Why there is no better screens in VR headsets ?

I think you're all slightly missing the point, so maybe I'm not explaining it right. There is an area that the eye covers where it does most of its work, this is before it makes more sense for the head to move. That space covers all the examples you've mentioned. Of course eyes flick about all over the place, but I don't believe that anyone drives around with their head facing forward and pushes their eyes as far as they'll go left or right. The area I'm talking about is the space between the most used area where the eye scans and the very edge of where your eyes can move.
 
Last edited:
which on current low FOV headsets is the entire screen area

and anyway, your eyes will flit to wherever there is movement, the head will follow IF it needs to - having low resolution without eye tracking will definitely be nausea inducing as you keep looking from high to low resolution areas

there is a reason absolutely no one is working on your "solution" but there are several variations on eyetracking to give high res where you are looking and low res away from the fovea, as pointed out already, there are already military grade applications working, they are just getting it down to consumer size/prices

I mean, in current headsets with uniform pixel arrangement, the optics already give a higher effective density in the centre and a lower density around the edges, they can and do already do that with a fixed screen, so what would be the advantage of having a panel that has a higher pixel density near the middle, as that is how it effectively works now and its awful exactly because as soon as you look away from dead center it looks crap.
 
Last edited:
Something people seem to not get, a higher res display has more benefits than they realise. You don't have to render at the display resolution for it to make a difference. Higher res displays will get rid of the screen door by itself, which will look much better than what we have now.
 
bloodiedathame said:
Of course eyes flick about all over the place, but I don't believe that anyone drives around with their head facing forward and pushes their eyes as far as they'll go left or right.
Click to expand...
Come for a drive or a pillion ride with me, then, and see for yourself.
A not-uncommon cause of failure in driving tests is people failing to use mirrors, not because they don't but because the examiner doesn't see any head movement. Instructors sometimes have to teach students to make these mirror checks more obvious.

bloodiedathame said:
The area I'm talking about is the space between the most used area where the eye scans and the very edge of where your eyes can move.
Click to expand...
In my case, both of those are inside the limits of the Vive and Rift.
When I play in VR it's like I'm wearing a diving mask. I can't even see all my forward instrumentation in Elite because it's so blinkered!
 
ttaskmaster said:
Come for a drive or a pillion ride with me, then, and see for yourself.
A not-uncommon cause of failure in driving tests is people failing to use mirrors, not because they don't but because the examiner doesn't see any head movement. Instructors sometimes have to teach students to make these mirror checks more obvious.
Click to expand...

I rode motorcycles for 7/8 years (taking a time out after accident/kids). I remember well being told by my instructor to move my head because it didn't look like I was checking my mirrors. That's not at the edge of your vision though.
 
ttaskmaster said:
In my case, both of those are inside the limits of the Vive and Rift.
When I play in VR it's like I'm wearing a diving mask. I can't even see all my forward instrumentation in Elite because it's so blinkered!
Click to expand...

Exactly, the FOV is so small in sonething like a rift or vive that its easily within the area your eyes woyld normally scan. Add to this that the optics already prioritise the centre for density that having a panel that also does the same thing makes the problem exponentially worse.

We are so close to having foveated rendering (or that varjo thing that uses a low res panel with a high res and mirror arrangement) that having a mixed density panel seems like a very short term dead end. It doesnt really solve a problem that exists, hence why no one is workibg on it as a solution.

You couldnt use a panel with a high density centre in a wide FOV headset either as the "centre" vision area would actually be more toward the edges of each panel for each eye as there is less binocular overlap in a wide FOV headset.
 
That foveated rendering is excellent, I'd not see that before. I guess eye tracking and a constant DPI across the whole FOV is cheaper than variable DPI screens, the overall effect would ultimately be very similar.
 
bloodiedathame said:
I rode motorcycles for 7/8 years (taking a time out after accident/kids). I remember well being told by my instructor to move my head because it didn't look like I was checking my mirrors. That's not at the edge of your vision though.
Click to expand...
Depends on the bike. Mine have always been big wide Cruisers and Tourers, where the mirrors really are well outside the main visual area. Added advantage is that I have almost no blind spots and don't have to body-shift to see more than my own shoulders!
 
spoffle said:
Something people seem to not get, a higher res display has more benefits than they realise. You don't have to render at the display resolution for it to make a difference. Higher res displays will get rid of the screen door by itself, which will look much better than what we have now.
Click to expand...

This is what I've been banging on about. I couldn't care less if it's a 1080p image upscaled to 4k. It'd be a massive improvement wrt the screen door effect. If they released a headset with the exact same spec as the rift, but with that solution in place, it would make desktop work completely doable and would make VR worlds immensely more beautiful.
 
bloodiedathame said:
That foveated rendering is excellent, I'd not see that before. I guess eye tracking and a constant DPI across the whole FOV is cheaper than variable DPI screens, the overall effect would ultimately be very similar.
Click to expand...
The display is only part of it. The problem is you'd still have to output at something like 4K x 2 x 90Hz, foveated or not. That's a LOT of bandwidth for a flexible 4-5m DP or HDMI cable. Paradoxically, going wireless and using compression might solve that issue.

Pimax's approach of upscaling on the headset is the most feasible solution at the moment. Once all the other bits fall in place and become affordable, true foveated rendering becomes a reality.
 
Like the CEO of polyphony said a few months ago, we're at least 10 years away from decent VR.

Everyone wants more pixels, larger fov, wireless, lighthouse tracking, lighter weight, faster response to limit motion sickness, better motion controllers

Unfortunately technology isn't ready...........and if it was, it would cost £££££

The trick is to get customers buying into VR so they can justify the R&D
 
I would say we have decent vr now :) Decent enough to still wow me coming from a monitor, and I feel the last year of gaming will be forever in my memory. Doubt ill remember much of any games in the last year when compared to the vr memories made.
Of course it can get even better, but to me those are more than decent if they came today. It would be a dream come true :D
 
Everyone's eyes move around large distances all the time no matter what you're doing. It's called saccadic movement, and foveated rendering needs to take it into account because although it's largely unconscious, the brain uses information picked up during saccades to build a mental map of your surroundings and fill in the detail that your brain perceives but you aren't really "seeing". Adapting VR to this highly tuned and sophisticated system that's been built up over millions of years isn't easy!

I personally think the VR we have now is pretty fantastic though....
 
You must log in or register to reply here.
Back
Top Bottom