numbers does not always equal better. |
It's no quite as clear cut as that. Double blind tests have shown that humans can still tell which is the better display above 100 pixels per degree, but it starts dropping off rapidly after that to become guesswork between 150 to 200 pixels per degree. 20/20 vision is said to be comparable to 60 pixels per degree, but that's a lower limit at which you can pass reading tests, and see all the jaggies.
Test with 3 static images show probability dropping to 50% at aout 100 cycles or 200 pixels per degree.
Moving images could betray jaggies more easily though. But as you can see at about 85 pixels per degree (8K at 90 degrees) 90% of test subbjects can still tell which image looks more clear. For a 70 inch 8K screen you get 60 pixels per degree at 2.3 ft viewing distance, 200 pixels per degree at 7.6 ft viewing distance.
16K is only relevant for VR. Ofcourse the human eye can only resolve that much detail in a 2 degree fov, outside 10 you hardly see any detail. With proper eye tracking and foveated rendering, giving the illision of a full 16K VR display won't be anymore taxing than rendering 4K currently. You do first need dual 16K panels small enough to fit in glasses and the bandwidth to display dual 16k at 120hz. (Since only a small part of the image has actual 16K detail, smart compression will be key) Anyway that way you can have 102 pixels per degree at 150 degree fov per eye. (human limit)
And sure, some will argue 32k glasses will be better and in theory you should still be able to tell the difference.
Outside 10 degree fov your eyes resolve less than 20% of resolution. So basically your 16k image can drop to a 2K render 15 degrees away from where you're focussing. Ofcourse current headsets are only 1K ish per eye horizontally. Still foveated rendering can already begin to help.
