Would it be possible to design lenses that have the sweetspot axis angled at 100° to the lens plane? If that were possible, I could imagine the periphery becoming even more blurred.
My understanding is that it’s because of the slanted screens; the distance between the centers of the lenses is not the IPD.
4 Likes
This would indicate that the correct IPD for you would be when the HMD has a clear image but the geometry (distortion) profile is wrong and dose not match the IPD (sweet spot) of the lens.
1 Like
I would really like to know this! They are so silent! They designed this hmd, I believe they already know the answer…
1 Like
No. It would not indicate any such thing.
When I talk about the clear image here I mean clear image of the panel pixels. In other words, I cannot see clearly the pixels (or the panel matrix if you want). This has nothing to do with the distortion profile. The distortion profile makes sure that the geometry rendered by the game into perspective projection is perceived the same way through the lenses (which would otherwise add a non-perspective warping).
In fact, for being able (or not) to see the pixels, I do not care what is displayed on the panel. But I thought you already knew that (https://community.openmr.ai/t/pimax-is-there-a-plan-to-fix-eye-strain-and-ipd-issues-to-make-the-5k-and-8k-headsets-more-useable-for-people-with-less-than-60mm-ipd-small-ipd-160-fov-lens-replacements-and-adaptors/16925/312).
My point is that the image gets blurred because I am not looking through the lens center but through the outer ring. And I am not looking through the lens center, because the headset was designed this way.
3 Likes
@VTS @PMXskciVR
You can find answer to your questions in my post here https://community.openmr.ai/t/clarifying-near-ipd-x-distant-ipd-confusion/14809.
4 Likes
If your image is blurred then you are not using the HMD correctly, I can get the image in focus (JUST, NO BLUR) but I am using the center ring of the lens for looking straight ahead. If I open up the lenses and use the HMD the way you are using it (not looking through the lens center) its god dam horrible.
And how do you know this exactly ?
You might be right, but if you read my posts about the IPD, FOVs, and rendering, you will see that I spent quite some time in analyzing and understanding the design of the headset.
Besides, if you read my first post on this subject you may notice that I actually can see some parts of the panel clearly, they are just not corresponding to the same spot (in the virtual space). Anyway I guess I have no more to say on the subject (which I have already not said).
3 Likes
What basis do you have to make that claim? Seems like disinformation, because IPD is the distance between the middle of both pupils.
You’re trying to say the distance between the middle of the sweet spots of both lenses do not translate to an IPD reading, when they do because your pupils are supposed to sit in the middle of the sweet spots of the lenses.
Sorry but I have to disagree with you seeing as every other headset on the market has the same measurements and values for on screen IPD readings and the distances between the middle of the sweet spots of both lenses.
@VTS made the correct analysis of the issue at hand as have many others in this thread already.
Regardless of any cantering, curved lenses or Near VS Far IPD measurements, none of those explain why the on screen IPD reading is 10mm less than the physical measurement between lenses. Cantered displays and near IPD measurements would reduce the IPD, not increase it. Your eyes move closer together to look at something closer to your face (That’s Near IPD and it REDUCES your Far IPD by 3mm, it doesnt increase it.)
I really cannot see any way any of those issues or design choices can justify or compensate for a 10MM discrepancy in IPD reading. Cantered displays/curved lenses have NO IMPACT on the fact the minimum physical measurement is 70mm between both lenses, those arguments cannot and do not change that.
1 Like
True, and I am not in disagreement, I said: “the distance between the centers of the lenses is not the IPD.”
True for a “flat front” VR headset. My understanding is that the center of the Pimax lenses are adjusted outward, so that more of the screen is in focus, but maximum clarity is not straight ahead. Maximum forward clarity appears to have not been the design goal. I see “fine” straight ahead when the IPD of the Pimax is set to my IPD, but maximum clarity is slightly outwards, towards the middle of the screen.
I think there is a misunderstanding about what the balance of visual quality should be, between some users and the engineers. Currently, VR is a matter of trade-offs and it sounds like you don’t agree with the engineers’ decisions. That does not imply that Pimax must redesign the lenses or headset; they just have a differing opinion as to the weighting of the various factors, just like their choice to use Fresnel lenses.
2 Likes
I respectfully disagree, i believe most definitely that there needs to be a lens redesign considering the sheer amount of users reporting clarity problems and IPD issues.
The fact the lenses are angled outwards has no impact on the fact the minimum measurement between both lenses is 70mm, that doesn’t change anything.
Also the fact the Pimax isnt a flat front VR headset does not change the fact the IPD reading should be consistent with the physical minimal lens measurements, as they are in every other headset.
70mm does not equate to 60mm no matter how you look at it, curved lenses, cantered displays or Near vs Far IPD readings. That should have an impact of 3-5mm at most, yet here we are with minimum 60mm on screen, actually minimum 70mm when the headsets were advertised to be 55mm minimum IPD.
There is an IPD problem and Pimax needs to address it, likely with a lens redesign.
@anon74848233 @mozi @PimaxUSA @Sean.Huang
2 Likes
Respectfully, it seems like such a minor thing.
My IPD is 63 mm, when I set my 8K to 62.3, I get the best overall visuals. No noticeable distortion and my straight ahead view is SLIGHTLY less clear than when I look a bit to the side (through the center of the lens). This doesn’t bother me at all; the difference is hardly noticeable. Here’s the list of issues I have with my 8K, in order of annoyance level:
- Difficulties get the headset to connect to my base station (initializing tracking)
- Occasional black or white flashes (hopefully this will be fixed by a replacement cable, which is on the way)
- SDE
- Occasional lockups, when trying to initialize the headset (hasn’t happened in-game)
- Minor color fringes on text and thin lines
- Straight ahead vision is slightly fuzzy
- Slight pixel sparkle (hopefully due to a bad cable)
- Black level is dark gray, not black
Honestly, I think it’s a matter of expectations; I never expected a “perfect” headset. My 8K is better than I hoped for (assuming the new cable fixes the flashes). I can wear it for 4+ hours without eyestrain. Even with the issues above, it’s great, in games or for watching movies.
2 Likes
Actually, it does. I have already posted that many times in this thread and probably several times to you, but I will give it the last shot once more here. Please read my two posts here (https://community.openmr.ai/t/some-thoughts-on-the-ipd-discrepancy/14754) and here (https://community.openmr.ai/t/clarifying-near-ipd-x-distant-ipd-confusion/14809). They explain, how I made an analyzis of the headset optics geometry and how I tried to explain to people that because of this geometry the headset would not have the lenses (centers) in the same distance as are your eye pupils.
I tried to made it as clear as possible. If you do not understand any part, I will try to elaborate more, but please, first read it and digest it yourself. Once you do, you will also realize, that there is no mystery and every and all observations I made could be done by anyone else if he has the headset. There are no secrets. What I did, was just to write it down and draw the conclusion. You may dispute the conclusion, if you have a valid point, but you cannot dispute the facts, especially if anyone can verify them exactly the same way I did (and which I described)
4 Likes
A minor thing that’s causing too many people to count on these forums, reviewers and customers to sell their headsets because they simply cannot use it at all without causing eye strain or being unable to have clarity in both eyes at the same time?
Yeah nah seems like a massive understatement and downplaying of the severity of this issue. Headsets being totally unusable by a substantial amount of the community is in no way a “minor” issue.
Unfortunately a substantial part of the community can’t share that same opinion, because some users cannot use the headsets comfortably without eye strain or IPD issues at all, and as a result they’ve had to sell their headsets (many cases of this happening in the forums)
Whilst it’s great to hear your experience and (mostly) satisfaction with the product, don’t be so quick to discount the large amount of users here who don’t share the same experience.
Issues which have prevented users from enjoying the product at all, even after trying all the extensive suggested fitting techniques.
@risa2000 I’ve read both of your posts and what I can’t seem to wrap my head around is the fact the lenses are curved, by my understanding divergent displays and curved lenses should be giving a lesser IPD reading than what’s shown on screen, not more.
Take a straight line, and a curved line, both of the same lengths respective to each other.
Straighten the curved line out and it becomes longer. This is true for any line divergent from a 180 degree axis as well once straightened out, it becomes longer, not less.
So by that logic, curved lenses and divergent screens should result in a lesser physical minimal lens distance between the sweet spots compared to the on screen IPD reading, not 10mm more.
To me it seems Pimax intentionally misrepresented the on-screen IPD in order to get as much extra lens real estate as possible.
I am not sure what do you mean by curved (made it bold in your reply). If you refer to my schematics, there the shape with which I draw the lenses should not be taken literally, I was just using a known symbolic.
In fact the actual lenses in Pimax are supposed to be flat on backside (I did not verify it, but I believe @Sjef or @SweViver removed them and made such an observation) and slightly bulging on the front side.
If it is the actual shape you meant then the curvature of the front shape should not have a (significant) impact on the observations and claims I made. For all the purposes (of my statements in this and other threads) we can consider the lenses to be flat (not curved).
This one is difficult to grasp for me, if you could make a drawing, maybe it could help me to better understand it.
Regardless though, divergent displays (panels) alone do not imply any particular view, and I can imagine a different design which could use divergent (or angled) panels but coplanar lenses. Which means basically the same views as any other current headset.
I do not know if such a design would work (or work better), but it is technically possible. Pimax however chose to equip the panels with lenses oriented in such a way that they are parallel to panels, probably in an attempt to maximize the view and minimize the overall distortion over the whole FOV and because the standard rendering pipeline expects the view axes (optical axes of the view) to be perpendicular to the projection plane (i.e. the panel in this case) . The price they pay for this decision was that the left and right view axes became divergent by the fixed angular offset (by which the panels are angled).
The way we should look into the lenses is ensured by the headset driver which reports the view and projection geometry it to the VR subsystem (OpenVR, Oculus) and it subsequently reports it to the application.
Once all is set up this way, the application renders the (divergent) views (when parallel projection is off) and there is no other way for us to view them than the way Pimax designed it.
2 Likes
I wish people would let go of the near and far ipd red herring.
(EDIT: Convergence is real enough, as is centering on the lens for optimal use, but the concept is being brandished dogmatically and without understanding situational circumstances.)
Thanks @risa2000 for a once again so detailed post 
To all people having problems with the pimax optimal optical ajustment I suggest removing the padding and try to understand what is the best position vs your face. Take your padding less headset in your hands in front of you evidently 
and slowly move it toward the best positions for your dominant eyes. Then ajust and refine the ipd and position values for your other eye.
Dont’ forget to play with the parallel axial position (rolling the headset on itself front/rear)
Things I’ve found for me
0- original fabric head strap is crap, I use a modified deluxe head strap. Can’t have a good image with the strap provided… your nose should not hurt…
1- 6mm padding simply don’t work too blurry
2- 18mm padding is more confortable but I loose FOV and sharpness around the edges whatever position & IPD I place/ choose the headset on my face
3- 11-12mm cushions are the best for me with a strong inclination upward, this give a really big space between the headset and my nose
4- my IPD is 63.25 and the IPD value is not so important in my case and the Pimax gives me a really better sharpness and overall bigger usable region of view then the htc vive or the the Samsung OD+ Even with the small FOV option.
5- I tried to see distortions that would distract my experience but that is not the case with my 5k+.
6 - no more eyes strain then when I use a normal monitor for to long. My VR sessions are about 1 hour long.
7- both eyes see raisonnably clearly
8- I normally wear glasses both multi-focal and fix focal, but I don’t use them in VR
9 - use this application for your tests
https://knob2001.itch.io/test-fov-rv-testsglare-readbility-lens
4 Likes
Things are more complicated then that :
1-the current head strap is crap, I would be the guy telling you pimax headsets are blurry if I was not using my own head strap.
2- the first thing pimax must do is assert the problem in a controled environment with numerous people selected with racial and gender variations in mind.
3- Pimax could redesign the lens and simply have another 20% of people not been confortable with the new design. This is why big cie will avoid wide FOV headsets in main stream market because the optical system become to critical and it’s difficult to achieving the normal sigma value for those kind of commercials products
4- having a one design fit all may even not be possible…
2 Likes