PPD discussions

Hi guys,

I tried to estimate the PPD of headsets with this basic calculation (ignoring the lenses)

PPD = Horiz. Res * Panel Untilization (%) / FOV (horiz)

I assumed a panel until on the headsets of 95% and on the Pimax around 85% large FOV and 70% normal FOV (taken out of Swevivers Review)

Horiz. Res Panel Untilization (%) FOV PPD
Rift 1280 0,95 95 12,80
Odyssey+ 1600 0,95 105 14,48
Reverb 2120 0,95 105 19,18
Pimax 5K+ Large FOV 2560 0,85 160 13,60
Pimax 5K+ Normal FOV 2560 0,7 145 12,36

Am I right with these calculations?

3 Likes

Panel utilisation seems a bit off of it’s percentage… :wink:

3 Likes

No 100 % means 1 …
Easier for multiplication

1 Like

I figured that’s what You meant but it’s not what You wrote. Maybe just write that (1=100%) instead of “%” for weirdos like me… :face_with_monocle::grin:

3 Likes

Pimax 5k+ FOVs are 160° Large, 140° Normal, 120° Small (https://community.openmr.ai/t/all-the-different-fovs-of-pimax-5k/16053).

The PPD for Pimax5k+ is around 14 PPD inside 90° FOV (https://community.openmr.ai/t/pixel-per-degree-ppd-of-pimax-5k/19389). If you read the post it will also give you an explanation, why the simple formula you are trying to might easily oversimplify the stuff and lead to completely arbitrary numbers (not related to the real figures at all). Guessing the panel utilization cannot really mitigate that.

The procedure to measure/calculate the PPD for Pimax 5k+ is equally applicable to any other headset, which uses OpenVR, if you can render “the static image” to the headset and can get the grab of its warped version afterwards.

Of course far deeper and educational post on the subject you can get from Doc-ok (The Display Resolution of Head-mounted Displays, Revisited | Doc-Ok.org).

4 Likes

Not forgetting that even then we assume a 100% match between the convolution profile vs the lens properties. So on the Center of lens max ppd is obtained but the real value is lower the computed value towards the exterior of the lens

1 Like

Your calculations conclude that PPD changes when you change the fov in Pimax. It’s the same in all fovs at least in the centre.

6 Likes

Due to reducing rendered pixels on the display it can be diffrent. I have not the exact numbers of Utilization and FOV is in my case 144 degree that can also differ from person to person.

Thats your value maybe, but if I do the ROV test i get different values…

Also i wanted to have a “simplified” PPD calculation to see the relative increase

I agree with @Flinnt, PPD don’t change when you reduce FoV.
BTW you’re missing the 8k.

1 Like

The 8K uses upscaling, has theoretically high ppd but in reality not really…
PPD depends on the panel untilization…

Horiz. Res Panel Untilization (%) FOV PPD
Pimax 8K Large FOV 3840 0,85 160 20,40

The 8K-X should be similar to the Reverb

It is not “my” value. It is what the headset reports to OpenVR.

2 Likes

Ok and why do I see 144° on normal?

Either the test you are using does not work correctly, or you do not know how to use it correctly.

This is ROV test tool I doubt this is incorrect nor I think I can’t read numbers…
FOV can’t be determined by OpenVR there are to many variables… Eye distance, IPD settings etc…

FOV at which the scene is rendered actually must be determined by OpenVR, otherwise the app will not know how to render it. Plus the FOV reported by the headset is the technical FOV - something the application renders, but is not guaranteed that the user could observe all of it.

You are confusing several things here. Your FOV might be smaller than what the application renders, i.e. you might not be able to see full of the scene for various reasons, like your eye distance from the lenses, or having the part of the panel obstructed by the lenses, or distorted to the point it will no longer will be recognizable.

But you can never see bigger FOV than what the application renders, because it is simply not rendered and if it is not rendered you cannot see it. No matter how close or far you put your eyes, or whatever else you do with your headset.

On the other hand, IPD change does not change your FOV.

I do not doubt that you can read the numbers. I doubt that the numbers you read represent what you think they are. ROV test has been already raised up on the forum several times and I tried to summon the author to explain, why people are observing contradicting results, but he seems to no longer follow this forum.

Anyway, I do not know why it happens, and I really do not need to explain it. If you doubt my results you can verify them yourself on your headset by using exactly the same procedure I used. There is nothing hidden and no mystery about it.

3 Likes

A good point. twenty dingens

1 Like

The only way PPD would change with lower FOV is if you had lenses that had been designed for lower magnification, at the same or greater panel utilization. One of the many reasons early on why I suggested Pimax should have a modular lens design.

3 Likes

I think most of what makes 5k/8k unique and expensive is the lens design, the rest is “just” stitching existing components together and following Valve’s blueprints. If Pimax had designed 3 lenses we’d still be waiting for our HMDs…

1 Like

Not so. Testers recieved revisions of the optics for the M2 unit, and were able to swap them out without compromising the components. Pimax makes lenses in house.

It would indeed cost Pimax about $50,000 to retool to make new lenses, but, if they were purpose built for a lower FOV while using as much of the panel as posdible, the PPD bump could put in stricking distance of more expensive HMDS.

Actually, they just need a lower magnification lens. That shouldn’t be as expensive as a whole new design.

Point is, we know for a fact that they could do it.

1 Like