kanetw

21 ms·

211.
by kanetw 5y ago
Electrical engineering. Getting a working PCB design for a modern CPU, especially in tight space constraints, is quite challenging. It's probably something we will do in the v2 (the NUC solution has its drawbacks in terms of exposed IO
212.
by kanetw 5y ago
Really simple reason: Intel talked to us, AMD went "lol here's a bunch of embedded boards, go fuck yourself." Intel also has the NUC compute element solution which significantly simplifies the EE and allows for a seamless u
213.
by kanetw 5y ago
Worth to note is the display is only 70Hz at 2560p. The "95% light path amplification" for their pancake lenses is also either misleading or an impossibly high transmission figure for pancake optics. Maybe I just misinterpreted th
214.
by kanetw 5y ago
Stuff like weight/dimensions are pointless until we have the integration finished, and we don't yet. We're approaching the limit of what we can do with a 3 person team, and we need to place orders for our components so the in
215.
by kanetw 5y ago
We're looking to mount the compute unit at the back of the head, which gives both more space for heat dissipation and allows for a balanced weight.
216.
by kanetw 5y ago
They actually run on the headset. It's a full laptop equivalent.
217.
by kanetw 5y ago
We're actually focusing primarily on the hardware right now. The website is very much a placeholder, but it does provide a rough ballpark. It's a challenging task, but by focusing on productivity applications and a higher price po
218.
by kanetw 5y ago
We're trying to hit as wide of an IPD as feasible. Right now with the BOE displays we can hit an IPD of 55-77mm, but the design isn't finalized yet. Lens glare, too early in the design to tell yet. I'll have more info in late
219.
by kanetw 5y ago
I actually tried to get us to rename but we never managed before we had to incorporate and, yeah, we're a bit stuck now. When I was interviewing for a programming language theory PhD program in... Bristol I think, back when this was pu
220.
by kanetw 5y ago
We will be, once we've sorted out all the blocker issues and our prototype is complete.
221.
by kanetw 5y ago
Compute module based on NUC will be included, and is pluggable on the back of the headset. About 2-2.5k predicted price point. Prescription lens we'll figure something out. We're trying to keep enough eye relief to support glasses
222.
by kanetw 5y ago
You get position/velocity/angle data directly, and it's less power hungry than running high-res cameras specifically. Also some research papers show an increased tracking accuracy with mmwave+IMU than RGB+IMU. So less process
223.
by kanetw 5y ago
both me and george are avid wolfram users. incredibly good stuff
224.
by kanetw 5y ago
I've been meaning to use a proper service but it works right now and once we're actually sending I'll import the emails and do an opt-in and such.
225.
by kanetw 5y ago
Today I'd definitely use Rust. Most of the rendering work is gone in Godot, which is C++. GC hasn't been an noticeable issue as a result.
226.
by kanetw 5y ago
Tentatively 2k.
227.
by kanetw 5y ago
We're planning for 2k. Due to our low volumes etc going hard on optimizing the unit cost is difficult.
228.
by kanetw 5y ago
Appreciated. FPGAs are something I've been aware of for a long while now but haven't used before, so recs are always good.
229.
by kanetw 5y ago
Good question. That's an open research area for this particular use case. Some research about flight helmets/night vision goggles indicate that a counterweight alleviates neck strain. But it doesn't have the specific up/
230.
by kanetw 5y ago
The absolute killer app is compilers, static analyzers, and similar. Other languages don't even come close for anything in that space. However, I think that purely functional languages are useful for basically everything. Haskell is my
231.
by kanetw 5y ago
There's a lot of research papers that I found, but nothing hardware generic unfortunately. Hand tracking is a difficult beast especially, and we would like to just use the new Ultraleap module for that, but they don't support Linu
232.
by kanetw 5y ago
We're evaluating 2880x2880 displays right now. With our upcoming optics, we can get up to 45PPD in the foveal area if everything works out as planned (they're basically a variable magnification optic). Depending on vendor support
233.
by kanetw 5y ago
We plan for at least 100deg DFOV.
234.
by kanetw 5y ago
We're making our own headset. Nothing off-the-shelf was available or satisfactory. Most of the recent Github commits are in the dev branch, but right now it's 90% George working on the code. I'm busy on the hardware and wrang
235.
by kanetw 5y ago
Nothing that's satisfactory in one way or another. Probably Luxonis DepthAI? The main problem with off-the-shelf solutions is that they add another set of cameras, and afaik nothing exists that allows custom cameras. We're gonna n
236.
by kanetw 5y ago
Another founder here, with some more comments on the tech side of things: 1. The software is relatively usable, and you can try it out right now on https://github.com/SimulaVR/Simula 2. The hardware is still being work
237.
by kanetw 7y ago
We don't actually use X as the framework. The underlying code is Wayland, and then Xwayland as a compatibility layer. It's just that most apps we test against are X11.
238.
by kanetw 7y ago
I use a curved monitor (Samsung CRG9) myself as my daily driver, and it appears curved when I use desktop view in SteamVR, so I definitely see the appeal. Just not sure how nice it looks on smaller windows.
239.
by kanetw 7y ago
I end up needing a lot of screen real estate whenever I work on some EE project --- datasheets, schematic, PCB editor, calculator, notes. Additionally, being able to e.g. manipulate the enclosure and PCB to do a test fit and immediately edi
240.
by kanetw 7y ago
IIRC the Quest is too different, but our medium to long-term goal is to design a portable computer in HMD format, if that's feasible. I'll take a look at Quest docs and see if it's doable, but I don't have high hopes for

More ›