Demostrating Pimax 8k has real RGB subpixels with an 4k panel resolution. And explaining the source of its image problems


#122

choose 5k+, get more $100.
if have 8k+, paid more $100, lol.


#123

As an early backer I sent several mails about switching but no longer getting reply, can you tell if the first 100 got send yet please?


#124

I found a spec sheet for a Sharp 3840x1440 “rainbow RGB” panel. It has a zigzag subpixel distribution with full RGB rectangular pixels. That would be far better compatible with our scaler, not least since it wouldn’t need any vertical scaling. However, it’s only specified for 60Hz. I did not find what’s different about the 02 or 05(g) models, though they both list as 3840 X 2160. Compare with an AU Optronics RG/BR/GB panel. This concept of refusing to describe what you want to sell is alien to me.


#125

This really. I was all for sticking to the 8K until I saw Swevivers latest comparison and this was what shocked me. There are simply not more pixels available for the same picture wich is puzzling because there should be. The diamond pattern is a disadvantage by itself but the panel usage seems to be the bigger problem.


#126

http://www.panelook.com/LS055D1SX04_SHARP_5.5_LCM_overview_27124.html

Looks like the same panel. It says the panel refreshes at 120Hz, backlight flashes at 120Hz, but input is 60Hz. Don’t know what to make of that. Can you get 120Hz if you bypass the included controller (or whatever the thing that drives the panel is called)?


#127

Disassembling the panel is hardly an economical option, but sure, you could replace the driver. It likely wouldn’t change the rates you can get out of it; just because it’s specified at 60Hz doesn’t mean the driver would stop responding at higher rates. Overclocking is not guaranteed, so if you aim to do such a thing you’d contact the manufacturer about binning faster parts. That’s the sort of deal you’d expect for naming like CLPL.
As for the 120Hz strobes, like I said, backlight provides the exposure. Running a low duty cycle 120Hz backlight yields a perceptibly sharper image when motion is involved. It’s not as good for VR because we really want to align the image for every exposure, so we need the update rate too.


#128

All depends. It is not a use case scenario with different types of panels.

See also this comment:


#129

I don’t think so. Take a look here
http://forum.pimaxvr.com/uploads/default/optimized/2X/e/e6d332793b4c98d823cdeaa7636882738fbbc9cb_1_510x500.png

These are our oldest through-the-lens shots, showing the original 5K. You can’t see the sub-pixels but the grid of the 5K looks similar to what we see with the 5K+ at that distance.


#130

[quote=“antiflash, post:1, topic:8629”]But that is only true if the source image resolution is also higher than the panel to account for the re-sample to different pixel positions.
[/quote]
If to speak about 180 movies. We already have resolution 6144 x 3072. And it will be even higher I suspect. In this case, could this high resolutions overcome 8K’s bad subpixel pattern?
Is 8k better than 5K+ for 180/360 movies in your opinion?


#131

Yeah the headstrap, I chose from the far right of the screen where it wasn’t so dark.

If you look at the top blue border as it’s zooming in, you can distinctly see an individual small blue sub-pixel. There is also room there for a small red sub-pixel which is not turned on. There actually seems to be a lot of space between the sub pixels, this space just gets flooded with color normally I guess.


#132

Yep, I see what you are saying - it does look like a smaller, properly delineated square, about half the size of the green. On this picture, too, of course, we have the flooding, and blowing out, and there is a blue tint to all of it.

We need a microscope picture of the display panel directly, without the HMD lens. :7


#133

The @Pimax-Support is collecting info from the first 100 backers.


#134

Q) Have the first 100 got sent?
A) Support is collecting info from first 100 backers

What, but, how, ah forget it…

Think I need a break from all this lol


#135

I think the pixels are lizard shaped


#136

:sweat_smile::joy::joy::joy::joy::joy::joy::joy::sweat_smile:


#137

Do you realize that without NDA these problems would have been resolved by the date set for the mass shipment?


#138

Not 100% but I’m taking that as a “no” :sweat_smile:

I think “collecting info” means just compiling the addresses and the 5K+/8K preferences.


#139

Yeah that’s the one PImax used in the Pimax 4k. Like I showed in the other thread,it indeed really scales great: For reference: pimax 4k vs Vive Pro


#140

Wow! Just in a couple of days there are tons of post to read about this rare 8K panel.
I wonder if someone could do a summary about what we conclude about that


#141

I don’t suppose someone could just get a macro lens to actually take a picture of the pixels. ie