Essentially, this product seems to actually be on the way out. It's $950. I wonder why they didn't just go $999. 

Allegedly, there's a Dyson engineer in the comments talking about the weight and how it fell off:

That movement sounds like leaning forwards (for momentum) and then standing up quickly. I should take note to keep the headset secure even without the bicyle helmet strap.

Last log, I expected these to come a week ago, but then the package decided to sit in a warehouse for a week.

It's pretty cool. The flat side closer to my eye is the one that allows for a full FOV when the subject and my eye are close to the lens.

I've got no idea why the camera captured all this glare, but I didn't really see it when I looked though. It also looked a tad sharper. I certainly got to see the "screendoor effect" first hand though.

Looking at a monochrome grid such as: I could see that the edges were blurrier than the centre.

What's cool is that none of the rings are visible. It just acts as one flat magnifying glass. I even opened the camera app to see what VR passthrough these days might look like, though my setup was essentially Google Cardboard without the card.

The weight of the lens is 5.8g.

Sweet. Today, the glass elements for the test rig (lenses and 90 degree prizm) arrived today, which I wasn't expecting because they were the last things to dispatch, according to AliExpress. No idea how they jumped the queue but I'll take it. Looking at tracking, I predict to have everything else this week, so I probably should get started on designing the test rig.

I wasn't expecting the lens to have a way deeper curve on one side than the other. Good thing I didn't assume they'd be even on both sides and tried a computational simulation.

For (checks notes) £2 worth of lens, the first one inspected looks really nice. I'm glad I haven't yet needed to attempt to 3D print something custom, because I doubt even a theoretically perfect #SecSavr Suspense [gd0105] would get this visual clarity.

[some time later]

Checked all the glass and it all looks divine.

The prism has a mirror-coated face but it doesn't extend all the way to the edges so this would only be usable for moving the beam around. 

The total internal reflections look cool though, and it indeed does the dove-prism reflection thing if looked at from the right angle. I feel like I'm not going to get a perfect edge-to-edge merge using plain prisms, but I do think I can get close. Even Samsung shipped out a phone with a crease line, so I've got to give my hobbyist run some slack.

Lastly, the F-40 lens (deep/shallow curved face)

Assuming the packaging is massless, the lenses weigh 140.9g. 92.6g is the 3 F-30's and 48.2g is the 2 F-40's. Average lens weight is 28.18g. Prism is 10.7g, so I'd assume a beamsplitter is 2x that at 21.4g and 10mm prisms are 1/8th at 1.3g. Thus, in glassware alone, I'm estimating a weight of 496.2g. That doesn't sound great. Google suggests that the average motorcycle helmet is 1.5kg, so perhaps I could still obtain a comfortable solution for prototypes.

Chromatic abberation of the lenses looks minimal, but the entire view is minimal so it might be more apparent in testing.

Whilst impatiently waiting for the optical gear to arrive from AliExpress, I did a bit of looking around on the internet for things related to Mixed Reality. It's not strictly TyMist related, but with all the different systems I'd need to design and test, I don't really want to find out I could've just taken it easy for a month or two and a company to come out with the next generation in MR technology. I'd still accept it, since I'd rather not invent everything here -- it sounds exciting until you have to spend hours grinding in some software because some things you think should've been a 30 second action is actually 600.

Pixel Counts

For starters, I wanted to see how many pixels I'd actually need to fill the entire area of vision reachable by the centre of my eye. It seems that a 45 degree cone was a good rough estimate. The pancake lenses for the 2560 OLED is also 90 degrees, so it sounds like a sensible number. 

The Nvidia ppd calculator says that an 8192x8192px at 90ppd center would give me an FOV of 91 degrees, and the reason why I chose that number is because, if the image merging idea of TyMist actually works, it could be possible to combine 4 4096px displays. I also looked into 8640x7680, which is 2 portrait 8K HD displays side to side. That would give a vertical FOV of 85 and horizontal of 96. The edge PPD is 118, so there should be enough to get a decent amount of peripheral vision of the eye covered too. But still... 16K. That's what the equivalent of 4 8K displays is, counting the pixels. Obviously, foveated rendering needs to be a thing that exists in the future.

With the screen panels of today, 5760px square is likely the best doable using the quad 2880 MiniLED panels. This theoretical headset would have a peak brightness of 760 nits, but won't be cheap since each panel was £150 each on Alibaba. 

I found though-the-lens videos of the Pimax Crystal, which uses 2 of these panels:

Since I don't have a VR headset, it's nice to see what it looks like. Considering the difference to the Pimax 8K, all I can think of is "Wow. People can play with all this blur? Wow, the human mind is impressive." 

I also found this image:

The ppd calc says that 5760 equates to 60ppd, which seems to match up with doubling and averaging the Crystal's PPD numbers.

Next, I'd like to say that I'm very excited for Porotech's new... tech, DPT.

Basically, say goodbye to individual red, green and blue subpixels. I've always wondered what would happen if someone had a 4th retina cone evenly spaced between green and blue, and how the world around them would look new and even more vibrant, but all screens would have a bright but unnatural colour (imagine if the rest of the world only saw red and blue. White to them would be magenta to you with your RGB cones). Well Me In That Imaginary World doesn't have to worry any longer. This single pixel would be able to ouput near-UV blues, real cyans, authentic yellows and White(tm). It might even be usable for 3D printing since there has been research into resins that react differently based on exposure to different wavelengths of light. Another benefit is that this tech could double the resolution of a screen since now a subpixel can be an actual pixel; the 4K displays of today can become the 8K displays of tomorrow, making a 4K Nreal Air a reality.

Future Headsets on the Community Wishlist

Whilst scanning, I noticed that PSVR and Apple's forever-rumoured headset usually come up. 

For Playstation conversations, some are hoping that someone out there cooks up a port for PCVR. 

For Apple, it seems that the conversations are more along the lines of what the VR landscape will be like after they announce and start selling their product. It seems like they've got high expectations of Apple to unveil something revolutionary. Even I got on that Dynamic Island hype train, so I can see where those expectations are coming from. Additionally, it's expected that AR is where the real mass-adoption is going to be.

There's also the Pimax 12K, which people are excited for because "it puts it all in one box, and nobody else is doing that".

Concept Looks

This was a design idea I've pocketted, along with these headphones:

I can't think of any design solutions at the moment, but I'd like to keep the concepts in mind.

Below is the 3 step process I did:

Now I think that the design is decent enough to move on, since it essentially needs to fit a pillow, air purifier, bike helmet and audio headset in there, even if the TyMist part fails. I'm just putting Solyndr and TyMist stuff in the same project for now.

I'm starting to see a resemblance to those thick, cold climate hats. Since nobody wears much-anything other than glasses, hats and audio equipment, any solution practically possible is going to be noticed by anyone even half atentive in a 50 metre radius. I think the best solution is to aim for a "Wow. That's a MASSIVE hat/helmet. Never seen that before. Anyway..." thought process in the public eye. Hopefuly I'd be wearing #TEOSS [gd0037] + #TetInventory [gd0039] too and I'd just be considered another food delivery service person. Maybe with some reflective stickers, it'll look more like a (road) safety device and less of a fashion statement; I think people are more tolerant if it looks like something is being worn for work/safety.

I think this concept is at a point where any aesthetic changes would be texture based:

The textures actually make it look smaller and more like a bike helmet I'd find hung up on a Halfords or Wilko rack. Texturing is the last of my worries though, and since I've got a lot of White PETG and Transparent White PLA, all projects are still likely to be white.

My university just happened to have a Microsoft Hololens 2 that I was able to try. The guys there said that it was in the region of £2k, and I'd assume that's after an education discount.

Thoughts:

1. As with most Microsoft hardware products I know exist, Hololens looked nicely designed.
2. I like the rear dial to tighten over the head, and the optics can be swivelled out of the way when putting it on/taking it off.
3. The first thing I noticed was the light non-uniformity. I think it was on this website but I remember seeing an image that showed how the brightness noticably decreases closer to the edges of the FOV.
4. The FOV was better than expected. I really think it's because there's less light on the edges, so the projected image more guadually comes into view instead of a sharp boundary.
5. The 3d... didn't seem that 3D. Like there was depth but it was more like a 3DS. I assume that this was the thing that VR users talk about where everything, no matter if it's close or far, is at the same focus. 
   1. I quickly compared the view that my left and right eyes were seeing, and the difference was noticable but small. Perhaps I was supposed to set IPD or something to that effect, but I think dual cloned images for TyMist is sill better than 1 eye only.
6. I initially noticed how choppy the hand tracking was, but not how fluid the air-drawn text in the background was. I don't plan on any hand tracking so I'm more interested in the FPS needed to achieve the latter. Googling seems like 60fps, which is suprising.
7. The experience was cool, but not sharp. I just searched and the Hololens 2 has a PPD of 47.
8. The brightness of the world is higher than the 2 way mirror.

Now, because I've got a project called #TEOSS [gd0037] that will see me walking at elevated speeds, Moonwalkers inventor (a walk assist solution like TEOSS that's currently on Kickstarter) started to mine for a solution after almost colliding with a car on is scooter, and because this is a helmet which would be hard to transport in #TetInventory [gd0039], I'd very much like to know about and integrate safety features into TyMist. Just because certification is unlikely doesn't mean I shouldn't try and make projects safer. 

Initial expectations:

• High visibility
• Helmet needs to be secured by an under-chin strap when in transit to prevent it from falling off
• Shocks need to be safely absorbed.
• Helmet needs to be able to slide.

Initial mental simulation findings:

• High helmet diameter may result in higher moment forces on neck.
• A front crash may slide the visor out, causing it to only be supported on one side.
• The discontinuous edge between the cylinder and flat top face may be caught on rough terrain instead of allowing the helmet to slide across.
• The floor or collision object will initially be an impact, but it is most likely that I'd have enough momentum to slide across the floor afterwards.

Search map:

• Google: 3D printed helmet
  • https://all3dp.com/2/incredible-prints-can-you-3d-print-a-bicycle-helmet/
    • https://www.bicycleroots.com/blog/mips-helmet-technology-what-it-is-and-why-you-need-it/
    • Seems I'm thinking along the right lines with the 3 mentioned bullet points of CPSC.
    • Mental simulation result from implementing the understanding of MIPS:
      • "Wow. It works."
      • Expected moments on the head and neck have noticably decreased.
        • The fact that TyMist is a cylinder may already help in this regard.
      • A filleted chamfer on the top edge is expected to increase helmets ability to slide across the floor substantially.

      • The MIPS system may also reduce the forces felt on the head when rotating during normal virtual monitor use, and could even be a mechanical way of keeping the monitor view looking semi-stationary when turning to the next virtual monitor.
    • https://www.kupol.ca
      • This is from the same person that makes long and sharp FFF nozzles for non-planar printing: https://www.nonplanar.xyz
    • https://www.designboom.com/design/smart-bike-helmet-3d-printed-nylon-shell-high-tech-nfrontier-08-14-2022/
    • https://www.youtube.com/watch?v=YdiegagK0ek
      • https://nexa3d.com/3d-printers/nxe-200pro/
        • Sounds like they get speed gains from a flexible vat. Perhaps the #SecSavr Suspense [gd0105] could be faster than expected.
    • Simulation warning: Impact with nose on combiner
      • This combiner was already throwing a solution warning because of the fact that it's dark all the time. 
        • In an infinite possibilities world, the combiner would actually be a probably-non-existant pane that can be PWM driven between very reflective and very transparent, or some other way to electrically obtain different reflectance values.
          • Me, approx 180 seconds in the future: https://pubs.rsc.org/en/content/articlepdf/2015/sc/c4sc01912a looks to be the kind of thing I want.
          • Me, approx 8 minutes in the future: [spits cereal]

          • Source: kentoptronics.com
          • The future is wow thanks to science! Liquid crystals can do so many things apparently.
      • I'm going to need to figure out how to move the combiner out of the way when not in use.
        • I might need to obtain a different solution if I ever install a PC into TyMist, but currently, there'd be nothing to see without #Teti [gd0022].
        • Perhaps the simplest approach will be to remove it from TyMist and hold the combiner inside a bag.
• Google: bike helmet design guidelines 
  • https://helmets.org/idealstd.htm
    • This sounds like a good set of standards

I punched the numbers into the calculator and this should achieve a PPD or 94.5. A PPD of 90 would be a 32 degree cone. If 52mm is larger than this cone, the PPD is less than 90, and higher than 90 if smaller. I've angled it so that the image starts at the precise angle that the user can see past the top of the helmet, which could allow for some cool "TyMist On" visual animations.

I think this concept looks extremely clean and fresh now. Kind of like Blanc:

Wait a second. This reminds me of Kerwhizz, one of my favourite shows over a decade ago:

I can't find a 350mm OD transparent tube, and I can only imagine that a wood jig to bend a pane into the half cylinder I'd need would need a lot of wood. I'll try and compute a solution later on. For now, I could have a black window for a cinema blackout. Alternatively, I could print a mesh or grid to adhere the film onto. 

Anyway, I've reduced the extra amount of helmet under the door area (the part with the steep chamfer) to increase head rotational freedom. I've also removed the eagle-like front section, so it's now just a cylinder all the way.

If I then add some soft materials, I think it's going to look quite comfortable. I'd still like to avoid a nose cutout on the combiner though. It increases manufacturing complexity, and most likely would still be an issue when I'm lying to the side and the helmet gets pushed to the left/right.

I slept on it and yeah I need that Thunderbird 5 solution: 

I belive, however, that as a user inside the helmet and not looking at it, 350mm will feel much more spacious.

The front window is 100mm, which looks to be enough to cover the nose and hopefully enough to provide fresh breathing air.

I'm probably keeping the TyMist name. Or perhaps TyCyMistify.

I should've mentioned this earlier, but the LED ring is like the RGB one on Alexa powered devices. The idea is to get the PLDC to reflect some of that light for a Phillips HUE like background effect.