I am not sure I am able to visualize “turning HMD in different directions, while lens faces straight ahead”, seems contradicting.
Anyway, I believe there are two possible reasons for what someone may observe as a “focus fall-off”.
Looking through the Fresnel lens in “non-canonic” way (i.e. through the peripheral ring and at an unfavorable angle). This (I believe) should be possible to mitigate by the headset placement (changing eye-relief and the lens position) and that was my suggestion I wrote earlier.
Simple inability to focus on the virtual image, because it is either out of focus by itself (i.e. the projected virtual image is defocused by the lens optical property), or because the eye acuity cannot handle it. For example, if the projected image is too close and your eye cannot focus at such distance.
Since there are people who claim to not have such a drastic focus fall-off, I would assume your eyes play some part in it. If you wear the reading glasses in the headset, does it change the perception of the focus?
Including (… amongst other directions… holding it in such a way…) so that the lens faces straight ahead; Not: “while”. :7
Just referring to turning the HMD ten degrees, so that one of the lenses sits in front of one of my eyes the way it would in an HMD without canted lenses; Essentially the same, optically, as looking ten degrees out to the side, when wearing the HMD normally. :7
You may have picked up on a general streak of inability to get anything across coherently, in my writing.
Yes; Aligning with the axis should be ideal, peering along it, with concentric lens symmetry, and this is what I usually strive for, but I have tried every possible position/orientation, to see what that does.
(Turns out the aspheric “conventional” side of the lens seems to line things up in such a way that even if you look at the lens whilst holding it at an angle, the fresnel rings kind of appear as hanging almost unmoving in the air on its other side, as were they gimballed – kind of cool (EDIT: And optimising alignment with the fresnel part, so that the negative effects of off-axis viewing should be strongly reduced). :7)
Yep, which is why I’ve been so curious about the virtual focal distance of the HMD, and whether it could be favouring the nearsighted (…which would still be ok for people without eyesight deficiencies). I use +1.5 reading glasses, when reading at arm’s length, but that only makes things worse in the headset – whatever my eyes need, that +1.5 is too much.
Experimenting with the lens loose, picked out of the headset, against a sheet of paper with fine print, I can find no position/orientation where I would call the view “good”, but I can get a slight focus improvement, by modifying the lens-to-screen distance by a fraction of millimetre; Past that, it goes off again. I have tried scraping a little around the edge of the lens, to get it closer to the screen (as opposed to shimming it up, which I have previously done), without cutting into the “shoe” it sits in - results are inconclusive. :7
This is just the basic focussing, without looking into the field curvature of the lens, and how it devitates from the plane of the flat display panel…
(Mind you: The lens has to do something fancy and asymmetric, because the screen is canted an additional 9 degrees out, on top of the lenses’ ten (whilst keeping the lens-to-screen distance constant, as far as I can tell) (EDIT2: possibly just trying to approach the field curvature in the periphery, by drawing closer there, than on the opposite side)).
Now it is easy to understand and visualize even for me . You should definitely be able to find a good view in this configuration (unless point 2) above applies).
Now, if +1.5 is too much, I would expect that “too much” will be exactly in the center and should be less “too much” when diverging from the view axis (center of the FOV). You could try glasses with +1 or +0.5 dioptres then.
I guess this fraction means moving the panels a bit further from the lenses, right?
Indeed. Such ‘focus fall off’ seems a bit weird to me. Not saying it doesn’t happen, but that I can’t understand how it would happen anywhere near so readily.
FIY: the closest it is physically possible to get the lenses center sweet spot is around (tolerance + 0.5mm) 62.5mm. Before the extreme mod is was around 66-67. I have em set to around 63.5-64 now and its perfect.
Sorry @jojon you’re screwed Maybe just maybe if we get index++ quality lens drop in its possible
So Pimax need to redesign the IPD slider, allow greater range, and the most important. A high quality drop in lens replacement.
This area could also be better designed. The headset is also unusable for people with a big nose or a high bridge of the nose. normal people also complained of the pressure on the nose. there is much to be improved. But it doesn’t matter what the community says, only what is heard.
Yes - at first at least. I have subsequently experimented with getting a micrometre or two closer instead, by carving into the edges of the lens - not sure where I am going with that… :7
Could be a good thing - I’ve never had my head properly screwed on before.
I guess the question I have is… how much better would the G2 lenses be at 160-170 Horizontal FOV? or even at 150 degrees?
I have this hunch- not based on any physics or optics expertise- that the wider the FOV becomes, the harder it is to make the lens, and the more difficult it is to expand the ‘sweet spot’.
I’m probably wrong, but- if wide FOV lenses were easy to make, wouldn’t the new competing lenses have them? The G2 likely has a scuba mask FOV for a reason.
Cheers
Easy no but my opinion on that is the hw at the time wasn’t there for the reverb and they simply don’t have the focus on fov and choosed to optimize. Hw is still a big deal though (lots think they really need a 3090 for the G2) and the other developers keep their gpu requirements lower while pimax kind of don’t care (end user experience is a huge deal for the others too but pimax neeh) and push out which is a really good thing but the rest is catching up fast now.
Also with the Index the hmd by itselfsn’t that expensive with great lenses so when the index comes with 4k, less glare on their lenses with a huge sweet spot while perhaps close to normal fov equivalent it will be bye bye Pimax unless they upgrade their lenses very soon. I really really hope there will be a drop in option. Like others also mention and me to a id happily pay for that.
Edit to add. 4k while having some slight sde (8kx don’t know G2) is so good now that you really should put the main focus on lenses anyways so regardless of anything else its the next logical step. If they could increase the sweet spot by 20-30% it would make a big difference
No idea what an ‘eyebox’ is. Sorry, you’re way over my head.
I thought swee tspot meant the area of the screen where the image was the clearest. So would that be angular range in focus?
For the most part the 8kx lenses deliver a good image to me right out to about 140 degrees. In real life, I don’t think my peripheral vision is much clearer than on the Pimax. What I’m focused on is clear and things on the edge ‘call’ my eyes over to focus with movement. (I think that’s how it works for me. I’ve been trying to sort that out!
Cheers
Dave
A lot of people talk about it (how badly designed the Pimax lenses are) but You know nobody’s walking the walk…
It might just be that it’s actually not that easy seeing that Valves lenses aren’t perfect either (god rays) and they’re pretty far from Pimax FOV.
Other high FOV headsets have other downsides is seems.
StarVR is blurry as far as I recall and XTAL is not something I’d be wearing for more than a few minutes it seems (pretty bulky)…
Oh, and the price…?
I’m no expert but it seems to me that we can’t have it all, for now.
The Index is probably the best compromise for now.
If it had less apparent god rays and a higher resolution it would be hard to beat. I think…
Or, if Pimax fixed the eye relief/IPD related issues (different adjustable cowling and maybe other slight adjustments) maybe the X would be the clear winner for those that currently don’t think it is.
I still never use large FOV as it’s way too heavy performance wise and the gains are tiny (for me).
Maybe a 2 x 4K 140° FOV headset with less “aggressive” lenses and an adjustable cowling (eye relief) would do wonders?
Or maybe they should just ship the headsets with a thin velvety foam piece, a thin PU leather foam piece and two “spacers” (thick and thin) and instructions on how to find the right setup for each person…
They could even sell the spacers separately (a 5mm and a 10mm spacer with velcro on both sides).
Eye box is a space area where you can put your eye (pupil), relative to the lens, and still see the correct image. “Sweet spot” for me means the part of the eye box, where the image is the best. This is a kind of analogy to an audiophile definition of the “sweet spot” - i.e. place where the sound sounds the best.
For some reason however, people are using “sweet spot” in the way you did, which confuses me.
I received a 3090 founders recently and swapped my 8kx with my index.
Because of the sweetspot, which is much better on the index. The sde is worse, yes, but not “50%” as others say. What use is an (almost) perfectly clear center, if i have to live with blur all over the place just a bit outside of it. Reading any text in nms across the screen is a burden. You need to move your head to read. With the index i can read with just moving my eyes.
Edit: want to add that i have a 69.5 ipd and have the hmd as close to my eyeballs as possible. Using the index to max the fov and using the 8kx because that way i have less distortion (deep poclet eyes)
Everyone can make the test: activate the performance graph in steamvr and try to read the ms numbers below the graph sharply in the 8kx. Simply not possible. While very well possible in my index.
Pimax really need to improve the lenses. But thats not all. I play nms with the index @120hz with steamvr motion smoothing on. >3000v pixels in steamvrss. Yes, sometimes a little wobbly because of the inserted frames, but the overall clarity is amazing. I have my woojer vest connected via index frunk-usb. And all works all of the times.
I really tried to love my 8kxes, even lately used the das on the 8kx. But the hassle with the connection (sometimes not detected, always a need to wobble in front of the lighthouses), bad tracking (loosing my tracking when coming too close to the sides of the hmd), less comfort (rugged housing plus mas is nice, but much heavier than my 8k or 5k+ with das), and i have marks on my nose (big nose guy here) and the lens -ipd hassle, etc. not to mention the disappointment with the eye tracking (usb 3.0 speed issue) or the hand tracking (way too laggy, maybe also only a 2.0 connection?) brought me back to my old index. Just ordered a confy face foam set even so it’s already confy stock.
. You should definitely be able to find a good view in this configuration (unless point 2) above applies).
.
Maybe just maybe if we get index++ quality lens drop in its possible
