Quark VR Working On Wireless HTC Vive Prototype

[Megalith]

Dealing with a tethered device is probably one of the biggest gripes with VR today, but this company is looking to change that. One can only wonder how they will go about solving this with what sounds like traditional Wi-Fi, however.

So here’s the thing — they’re not working on a wireless headset as much as they’re creating some sort of Wi-Fi transmitter dongle that connects to the headset. This leaves you free to roam the room untethered from your PC. You can stick the device in your pocket, where it’ll send signals between your computer and the Vive. That sounds all well and good, but what’s the latency like? Quark VR admit latency is “a huge challenge” in designing their little wireless gadget but they expect they’ll be able to show a successful prototype this autumn. No word yet on quite what they’ll do with it after that, but they do intend to open source the design.

 

[Jim Kim]

Tetherless will be a VR gamechanger. Patiently waiting for gen 3.

 

[MikeTrike]

Day 2: The Vive is good as is, though a good wireless module will just make it even better.

I'm excited to see what's going on 5+ years from now. :-o

 

[Derangel]

Wireless VR is a pipe dream at the moment. The prototype will be interesting to see, if it's ever shown off to the public, but it will be years before it's viable in retail products. Not only is transmitting the data without any interruption an issue there is a massive issue of latency. No current wireless tech can handle it. So unless they develop something magical we'll be waiting on wireless tech to catch up to everyone's dreams.

 

[JDanser]

I can see it being doable over 802.11ad. It can push 4-10gbps in a room-scale network, and 60ghz WiGig latencies are DAMNED close to the same as wired.

 

[Derangel]

I can see it being doable over 802.11ad. It can push 4-10gbps in a room-scale network, and 60ghz WiGig latencies are DAMNED close to the same as wired.

The math I've seen pegs the required bandwidth at around 5.6Gbps. So ideally you'd probably want, at least, 6-7Gbps constantly in order to ensure nothing gets dropped.

Edit: And just to note; The 5.6Gbps number is just for the current Vive. Higher resolutions will significantly increase the required bandwidth.

 

[Spidey329]

Wireless VR is a pipe dream at the moment. The prototype will be interesting to see, if it's ever shown off to the public, but it will be years before it's viable in retail products. Not only is transmitting the data without any interruption an issue there is a massive issue of latency. No current wireless tech can handle it. So unless they develop something magical we'll be waiting on wireless tech to catch up to everyone's dreams.

I agree. Although there are some wireless transmitting devices that state (in their spec sheets) they're capable of less than 2ms latency. One is the Connex devices. They aren't traditional wifi though, it uses a MiMo setup and acts in wifi broadcast mode, gives it much higher bandwidth.

That latency most likely will go up with the much higher resolution. On top of that, you have the controls as well.

It's doable, it's just going to need a lot of funding behind it.

 

[Derangel]

I agree. Although there are some wireless transmitting devices that state (in their spec sheets) they're capable of less than 2ms latency. One is the Connex devices. They aren't traditional wifi though, it uses a MiMo setup and acts in wifi broadcast mode, gives it much higher bandwidth.

That latency most likely will go up with the much higher resolution. On top of that, you have the controls as well.

It's doable, it's just going to need a lot of funding behind it.

Mhm. Making the hardware open source once they've got stuff developed is a great way to get other companies on board and working on ways to make it viable. I wouldn't be surprised if within 6-8 months of the hardware going open source there are a handful of companies showing off modified prototypes using various different wireless solutions.

 

[JDanser]

The math I've seen pegs the required bandwidth at around 5.6Gbps. So ideally you'd probably want, at least, 6-7Gbps constantly in order to ensure nothing gets dropped.

Edit: And just to note; The 5.6Gbps number is just for the current Vive. Higher resolutions will significantly increase the required bandwidth.

IIRC the 802.11ad spec specifies a theoretical maximum of 7gbps PER CHANNEL, so a 16x16 MIMO array should do the job pretty nicely. I've heard talk of being able to shoot PCIe over a 32x32 array.

 

[Nenu]

The math I've seen pegs the required bandwidth at around 5.6Gbps. So ideally you'd probably want, at least, 6-7Gbps constantly in order to ensure nothing gets dropped.

Edit: And just to note; The 5.6Gbps number is just for the current Vive. Higher resolutions will significantly increase the required bandwidth.

Yep.
It will need to handle dropouts as well, whether due to bad reception or interference.
The current interpolation does 1 frame at a time, that probably wont be enough.
Multiple sequential lost frames could be its undoing anyway due to loss of continuity.

This will be heavily dependent on spare computing power and pre-emption algorithms, more so as resolution rises.

Multiple transmission streams could be helpful reducing lost frames to a max of 1 sequential.

 

[Uvaman2]

Wouldn't it still need a power cable though?
I think making it a single wire, composed of power and fiber optic fiber (1 or 2 filaments ?) would be a better choice.. one fairly thin wire, might be easier to deal with I assume.

 

[lostinseganet]

IIRC the 802.11ad spec specifies a theoretical maximum of 7gbps PER CHANNEL, so a 16x16 MIMO array should do the job pretty nicely. I've heard talk of being able to shoot PCIe over a 32x32 array.

In parallel like a gpu. They should be open to this. how much power would it take? Would we need a tesla battery strapped to our backs or thighs? Also wasn't there a technology that could send information server to server using the ceiling? It was to prevent bottlenecks during high usage for Internet servers? Maybe that could be used in some way? Kinda starting to worry about ow much energy would need to go through the skull...cancer and all that.

 

[Nenu]

Wouldn't it still need a power cable though?
I think making it a single wire, composed of power and fiber optic fiber (1 or 2 filaments ?) would be a better choice.. one fairly thin wire, might be easier to deal with I assume.

Battery.

 

[LurkerLito]

I think they should make a haptic feedback vest for better immersion when things hit you from outside your field of view that doubles as a battery for wireless. That way you can increase the immersion at the same time have a way to use a large battery array that can last longer than something you'd say strap on to the head gear. Definite safety concerns though especially with the recent batteries catching fire with damage from say a fall.

 

[Ducman69]

Tetherless will be a VR gamechanger. Patiently waiting for gen 3.

The Gear VR is already wireless.

To reduce latency though, I'm sure you'd want your desktop to be wired to the router, with the router sitting on the desk @ 5Ghz to the headset on AC1900. At that point, you should have low latency and high bandwidth.

Wouldn't it still need a power cable though?

Would obviously need a battery pack, ideally just a small lithium pack on an armband IMO. Just enough for a few hours of play.

 

[Methadras]

Tetherless will be a VR gamechanger. Patiently waiting for gen 3.

This is why I said in an earlier article about the Vive that it is the cleanest of all the headsets and now having this new development making it wireless will not only make it a game changer, but will make it the defacto standard.