More Distortion On 8k Than 5k Plus (MRTV)

Thank you for the important information!

2 Likes

I haven’t watched the video, but have to say that at a glance it doesn’t really sound like a distortion issue at all, but a persistence one. Only way I can see that being fixable in software, is if the backlight strobe periods are software configurable, but if they are, it should be a really easy fix (EDIT: Just one value change. :7). :7

4 Likes

Yeah if the display sizes are different it would be stupid to simply apply the same distortion profile to both of them without adjustments or am i missing something here?

Maybe Pimax optimized the distortion profile using 5K+ slapped it on the 8K and called it a day but forgot that maybe the display on the 5K+ is smaller and now you have a shifted distortion profile on the 8K.

Could have been a simple oversight and if so they only would need to adjust the distortion profile to the changed display size and resolution. Maybe thats also why the 8K is perceived more blurry. mmhh

1 Like

Nah, the chassis is the same. If there is any difference to the screen size it won’t affect the relationship of the lens to the flat bit of the screen the viewable image is scaled to fit on so it won’t make any difference to the lens correction warp.

1 Like

@mixedrealityTV Can you please check if you see the same distortion on Oculus GO. Since GO has 60 hz or 72 Hz in some apps, this distortion should be visible, actually even worse. This is of course only if your assumption is correct, that the cause of the distortion is the 80 Hz refresh rate.

I do use the updated PiTool.

2 Likes

No, on the Go is no distortion at all. They have the perfect lens profiles since they only do 110 degrees FOV.

@mixedrealityTV I see. Then can you please check the Small FOV on the 8K ? Since the small FOV is 120 degrees. the distortion should then be lower or non existent, right ?

But if the the panel size is different the same pixel would be on a different spot on the panel and you would have to adjust to it or not? If so you would notice the slightest difference in panel size (see picture)

Let`s say the red square is the 5K+ panel size the blue circle represents the distortion profile for the red pixel.
If the 8K Panel is bigger represented by the black square the same pixel (black) is on a different location and therefore the distortion profile doesn’t fit anymore.

And as it looks like the testers reported different resolutions on the 5K+ and 8K at the same settings.
So if both Units would use the same distortion profile you will have a problem on 1 of the Units because you have a mismatch.

Ahh it’s late already maybe i shouldn’t think about things i don’t really have a clue of :joy:

2 Likes

@Digital You might be right. @deletedpimaxrep1 The same distortion profiles are used for both 8K and 5K+ ?

Don’t know it was just the first thing that was running through my mind and i thought better write it down before i forget about it again :grin:

2 Likes

Sebastian with your comment on the 91hz rate, we’re going back to the “golden” rule 90hz and more is required for a fluid VR experience.

Thanks again

Oculus GO has 60 Hz or 72 Hz and it provides fluid VR experiences.

1 Like

Considering how long it took any of the testers to realise this difference though, I’m not sure it could be classed as an iron-clad requirement fora fluid experience just yet.

1 Like

This is so weird, both @SweViver and @VoodooDE didn’t seem to ever notice that distortion on the 8k, and on paper @SweViver has tested several hundred games.

I hope @deletedpimaxrep1 can clarify this, but i’m not sure mrtv guess is correct, 80hz is such a huge speed that human eye can’t detect pixels turning on and off and that’s why low persistence works, i find it hard to believe that those 10hz difference explain that distortion, and why it only happens when you move the head fast in any direction but not when you move it slowly, it should be even more obvious at low speeds.

From what I understand, people originally argued that was because the oculus Go and gear vr are 3 degrees-of-freedom VR as opposed to the 6 degrees used in rift/Vive.

Regardless, I think how long it’s taken to highlight any difference at all using a 6dof device really speaks for itself whether 90hz is totally “required”.

And same thing with mixed reality devices on 60hz mode which is a 6dof device , it’s not a 90hz problem.

just a example

https://www.reddit.com/r/askscience/comments/1vy3qe/how_many_frames_per_second_can_the_eye_see/

level 1mrcaid60 points ¡ 4 years ago ¡ edited 3 years agoI
have done academic courses on cognitive neuroscience at the university
of Utrecht (Netherlands). It all depends on the training a person has
had. Fighter pilots have been recorded spotting 1/255th of a frame.
That’s right: 255 frames per second And they could give a rough estimate
as to what they’ve seen.
( more)
The average population would perceive about 45 frames per second (nice
going HFR movies). But on the other hand, you have 25% of the population
who will percieve more than 60 frames per second,

** Warning: don’t click on the link below / Don’t look if you are prone to seizers! ***

https://wolfcrow.com/blog/notes-by-dr-optoglass-motion-and-the-frame-rate-of-the-human-eye/

A clearly see 48hz blinking and it’s just 12hz short of 60hz. Could easily extrapolate than 90hz is required to see almost nothing

@SweViver Can you confirm the distortion described by @mixedrealityTV in the 8K ?

I’m not well versed enough in the back end technologies to know all the ins and outs, but a question:

If the scaler/internal pipe is only able to pass through a signal at 80 FPS, but the 4k panels themselves are 90 Hz panels (the original spec), then in much the same way a monitor without G-sync or such would just repeat frames if it’s refresh is higher than provided, is it possible that 80 frames are being sent to the panels per second, and they’re refreshing 90 times, so every tenth frame is a repeat? Do we know if these panels adjust to lower frame-rates? I know they were going to implement 60 Hz at some point, but that’s an easier one since it it more cleanly divides into 90… right?

1 Like