“Are we really nitpicking over nothing?“

Yup.

“I want more subpixels”

I want clarity. 4K with pentile beats the competition in that regard.

Then don't buy it. Simple.

You know you've built a good headset when the review gush about how clear and colorful the image is, moreso than any other headset they've tried, and the haters have to break out the microsocopes to generate talking points.

I hear subpixels are so hot right now

People don't realize that pretty much all video they've ever seen on TV, blu-ray, youtube, etc. all follows a similar technique, but it happens in software. This is an over-simplified explanation of it (I'd have to go into a lot of detail to explain it all), but YUV compression works in much the same way... the human brain does not perceive detail in color, we see detail in black and white. So, to take advantage of this limitation, YUV compression... which all video you have ever seen shot by a camera uses unless it is true RAW (not even talking blackmagic RAW or ProRes RAW, as both use YUV compression as well). YUV compression samples some of the primary colors at lower resolution than the black and white (luminance) data. So, for instance, you might get blue at full resolution, but green might be half, and red might be quarter (commonly called 4:2:0). All "delivery" video formats (the stuff you actually watch) is 4:2:0. Pro-codecs for editing and capture sometimes use 4:2:2, and 4:4:4 is typically only used for motion graphics or green screen use, where having that extra color resolution counts when you need a clean background removal or composite.

The above ignores the translation to the YUV color space, which further compresses the video by mapping it to a color space that can be described using only X and Y coordinates, vs. separate bit data for the red, green and blue primary colors. If you'd like to learn more, look up YUV compression on youtube.

So, this is doing something very similar, but at a hardware level. One of the sub-pixels (the red, green, and blue mini-pixels that make up 1 full OLED or LCD pixel) is shared across pixels... so in effect, it's like getting 4:4:2 (when describing YUV sampling resolution, the 4 means full res, 2 means half res, and 0 means quarter resolution).

If 4:2:0 is good enough for all TV, youtube (and all other streaming services), Blu-Ray, 4K Blu-Ray, etc. and we still don't notice the difference... and 4:2:2 is good enough for professional color grading and image capture, 4:4:2 is going to be a step above even that... and again, this isn't even using the further-compressed YUV color space, it's still uncompressed RGB where each primary color gets its own 10 bits of data to describe the color it needs to be. in other words, good luck noticing it, even if you try. These folks had to break out the microscopes to do it, after all.

All a bunch of FUD over nothing.

Source: 28 years in the video production industry producing everything from SVHS tapes back in the 90s to 4K HDR today.

As an early adopter of everything VR (PSVR, PSVR2, Meta, Meta 2, Pico 4) I have to say that PSVR2 is a huge step forward in the VR gaming space. With a few but not huge drawsbacks.
Pros: Light Weight, Confortable, Excellent display colors, great controllers, haptic feedback.
Cons: Cable, small sweet focus spot (needs adjustments to get the focus right),
As for the games I only got exclusives as I can play everything else on my 4090 with pico 4 (way better and powerfull).
GT7 is nice in VR but if compared with Automobilista 2 or Assetto Corsa Competizione feels a bit behind. Lot of options but the visuals are a bit blurry
Call of the mountain is mainly a climbing simulator with a story and some fighting. Excellent showcase for the psvr 2 but FAA does some weird stuff
Village VR, excellent and so I went back to finish RE VII to start this, but first 10 minutes gave me heart attach

For the subpixel I do not see the problem, people will try to play things down on convenience.
Cheers

When the replies start coming in with explanations on what the eye can actually see then you know people are just talking bull. lol

When people ignore decades of experience and scientific research in favor of their own bias, you know they are full of bull, lol. None of you even knew you were seeing a similar effect in every video you've ever watched on TV, your phone, your game consoles, or your computer... that is enough to prove the point right there.

@Fr0sty
"People don't realize that pretty much all video they've ever seen on TV, blu-ray, youtube, etc. all follows a similar technique, but it happens in software."
It's true that most people don't understand color models. I'll say I'm no expert myself.

"All "delivery" video formats (the stuff you actually watch) is 4:2:0. Pro-codecs for editing and capture sometimes use 4:2:2, and 4:4:4 is typically only used for motion graphics or green screen use, where having that extra color resolution counts when you need a clean background removal or composite."
Also mostly true.

"So, this [PSVR2] is doing something very similar, but at a hardware level."
No. If you look at the sub-pixel structure of an imaging device like your camera's CMOS image sensor you will see a similar arrangement of subpixels, it's called a Bayer Array and like the Pentile Array used to make up the PSVR2's displays the number of subpixels is actually HIGHER than in typical linear RGB pixel structures. I don't think anyone would say that Samsung Galaxy phone screens are blurry, but guess what? Nearly every SuperAMOLED panel from Samsung uses this pixel array format. Yeah your phone screen if it's provided by Samsung (the screen not the phone) will have this Diamond Pentile Pixel Array.

Now I know Bayer and Pentile arrays are considered unrelated due to their application, but if we ignore perception and look at the arrays similarities we see the same pixel structure! You can see a typical Bayer filter for an inage sensor allows for each pixel to have one red, one blue and two green sub pixels. If we look at the common modern SuperAMOLED mobile displays used by Samsung phones and indeed in the PSVR2 you see the same thing. A square pixel with a single red, a single blue and two green subpixels. The only difference is that the Diamond Pentile Pixel Array is rotated 45 degrees.
Compared to a conventional LCD sqare pixel which typical arranges three rectangular subpixels from left to right with one red, one green and one blue subpixel in each pixel.

And what's the difference between a square and a diamond? It's rotation.
Seriously these are still square pixels still capable of providing the same colors but the subpixel array is rotated a bit. Now obviously the complexity of the display controller and wiring is different due to the orientation of the pixels and I assume that there is a power savings from the Dimond Pixel Pentile Array structure, but ultimately having less obviously square pixels will allow for more organic looking images with even less risk of running into the screendoor effect (which was also considered low on the PSVR 1 though it was still noticeable).

In the end Dimond Pixel Pentile Array screens have a definitive resolution and are still clearly capable of producing extremely sharp images, the article is still trash.

But I replied because ultimately you're also mostly wrong. There's no HW "4:4:2" color model magic happening here not even in a sort-of-kind-of conceptual sense. These are standard pixels with a "strange" subpixel layout that most people are ridiculously familiar with already without even realizing it.

^

I will, but I think that contradicts the whole necessity of having to get a clear view of text when first putting the headset on. If foveated rendering just means you get a clear and crisp image wherever your eyes are looking, then in theory I should never see a blurred image, yet I do.

Well what I’m seeing is if it’s losing tracking of one of your eyes. When you go to the setting that shows your eyes in the eye tracking setup it shows how much of your eyes the camera can see. My wife needs the eye width setting a lot closer than I do. When looking at that screen you can actually see it “losing your eyes” because it’s not straight or your full eye isn’t visible. I’ve never had the Fov rendering not work for me. I’ve never seen anything blurry like it couldn’t keep up with my eyes. Anyway, try that and once you’ve fixed it you should never have that issue again. It can’t work if your eyes aren’t properly set though.

The lens separation setting. Sure, I'll re-calibrate that. Thanks.

@ Deathdeliverer

I adjusted the distance between the lenses and as you say, the original setting was too close and the headset wasn't seeing one of my eyes properly. I increased the spacing and completed the calibration. I do remember making an on the fly adjustment a few days ago which must have messed up the spacing.

I re-did eye tracking while I was at it.

Anyway, GT7 was a lot better but I still think eye tracking and foveated rendering are two different things.

@Knushwood Butt

Yes, eye tracking and foveated rendering are two seperate things but there would be no foveated rendering without eye tracking. Yes, GT7 uses foveated rendering, and quite aggressivly I must say. It should not be visible in the headset, though. When you're doing the eye tracking calibration, is everything working correctly? Is it following your eyes? I have a bit of lazy eyelid on my right eye so I have to be pretty concious about keeping it fully open when I'm playing on mine.

@Boing

Thanks for saving me the time, can’t have foveated rendering without Eye tracking. That’s why people with one eye cannot use it. A lazy eye, as you’ve pointed out, could cause issues.

@Knush

Glad it worked out.

Thanks. The eye tracking test worked fine last time I tried it.

Next time I'll spend some time just sat in the car to see what I get when only moving my eyes.