What does this have to do with the visible part of the panel?
1 Like
Seriously, please read something about how lenses work, this is really really simple and it is like trying to explain it to a 5 year old. I need to draw you a picture but I can’t do that on my phone right now.
You’re talking about what you see through the lens, so the virtual image. What does that have to do with all this?
1 Like
The lens can only bend the light that hits the lens that is not occluded.
5 Likes
Exactly. It doesn’t matter how the lens bends the light. In fact it doesn’t matter at all how it looks through the lens. The only thing that matters are the outer edges of the virtual image (so the part seen through the lens) and finding those on the real image (so the panel) to determine the visual part by the lens. If you know that and you know the panel size then you can calculate panel usage. It’s really that simple.
3 Likes
Oculus patented the idea & the software Dev that attended the Berlin meet posted am article on curved display challenges/issues.
He did though compare quality to the xtal & was as they say impressed.
1 Like
StarVR is using 2 displays but rendering 4 viewpoints.
So… what are you trying to do? Alter the render resolution so we don’t waste GPU resources on non-visible pixels? Increase FOV? Find the ideal dimensions of for the panels to get a bit less SDE?
What is being occluded? The lens is in front of the panel not behind any plastic.
Well finding a solution is only possible if you know the problem. Right now people are already talking about better upscalers etc, while that really seems premature at this point. It really might be panel usage that’s the key problem.
Ok thanks for clearing that up. As I guessed, someone had had the same idea before me 
1 Like
Yes, this is correct, see how nowhere in this post you’ve mentioned the SIZE of the lens, which you have tried to use in every calculation you’ve done. You need to measure the panel used and the panel size, NOT the size of the lens.
LOL, did you actually read my 4 points? Again: if the plastic is angled at 90 degrees to the panel, then the part exposed by the plastic is the EXACT same size as the lens diameter. THAT is why the lens diameter matters (in this specific case)
1 Like
Settle down. Because these are special shaped lenses yes the lens size can be equal to the veiw port of the display.
Regular headsets use smaller diameter lenses than the viewable area. So the view expands until it reaches the final circle. For example the 4k uses 53mm lenses (not measured) where as the circular part of the panel might equal 83mm.
The 5k/8k due to having angled displays & much larger view area the lens size could be closer to the amount of the panel exposed to the lens.
1 Like
If the panel is much larger thsm can be seen then yes you need panel size to determine actual panel utilization. For example if i have a 3 inch circle in a 4" panel i am losing more panel than putting a 3 inch panel.
Simple geometry no physics required.
Could you two take this debate into a PM? AFTER you have settled this between yourselves, prefereably with a bit more scientific evidence/diagrams/calculations/ and any number of things, then feel free to share your results.
4 Likes
AAAAAAAARGH!
NO, the physical size of the lens bears no relation to how much of the panel it “sees” because its curved, its a fucking lens for christs sake.
Confusing physics; when your talking geometry & area.
CAN BE SEEN yes. But a lens bends light, a 3" lens can “see” a 4" panel.
My 2" glasses can see a 65" TV at 1 meter. I’m not losing 63" of my TV.
Ok with a picture then, look at this:
Imagine that the upper circle is the lens. The floor is the panel. The yellow part is the plastic lens holder, so the part between lens and panel. It’s angled 90 degrees to the floor (=panel). Wouldn’t the part exposed by the plastic be the same diameter as the lens? In other words, the upper circle should have the same diameter as the lower circle, if the angle of the plastic is 90 degrees.