The current screen utilization is at 80% for both the 5k and 8k.
Does that mean that so far, those who have been wowed by the demoes are experiencing only 80% utilized screen? Near 100% utilization after optimization must be incredibly impressive then…
@Matthew.Xu @bacon can you please confirm if that’s the case? And also, will the 5k also be optimized to achieve near 100% screen utilization? Or is the optimization only for 8k?
I think it will be very difficult to get near 100%. First of all in the IPD adjustment process the panels themselves don’t move, the lenses do move which means that the projection onto the screen must move together with the lenses. This in itself means there’s space that can not be used for display and needs to be reserved for IPD adjustment. And then there’s the form of the lenses that most likely won’t exactly conform to the dimensions of the panel. 80% is already pretty good, maybe they can crank it up to 90% but I doubt it can get much higher.
You can calculate it: if you want IPD range 50-75 mm, then you need 25mm to move. Half of that per panel, so 12.5mm per panel should be reserved for IPD adjustment. I’m not sure how large each panel is, maybe 12.5 cm ? So that would translate to 90% maximum usage, IF you can adapt perfectly the lens to the panel, which is difficult in itself. So 80% already sounds quite reasonable.
For a bit of reference: http://media.steampowered.com/apps/valve/2015/Alex_Vlachos_Advanced_VR_Rendering_GDC2015.pdf described a 17% fill rate reduction by using a stencil mesh. That means pre-warp the blank corner regions were about 17% of the rendered area. 80% utilization isn’t all that low. The pixels per degree formula also in Pimax’ FAQ suggests it’s horisontal only though, but that might have its own uses (sjefdeklerk’s decription of IPD, my suggested porch tweaks to raise effective pixel rate).
Does more utilization mean more FOV? What about gpu requirements?
Id also like to know. They never really clarified what that figure means. Display utilization affects how big the fov is?
Or does it mean more display goes into the same field of view making the pixels per degree more dense?
In principle it’s the latter (if only display utilization changes, you get more pixels per degree). But that depends on the precise tuning of the optics; the pixels per degree aren’t uniform across either the screen or the field of view, and an adjustment might only affect a small part of either. For instance, if the optics were adjusted to reach further into the corners of the screen that has no effect on the density along the central lines (whether horizontal or vertical). There are also factors such as chromatic aberration, caused by light of different wavelengths bending differently. That means the display utilization for red is different from that for blue.
Can they use smaller panels(but with same resolution) in order to enhance screens optimisation?
Like from 5.5inches, got to 5inches…
Theoretically, if they found a source for panels that more closely matched the usable region for the lenses, and they matched the fittings or the case was redesigned for them - but that’s not going to happen. The display panels are already ordered and I think case machining has begun, so it’s basically tuning of the optics and electronics now.
@LoneTech Has Pimax actually explained what they consider to be 80% utilization? Is it something like described in the Valve’s paper, i.e. the inefficiency caused by nonlinear transform, or is it something along the line of only 80% of the horizontal size (scanline) is used?
The former would be actually an achievement (considering the circumstances), while the latter would probably put the actual utilization even lower than 80%.
I have a feeling that curved screens will be the next game changer for VR, probably resolve a lot of projection issues.
Not that I know of. The PPD formula they posted suggests the latter, but it is bound to be a simplification at best, misleading at worst.
And don‘t forget that we are talking of rectangular screens and the lenses providing you rather with an eliptical kind of picture, so there are inevitably going to be list pixels in the four corners.
Curved displays: I agree with @burstingtops and have been wondering about this one for some time already. You would think that displays should rather be kind of dome-shaped.
And of course they should ideally be fixed to the IPD adjustment and move along. This actually will likely be a weakness of the Pimax construction, because as has been pointed out above it will result in at least lost effective resolution, potentially even worse side-effects like differences in the lens distortion based on where the lense is when IPD adjusted, due to angled display construction the distance from lens to display may vary making things more complicated and less optimized for one fixed lens distance.
Yes. the utilizations will be same with 8k and 5K
Has Pimax every officially commented on this? Losing pixels in the corners is understandable, but I would hope the edges are being fully utilized.
It would be great to see an actual test result (easy to simulated, just render a line at the edge of the screen, and slowly move it into view).
hello bacon, what are the updates on the refresh rates and Max FPS of the pimax now? @PimaxVR @Matthew.Xu if anyone can answer 
Curved screens. Curved. Screens.
edit: Pimax team this is just a Skyrim reference - not a request lol.
Your not fooling anybody, thats a request lmao