Pimax makes Beat Saber more fun to play

Pimax 8K Series
62

I have a Doc-Ok article that might explain the issue.

Display Lag - the world follows your head
In an HMD, the screens move with the viewer’s head, and any delay will cause the virtual world to move along with the viewer until the display system catches up. Imagine wearing an HMD and quickly turning your head to the side. Say it takes 30ms total until this motion is noticed by the head tracking system, the application updates its internal state, renders the new state, and refreshes the HMD’s screens. During this interval, the world will turn with you, and it will snap back to its original orientation once the delay time has passed.

 

Bad calibration - the world changes as you move your head
Another, more subtle, effect is that in a miscalibrated display system the virtual world does not behave as the real world would. Do a simple experiment: fire up some first-person video game that allows view configuration, such as Doom3, and set a high field of view. Then rotate the view and observe. The virtual world will display a strong distortion effect, meaning that the sizes of objects, and their internal angles, change as the viewpoint changes.

 

Article #2
Wrong virtual camera distance - world changes as you move your eyes
TiagoTiago asked if it wouldn’t be better if the virtual camera were located at the centers of the viewer’s eyeballs instead of at the pupils, because then light rays entering the eye straight on would be represented correctly, independently of eye vergence angle. Spoiler alert: he was right. But I was skeptical at first, because, after all, that’s just plain wrong. All light rays entering the eye converge at the pupil, and therefore that’s the only correct position for the virtual camera.

Well, that’s true, but if the current pupil position is unknown due to lack of eye tracking, then the only correct thing turns into just another approximation, and who’s to say which approximation is better.

Turns out that I was wrong, and that in the presence of a collimating lens, i.e., a lens that is positioned such that the HMD display screen is in the lens’ focal plane, distortion from placing the camera in the center of the eyeball is significantly less pronounced than in my approach.

 

He also has a good general intro into how VR works and some of the issues that face hardware/software designers.

1 Like
63

is there an available tool to measure display lag?

1 Like
64

It’s an old error known due to an incorrect profile for lens distortion.
It has nothing to do with the delay.

1 Like
65

Or a combination of things as Doc-ok article points out.

Read this berlin backer blog as he describes the effect if I understand from what he said a 5 to 10°

https://blog.infinite.cz/pimax-8k-m2-5k-impressions-7dc661e024c4

Screenshots for reference/highlights

Image Quality

Screenshot_2018-09-28-15-18-16.png1440×2880 435 KB

Optics & Distortions -(moving world effect maybe?)

Screenshot_2018-09-28-15-17-56.png1440×2880 425 KB

66

@SweViver
I recall seeing a tool posted in earlier topics. It looks like a mannequin head with cameras for eyes. I think it was a bit on the pricey side.

All you really need though is to have a camera mounted to the HMD that can see the screen and the outside world.

  • Take the S8+ and tape it to the HMD
  • Take a slo-mo video
  • Move the HMD
  • Count the frames in between moving the HMD and the screen updating

The S8+ can take a 720p video at 240 fps, so it has a temporal resolution of 1/240 = 4.2ms. # of frames * 4.2 = display lag in ms.

1 Like
67

The SDE of the 5K+ is better than that of the Odyssey.

4 Likes
68

Thank you for your reply, it is very helpful.

1 Like
69

Better than Oculus Go? For me Oculus Go is perfect. But I see dots and some strange pattern in 5K+. And people who I knows, and tried 5K+, said that SDE has a strange “pattern” and prefer by far 8K SDE.

1 Like
70

That’s answer for me… is all. 8K is perfect then. Because I love the clarity of the Odyssey, but not the SDE.

1 Like
71

i think its worth doing. i bring this up because the 8k already has a lower refresh rate, plus supersample and the scaler operation in the headset. how much is that adding to the display lag on 8k. is this whats causing the feeling of increased sensation of scene panning with your head when you move etc?

1 Like
72

I have been switching back and forth in favor of 5K+ and then 8K and back to 5K+ and then 8K and it just keep going… I think Sweviver’s review by zooming into the pictures screwed up a lot of heads. For example, look at these pictures…explain this shit, it’s bullshit:

sweviver1.PNG1588×893 989 KB

It doesn’t make fucking sense. I think the 8K is probably not as bad as some of the pictures he took. Here’s another example:

sweviver2.PNG1336×928 1.14 MB

The only explanation for the crappy images on the 8K is probably because he’s not rendering high res at the time he took those pictures when he claimed they’re high res?

5 Likes
73

Oscar’s reviews, and those of others who have used his headset are such an outlier from the set of known dual HMD testers, as well as the feedback from Berlin, that I have to assume that

A) and I think this is most likely, he has a setting incorrect on his computer which is causing the distortion/frame dragging problems. Whatever that setting is, it’s making for a very bad experience.

or

B) He’s got some kind of axe to grind, because his review felt more like a hit piece than a balanced article of journalism for me. I guess I’ll know if his next Oculus review is glowing and fails to mention any negatives at all (this all happens during Oculus’ Connect 5).

Anyways, as I point out his review is such an outlier, that I feel comfortable essentially dismissing it from my calculations and interpretations of all the reviews out there. The general consensus is that the two Pimax HMDs now represent the #1 and #2 headsets on the consumer market. I’m really looking forward to getting mine :slight_smile:

These over-analyzed super zoomed in 1 degree arc photographs don’t represent the way you’ll see the world. Zoomed out video shots of the 8k and 5k+ show that they’re both AMAZING in sharpness.

VR Maniac’s through the lens 8k movie - the pauses were when he captured stills. Looks amazing. You don’t notice the blurring at the edges or see the pixels, because you see the whole image sharp and clear. The 5K+ is even sharper!

1 Like
74

For your A, it’s too much experienced. And he is a tester too, and TristanC is another, they know what to do.

For B, it’s too much professional. It’s a profesional tester from the Spanish national television, and has been in VR since the beginning, and tested all kind of devices before the Oculus DK1 and I think he has tested and has at home everything that has been coming since then until now to the market.

1 Like
75

To answer your question I took out my Oculus Go, and I was amazed by its brilliant screen again. You are right Oculus has outdone itself with Go. It’s screen stands very well against 5K+. I switched back and forth many times, but it’s hard for me to decide which one has less SDE. And Go screen has one plus over 5K+: it does not has the light tiny dots which can be observed in some color backgrounds on 5K+. Those dots could be a little bit annoying when watching movies, but it’s not disrupting, and they could be easily overlookd in games. Again this could be one plus for 8K for movies as 8K does not have those dots.

And I should point out that the perceived resolution of 5K+ is noticeably better than that of Go. So it’s still better than Go for watching movies IMHO.

1 Like
76

@yanfeng

May I ask you a question?

If you run 5k and 8k with same pitool settings and also same steamvr setting, does 8k look vast inferior in details compared to 5k or just a tad blurry? I’m not talking about text, I’m talking about game environment.

Thanks!

1 Like
77

Surely 8K is NOT vastly inferior in details compared to 5K+. My subjective impression of the difference is in accordance with the pictures I took in the test review. The difference is minor, and it varies in different scenes. The most noticeable difference happens in the text scene. In other game environment it’s less obvious.

2 Likes
78

@yanfeng
Sorry I didn’t make my question clear enough. ( cooking in full restaurant atm lol but I’m still surfing the forum) what I wanted to ask is at pitool 1.0 setting, and 100%ss which I believe is the ideal setting for GPU today. At this realistic settings would 8k detail be very noticeable loss in detail compared to 5k? I’m not talking about max pitool and max steamvr ss, sweviver already covered that.

1 Like
79

@oysta1109 My 8K has been borrowed away by my friend, but from my memory both 8K and 5K+ does not increase much graphic quality over PiTool 1.0/SS 100%, so my impression of last post should be the same under this “lower” setting.

2 Likes
80

Thank you. This is seriously a tough choice,
I’m leaning towards 5k+ but I’m afraid I will see more SDE that will make me focus on the SDE like I do on vive. Once seen cannot be unseen

2 Likes
81

If you consider that now we know the G/BR subpixel arrangement of 8K panel, the pictures are easy to explain. The first on (E:D text) was taken at different distance or the text in game has different sizes, which is also possible since the first one is from the cockpit’s HUD and the second one from mission briefing, or whatever.

You can count individual pixels on the letters on the left one and it seems the letters are 2 pixels wide. Given the flipping nature of G/BR arrangement (or diagonal, but I would prefer to call it flipping) only the pixels which form the checkboard matrix are lit. On the right pic, the letter width is more than 2 pixels as you cannot simply distinguish the pixels (as easily). My take is 4 as counted on “r” in “throttle”. So not the same font size, or different virtual distance from the text.

On the plane cockpit the speedometer on the left has the numbers almost precisely 3 pixels wide and since it has full RGB stripe, it can render the number in 7x3 matrix just fine (this is accidentally the smallest raster size width, you can render any font into). On the right size, 8K renders the numbers in 8x4 matrix (possibly 9x4), but 4 is not sufficient width for rendering two vertical lines on 8K panel, because of the flipping subpixel arrangement, you end up with pure checkerboard.

The speculative detour: The fact that on 5K+ all 3 subpixels (RGB) are clearly lit for white color rendering, on 8K it looks like only green subpixels are lit, since blue and red have much lower intensity. I would even dare to suggest that the green subpixel area in 8K panel is in fact composed of two subpixels, green and white. This would explain, why green pixels (or the area of green pixels placement) is so domininant on all 8K panel shots. The matrix would be like like this:

G/W   B/R   G/W
B/R   G/W   B/R
G/W   B/R   G/W

Now coming back to why 4 pixels horizontal are not enough to render two vertical lines spaced by one or two pixels, which makes all numbers on speedometer on 8K panel smeared inside:

Imagine those 4 pixels on 8K panel:

g/w   b/r | g/w   b/r
---------------------
b/r   g/w | b/r   g/w
---------------------
g/w   b/r | g/w   b/r

Technically you would want to light the leftmost and the rightmost columns to get the shape right (i.e. vertical lines)

G/W   b/r | g/w   B/R
---------------------
B/R   g/w | b/r   G/W
---------------------
G/W   b/r | g/w   B/R

But this would give incorrect colors. The panel driver clearly prefers color over shape, so it renders it this way:

G/W   b/r | G/W   b/r
---------------------
b/r   G/W | b/r   G/W
---------------------
G/W   b/r | G/W   b/r

Which is ending up in the visible checkerboard and completely loses the shape of the vertical lines.

The last pic (altimeter) is taken from greater distance as you can notice at the width of the “O ring” of number zero is 2-3 pixels wide on 5K+ and 4 pixels wide on 8K. This is enough macro for 8K panel driver to hit the color and the shape.

Added: @SweViver

3 Likes