georgewsinger

Agreed that VR gaming is bigger than most people realize, but VR for productivity (to replace PCs & laptops) is still an untapped market. VR pixel density is now within one doubling of "human eye resolution", so it seems like now's the time for VR to be useful for this use case.

Apple's headset is really cool, but is being built over iOS for 2D apps, so isn't realistically going to replace someone's laptop as their primary computing device.[1]

[1] https://news.ycombinator.com/item?id=36326677

An underrated fact is that the Japanese train & metro system is largely privatized: https://www.bloomberg.com/news/articles/2011-10-31/why-tokyo...

> Japan followed a similar path as the West before World War II, but afterwards diverged. As in the West, railroading started out very laissez-faire in the 19th century, but came under state control in the beginning of the 20th. But after World War II, while nearly all railways and intracity buses in Europe and North America were nationalized, Japan stayed its pre-war course, with the railway industry retaining its sizable minority of private firms.

They became further privatized in the 80s, and now there are multiple competing rail companies in Japan. It's my understanding that their business model is to build rails, run near breakeven on the actual trains, and then capitalize on the businesses that spring up around the lines (by purchasing nearby real estate and opening up shopping centers & restaurants).

The Varjo Aero has the same PPD as the Simula One :] So if you are able to daily drive it for productivity, that's a good sign you would be able to do the same with our headset.

The problem with the Varjo Aero is that's it's a cord-tethered headset which needs to be attached to a powerful GPU. It doesn't run Linux Desktop natively (nor even when attached to a host) since it doesn't have Linux support. So in our opinion it doesn't actually replace a laptop, which we're taking as the "holy grail" of VR computing.

Our critique isn't so much about Apple's locked down OS as much as it is "you won't be able to run powerhouse apps from macOS to get your more serious work done".

We hope & expect consumers to love the Apple headset(!), because it seems better/more focused on productivity use cases than the gaming headsets currently on the market. But in order for it to "actually replace a laptop", bringing iPad/iPhone apps into VR isn't quite enough, in our opinion.

I understand your concern, but if we were interested in a quick exit we wouldn't be working on this lol.

We're looking for venture backing because getting a consumer hardware startup off the ground is really expensive!

Japan also ranks 3rd (behind the USA & India, with larger populations) in ChatGPT usage: https://www.demandsage.com/chatgpt-statistics/

There's also been discussion of their government using ChatGPT to reduce red tape: https://www.bloomberg.com/news/articles/2023-04-18/japan-gov...

It's cool to see Japan and Japanese culture taking techno-optimist stances on AI.

Can you expand upon this? Curious to learn more (as an American who has consumed a lot of K-dramas, I often feel like subtleties sometimes fly over my head).

Google's stock is now worth more than it was on the date of ChatGPT's initial release (Nov 30th).

Good test for the Efficient Markets Hypothesis: how does this make any sense?

Arch is just for the review unit. We're not planning on using it for our final release.

Also, the headset is ultimately customizable, so you can use whichever distro you'd like (though will probably want to use our window manager :).

> Considering this comes from Japan...

Can you explain to me why this follows? I don't usually associate "hot air exaggeration" with Japanese culture.

Wolfram Mathematica.

Nice find (and makes sense).

If you're interested in a "turn on your headset, and launch straight into your open-source Linux VR Desktop" workflow (rather than straight into a gaming marketplace/social network), you might check out https://simulavr.com.[1] This is the sort of headset that we're building this year.[2]

[1] In full disclosure, I'm affiliated with this project. :]

[2] https://simulavr.com/blog/2023-timeline/

Though we finished our first Simula One "Review Unit" back in September (this being a stripped down prototype which demonstrates display, optics, and main board functionality), we haven't released a prototype to testers yet due to it being too unpolished. You can see some videos of it in action however here[1].

Here are some things on the horizon for us to finish before we release test units to YouTubers:

- Add an IPD adjustment mechanism[2]. Currently, our prototype's IPD is locked in place until we add this functionality, which makes it really poor for testing since there's lots of variation in IPD.

- Incorporate our front-facing cameras for our AR Mode. On this front, we recently shipped our Camera Boards to PCB fab[3]. Note that our cameras support 2448x2048 per eye resolution, and are placed more directly over the eyes than the Quest Pro (avoiding the issues discussed in the parent post). For a fuller comparison against the Quest Pro see[4].

- Replace our FPGA eval board with our own custom FPGA. (Just started on this; will likely discuss it during our next blog update).

- Integrate our detachable compute pack to the back of the headset (right now our compute pack is just dangling).

So all in all, while much of the core functionality is there, our headset is still too unpolished to send to YouTubers.

Also: I understand your skepticism RE preordering. We try to keep a consistent stream of blog updates to at least let people know what's going on engineering wise.

[1] https://simulavr.com/blog/first-glimpse-of-review-units/

[2] https://simulavr.com/blog/simula-one-auto-ipd-adjustment/

[3] https://simulavr.com/blog/camera-boards-submitted/

[4] https://simulavr.com/blog/simula-one-vs-quest-pro/

It runs Linux using our custom VR Linux distro. The core of this distro is our VR window manager (https://github.com/SimulaVR/Simula). It comes with 16GB-32GB of RAM and an Intel 12th Gen i7 processor. RE URLs:

  - Hacker News discussion[1]
  - More detailed spec overview [2]

[1] https://hn.algolia.com/?q=simulavr.com [2] https://simulavr.com/blog/why-is-the-simula-one-so-expensive...

You could also completely replace your Linux Laptop with a Simula One.[1] :]

[1] https://simulavr.com

This definitely sounds like something Alan Kay would recommend :)

SimulaVR (https://simulavr.com) is offering a $100 discount to Hacker News preorderers: use the code "DISCOUNT_HN" at checkout.

This has been a feature built into our design from the start. We just hadn't implemented it yet.

VR pixel density is arguably on an exponential curve. Most headsets are within one tripling of "human eye resolution".[1] A few are within one doubling.

[1] https://i.imgur.com/Ml1dv3w.png

This is why VR pixel density (PPD) is so important for VR computing. It's the #1 thing that impacts how legible text and other fine details are when working. Here's a list of headsets for comparison:

- Valve Index: 11.07 PPD

- Quest II: 20.58 PPD

- Quest Pro: 22.6 PPD

(In full disclaimer I'm affiliated with SimulaVR, and we're working on a headset with 35.5 PPD.)

The Nreal Air looks cool, but it only offers a 46 degree FoV[1]. That's less than half the FOV of the Quest 2, which makes it hard to see all of your windows/workspace. (To be fair: the tradeoff is excellent PPD within its narrow FOV).

[1] https://www.nreal.ai/specs/

The Quest Pro is a great headset in many ways, but it only offers 10% higher pixel density than the Quest 2 (which was already pretty low).

For comparison: it offers 22.69 PPD. The human eye can distinguish up to 60 PPD.

Thanks for your support.

RE resolution being low on the Valve Index: we absolutely agree :] The Simula One offers ~3x higher pixel density than the Valve Index, and 50% higher density than the Quest Pro. This dramatically impacts the extent to which you have to turn your head to look at things.

The Quest Pro has a better design than the Quest 2 (as well as many other benefits), but in terms of pixel density it only offers 10% higher PPD than the Quest 2 (which was already pretty low).[1]

(Disclaimer: I'm affiliated with Simula).

[1] https://news.ycombinator.com/item?id=28691358

We're still in preorder phase, aiming to ship most headsets in ~Q2 of 2023.

This is our first foray into hardware, but we already developed a stable Linux VR Desktop (https://github.com/SimulaVR/Simula) from 2017-2021.

We try to provide frequent updates on our hardware engineering progress. For example, here's a recent post showing some videos of our review units: https://simulavr.com/blog/first-glimpse-of-review-units/ If you're interested in community discussion, you might also check out https://hn.algolia.com/?q=simulavr.com.

Yep, you can try out our software at https://github.com/SimulaVR/Simula on the HTC Vive and Valve Index platforms.

If you don't have one of those headsets, there are some videos of people (whom we have no affiliation with) on YouTube testing it out: https://youtu.be/8gVLF8SnK84?t=424 Here's also (a pretty old) video of me hacking on Simula, in Simula: https://youtu.be/FWLuwG91HnI.

It's also true that we're more focused on hardware right now. We plan on reverting more towards software development in 2023. For more details: https://github.com/SimulaVR/Simula/issues/180#issuecomment-1...

Some things we have planned to implement include VR window tiling and some other UI stuff.

The Simula One (https://simulavr.com) was designed specifically for this use case (disclaimer: I'm a cofounder of the project). TLDR: it's a 100% office dedicated headset with bleeding edge pixel density that runs Linux Desktop natively. Everything from the hardware to the software rendering is optimized for VR Desktop and the clear display of text. We really pushed hard to make the Simula One specs as premium/bleeding edge as humanly possible. For example, our headset has more than 50% higher pixel density than the Quest Pro (35.5 PPD vs. 22.69 PPD), and our compute specs are comparable to a premium office laptop (Intel 12th Gen i7-1265U Processor, up to 32GB of RAM, and 1TB of storage).

For more discussion, here's a list of HN threads on the Simula One: https://hn.algolia.com/?q=simulavr.com. The most recent one discusses a comparison between the Simula One and the Quest Pro (https://news.ycombinator.com/item?id=33318956).

Pros of the Simula One: it offers native VR computing (i.e. you put on the headset, turn it on, and get see an uncapped number of your Linux Desktop apps running in front of you). Compare this to the Questo Pro, which for VR Desktop provides WiFi streamed screens from your laptop (placing latency and bandwidth constraints on what you can do, and capping the number of screens you can use to 5).

Cons of the Simula One: it's still in preorder phase, and has a target ship date of ~Q2 2023. Our small scale makes the cost of the headset pretty high (though to be fair we're putting very premium specs into it).

But we've also been openly declaring these claims to not have any teeth, for almost a year now: https://simulavr.com/blog/we-dont-want-the-metaverse/

TLDR: That post is from 2021, and we basically said that Meta's product vision is to start with a social network/gaming platform first, and then backwards engineer productivity into it as a third use case; whereas we want to start with productive VR computing as our first and only focus, and engineer everything we do around that.

Yes, thank you for catching. We just pushed the fix.

The original table was copied over from an old post (when we were still using 11th Gen Intel), and I somehow missed updating it.