Somnium VR-1 Q3 2023

I’m not sure I understand your point.

But I do wonder if Pimax resolved the issues with the 8kX fresnels if they could have made a v2 and kept the price low. How different would the VR landscape look if they had a £800 160FOV headset rather than completely dump it and go backwards for a Varjo competitor. Wide FOV has no competitor it just needed to be done right. The 8kX was very close.

2 Likes

DK1 era headsets had conventional lenses (…which happened to be aspheric), and among their weaknesses, was significant field curvature (low edge-to-edge clarity, due to the focus of the lens “peeling away” from the flat screen, away from their centers), and -pupil swim (the view distorting when the pupil moves within the eyebox, due to looking around). At least the latter was directly mitigated by using thinner lenses - these happened to be fresnels, which could conceivably also easer be tuned per-segment, to alleviate the former; And now when we’re going back back to conventional, we claim the exact same resulting gains going from B-to-A, as we did from A-to-B; Maybe we have discovered a form of perpetual motion device analog. :stuck_out_tongue:

3 Likes

I’m not sure why Fresnel couldn’t have been advanced more. It’s flat, it’s cheap and it can go wide. At least the 12k still uses fresnel on the periphery.

Going wide with glass is a nightmare when even 110FoZ gives wild CA.

I imagine one could “paint” the back “wall” of each ring segment in a fresnel lens black, to significantly reduce the annoying light scatter, at the likely cost of thin, but visible black crescents… Question is what mass-production-viable process there might be to do this, whilst leaving the lens fronts untouched.
I believe Sony filed a patent application for a similar, easy to apply, but probably nowhere near as good, solution, where they would print black to the top rims of the ridges…

I think any “simple” optics would preferrably need to be concave on the user side, so that it wraps the eyeball - otherwise we get a fair bit of that other fresnel effect, where you transition from see-through, to total reflection and strong polarisation, with increasingly oblique angles of attack.
For this reason, I think one of the pluses the optics in the Index has, is that going dual element lets them make each lens plano-convex, eliminating the “hybrid lens” convex bulge on the user side. (EDIT: Ever seen the Index logo? A circle next to two semi-circles – an eyeball, next to a stack of two plano-convex lenses. :9)

Somnium says they are experimenting with an alternative lens train, that eliminates chromatic abberrations optically, but expensively…
I still imagine with sufficiently good eyetracking, dynamic software distortion compensation could do the trick.

Argueably, the 12k is the 8k evolved, and I suppose they could offer it in a variety of configurations, down to a simple 2x4k variant, without any of the fancy stuff onboard - just what it takes to drive the panels, and track the HMD…

They just got sidetracked with the Crystal, whether as a stopgap solution, to keep themselves going through 12k delays; Due to creative distraction; And/or bucking to persistent demand from this very forum. :stuck_out_tongue:

I have to mention that 12k marketing renders show a sandwich of two lenses per eye, like the Index, where the inner one is still fresnel all the way to the center, there right behind the conventional portion of outer one…

( I happen to have here a simple… contraption, where I for a bit of a lark duct-taped two pairs of cheap, spherical, plano-convex, 73mm glass lenses to a pair of huuuge cheap display panels, and frankly my vision in it does not strike me as all that bad – quite optically clear across the view, and not extremely distorted – an estimated 180-200-ish FOV, more stereo overlap than I’ve experienced in any real headset… Kind of wish I had the chops to write a driver for it, so that I could see what it would look like with a proper VR view… :P)

4 Likes
1 Like

Such a shame. Either they delay the launch and fix the optics or go ahead and ignore the problems and hope the VR-1 doesn’t generate a reputation that puts potential buyers off. Good for me I’m not a potential buyer.

I’m not sure why these companies get YTers in at the final stages, only to find out potential problems. If youre going to get YTers in do it at the beginning!

I’m afraid egos get in the way of making a good VR product.

3 Likes

yeah really a shame, after all the hype … Certainly reminds me of Pimax haha, usually a lot of hype too and then disappointment. I really don’t believe they can fix the problems easily. Really too bad.

3 Likes

Me either. I think its too late. But the price is the biggest problem. Stuck bewteen the Aero, Crystal and Xr-4, all doing things better or cheaper even before we get to the 12k.

I bet Art is sweating buckets right now. He must of sunk millions at this point.

I think Somnium and VRengineers will merge in 2024 and they’ll just work on the Hypervision HMD.

3 Likes

Well let’s not forget HTC with Vp1, Cosmos, VP2. Oculus Go. Decagear, SegaVR and Virtual Boy

But yeah not really surprising. VR-1 was likely to have a lot of issues as it has been advertised for a very ling time without any real demos.

1 Like

Honestly none of those really look that interesting. I wish one of the big boys would make a high-end unit. Valve, Meta, imagine what they could do. Although SUPPOSEDLY the next Pico headset is going to be high-end.

It seems only the big boys know how to get things right. In that sense it’s just awesome that Apple has stepped in. Even though their headset won’t be very interesting for gaming, I’m 100% sure that they’ll set the new standard for high-end visual fidelity and that others will follow suit.

This VR1 failure really makes me think that good VR might just be too complicated to make for smaller companies. If you look historically at the headsets that had good lenses, they really only came from the big boys. Even the smaller ‘high end’ companies like Varjo and XTAL have had their own problems with visual fidelity. And now it seems Somnium is going to join them.

2 Likes
1 Like

Somnium Vr 1 will be the best hmd humanity created :slight_smile: Its the way to go 100%

Not looking good with several critical issues. Time will tell…

It’s a 1.3k HMD tops and 1.8k for the top model.

Somnium will be in the position where they’ve spent more on development than they’ll get back on sales.

2 Likes

Hardware is a necessary evil best left to others (who are prepared to burn their money)

I left hardware development after being threatened over unintended patent infringement, leaving me facing huge legal costs which would have bankrupted me.

Most small hardware companies seem to be surviving on rounds of investment funding

3 Likes

Sorry to hear about your past experience.

VR hardware companies are literally worthless while sales volumes are so low. Looking at the Steam HW survey, things haven’t improved over the last couple of years which mean the pie in the same size but the slices are getting smaller as more manufacturers join in.

How long before Valve deliver? Index will be 5 years old next year lol. That’s the headset to rule them all.

1 Like
2 Likes

Somnium sound more and more like Pimax every day.

“Why did we show the lenses without coating? Why not?”

“Coating brings the last 5% which actually improves 100% of the visuals.”

Oh Jesus.

1 Like

Here’s an idea.

Somnium buy an Index for £459.
Open it up slide in some 2880x2880 panels and two aspheric lenses.
Close it up and sell it as a Crystal beater for £1299.

4 Likes

Keep in mind though the “Big Boys” are all playing safe. While the Index has great Audio they didn’t improve res or FoV. Which is ashame. Even doing a minor refresh with new panels and improved optics to achieve a legit 120 to 140 wide would be great.

2 Likes