Highest perceivable framerate? [good discussion]

Advanced display talk, display hackers, advanced game programmers, scientists, display researchers, display manufacturers, vision researchers & Advanced Display Articles on Blur Busters. The masters on Blur Busters.
User avatar
Chief Blur Buster
Site Admin
Posts: 7709
Joined: 05 Dec 2013, 15:44
Location: Toronto, Ontario, Canada

Re: Highest perceivable framerate? [good discussion]

Post by Chief Blur Buster » 12 Jun 2014, 22:10

ScepticMatt wrote:In my agreemend with flood, I feel that eye-tracking motion blur is the only real way to solve motion perception beyond using unpractical refresh rates.
And while motion blur solves the stroboscopic issue (ala the "mouse dropping" effect for the rest of us), motion blur is a far more critical problem during virtual reality, than with television screens or desktop monitors.

Consider that virtual reality has head tracking. Turning your head creates panning. Panning creates motion blur. And objects in VR is sharper than objects in video. There's no compression, no video softness, and a lot less antialiasing in computer graphics. And panning speeds in VR is much faster. (slow head turning creates VR panning motions exceeding 2000 pixels/second. Now think of faster head turning speeds!). Even at the same pixel density in FOV, the stroboscopic effects will become a bit more visible due to that.

-- In real life, you can still read reading this computer monitor while nodding or rolling your head. You're eye-tracking in order to read the text on the screen, to compensate for your head movements.
-- But in virtual reality, you can't read text on a virtual computer monitor while nodding or rolling your head, UNLESS you eliminate the eye-tracking motion blur. Because you're moving your head, the screen is forced to scroll/pan to keep things stationary/level relative to the real world.

Recent tests have shown that many of the virtual reality nausea/headache moments is caused by eye-tracking-based motion blur, and that low-persistence virtual reality solves a host of problems, in a very staggering way. As we all know know here at Blur Busters (and people of the likes of Michael Abrash, John Carmack, and Palmer Luckey), there's no way to have your (low-motion-blur) cake and eat it too (no stroboscopic effect).

From what I've seen with Oculus DK2 (had a brief demo with it), I notice eye-tracking motion blur is massively amplified with head-tracked virtual reality headsets. The luxury of the new low-persistence OLED DK2, means you can finally read text on walls even while you're moving your head, because there's very little eye-tracking-based motion blur. I feel there's an approximately 2x-5x+ amplifier factor, e.g. where you required 700Hz to eliminate stroboscopic effects on a TV, you need way more than that (very probably 2000-5000Hz+) to eliminate stroboscopic effects for virtual reality even for the same angular pixel density on your retinas. Consider a medium-speed head turning (in a head-tracked 4K VR headset, running framerate matching refreshrate) creating 8000 pixels/second panning motion, which would still create 8-pixel-gapping stroboscopic effect at 1000Hz.
Head of Blur Busters - BlurBusters.com | TestUFO.com | Follow @BlurBusters on Twitter

       To support Blur Busters:
       • Official List of Best Gaming Monitors
       • List of G-SYNC Monitors
       • List of FreeSync Monitors
       • List of Ultrawide Monitors

Posts: 36
Joined: 16 Feb 2014, 14:42

Re: Highest perceivable framerate? [good discussion]

Post by ScepticMatt » 15 Jun 2014, 20:32

Maybe I was using the wrong word. Maybe eye-adaptive motion blur would be better.

I didn't condone hold-type blur. What I was referring to was motion blur simulating real life.
We perceive motion blur if the angular velocity of the gaze direction differs from the angular velocity of a object
(Obviously that assumes that the object is lit in constant brightness light)
You'd need low-latency gaze-tracking. If you track an object across the screen, you'd get no blur.

I've posted a openGL demonstration here before. I think it worked rather nicely.

Post Reply