Hands-on: TPCAST’s Wireless Vive Kit Really Works

by Az Balabanian2 • December 19th, 2016

As exciting as the concept may be, internally we’ve been skeptical about Vive X company TPCAST’s wireless solution for the HTC Vive since its announcement last month. We have so many questions: how is it set up, is it too heavy, does the battery last long enough and, most importantly, what’s the effect on latency?

Last week, we got to answer some of those questions for ourselves. TPCAST visited UploadVR last week, and we got extensive hands-on time with the kit.

What Did We Use?

We used a Business Edition of the Vive supplied by TPCAST that had a receiver attached to the top of the head strap. It connected via short cords to the headset.

receiver-tpcast-overhead

TPCAST’s receiver sits on top of the Vive head strap. This is the engineering prototype we tested.

Along with the usual two lighthouse stations, there was a TPCAST video transmitter which was said to have a 160 degree field of view mounted high in the room facing the play area. The company said the transmitter is best set above the play space in the middle pointed downward. For our testing, the play space was 4 meters by 2.6 meters. TPCAST claims it can cover a 5 meter by 5 meter area with ideal conditions.

There was also a router connected to the PC which we were told was for the transmission of movement data. TPCAST says for the consumer version that unit will be reduced down to a dongle attached to the PC.

To power the headset, a battery pack was slotted into a back pocket with a cord running outside my shirt to the backside of the headset. Two versions of the kit are said to be going on sale in China next year, and we tried the version with the bigger battery life, said to last five hours. Our playtest with the headset lasted just about that long (ignoring the time spent restarting SteamVR and rebooting things) but we’ll have to stress test it with the consumer version to give you a more accurate reading of their claims. A version with a smaller battery that attaches on top of the head with the receiver, and is said to last up to two hours, is what was made available for pre-order already.

Everything we tried was running on our Falcon Northwest Tiki (4 Ghz i7 PC with a GTX 1080), though we also sampled the kit on a laptop TPCAST brought along with a 1070 stashed inside and found identical results.

Does TPCast’s Vive Wireless Upgrade Kit Work?

Based on our five hour play session — which ran the battery flat — it appears that it works as promised. We didn’t notice any difference in motion to photon latency while sampling VR experiences like Google Earth VR, Tilt Brush and Valve’s The Lab.

We were able to move around with complete freedom, including fast movements like jumping and flipping. Yes, we even brought in a gymnast to test some extreme movements possible only with this wireless positionally tracked headset. Even with accidental drops of the headset, receiver, and the prototype transmitter, it all still worked.

There were, however, instances where we noticed artifacts in the virtual scene, which momentarily reminded us that the headset was indeed wireless. These artifacts, which looked like a lower resolution streaming video, were momentary and mostly negligible to the overall experience of enjoying completely wireless room-scale VR. The artifacts were also hard to replicate by repeating movements or positions that we suspected might’ve been responsible for causing them. The most prevalent technical difficulties we dealt with was the connection between the physical wires on the headset, the receiver, and the battery pack.

One of our testers, a gymnast, did some ridiculous physical maneuvers, such as flips and handstands, and at one point the unit fell off his head mid-flip. The TPCAST crew were able to get the headset back up and running quickly. While we had enough time to quickly film these insane physical movements with the wireless headset, we unfortunately didn’t have the time to set up a proper test that would have used both a wired Vive and a wireless one to see if people could tell which was which.

I have spent hundreds of hours in wired Vives over the last year and I did not notice any latency differences between the headsets I’ve been using and this wireless one. We plan to do extensive testing with the consumer version of TPCAST when it arrives to fully examine the latency numbers, but for now, we found the overall motion to photon latency feels way under the 20 ms benchmark required for comfortable VR headsets.

It seems the location and setup for the transmitter will prove to be a crucial detail in having a seamless experience, similar to finding the right places to mount the Vive lighthouses. TPCAST says that the transmitter can be mounted above the lighthouse to minimize the overall real estate taken up by trackers around the house. To get the best results, however, you’ll need to have the transmitter in the middle of the play space facing downward, according to the startup. Dealing with forms of wireless interference could be another potential headache, but we didn’t see the headset having issues despite how much metal we had in our room (tripods, lights) as well as wireless mic systems.

The headset felt lighter with the receiver on that it does with the Vive’s thick cables pulling at the back of your head when tethered to a PC, and I didn’t note extra heat on my head during my playtime.

How Does It Work?

The video transmitter is placed high with a 160 degree field of view to cover play areas. Photo credit: TPCAST

The video transmitter is placed high with a 160 degree field of view to cover play areas. Photo provided by TPCAST depicting production version of transmitter.

That’s another question we’ve long wondered about, and something TPCAST CEO Michael Liu helped answer. He described the kit as “a bi-directional communication portal”. It uses a compression algorithm described as the company’s “secret sauce” with “Wireless HD” transmission we believe to be in the 60 ghz wifi band sending 2160 x 1200 video data.

Input like head and hand movements are transmitted via the router, while video transmission is handled with the overhead transmitter, according to the company. It’s important that the transmitter is placed high up to give it a view of the entire play area you’re using so that in “most cases” the user shouldn’t feel any “dead angles”.

Liu wouldn’t, however, promise that anyone that picks up the tech would be able to get it to work in their homes if they have a lot of interference around.

“This is a complicated technology,” he said. “We will do our best.”

Liu said it would be “very easy” to set up.

Developing the tech obviously came with its fair share of challenges, the biggest of which Liu said was keeping latency low. Currently TPCAST claims it adds less than 2 milliseconds of delay.

“The latency that we talk about is actually our latency, latency cost by our wireless device,” Liu explained, meaning it stacks on top of the latency already involved with using VR headsets.

As for concerns about the safety of using such a device, Liu wasn’t worried.

“I think wireless communication, the receiving part is actually okay,” he said. “I think a cell phone is more trouble.”

How Did It Come Together?

TPCAST itself has been specializing in making technology wireless for the past “five or seven years”, though the Vive kit has only been in the works for the last year.

“We started the project as an internal project because sometime last year we saw this big opportunity, this big wave of VR coming,” Liu explained.

The Vive kit got off the ground after a meeting with Alvin W. Graylin, China Regional President of Vive at HTC. Following that, TPCAST was one of the first companies to be inducted into HTC’s Vive X accelerator program, giving companies access to resources, working space, and mentorship as well as investment opportunities through other programs like the VR Venture Capital Alliance (VRVCA). HTC itself was one of the first outside investors in the company.

What Does The Future Look Like?

This pocket-sized battery offers five hours of play time. A small one with two hours is also coming. Photo credit: TPCAST

This pocket-sized battery offers five hours of play time. A smaller one is also coming. Photo provided by TPCAST depicting production version of the battery.

Chinese pre-orders for the kit opened in November and sold out very quickly. Liu declined to reveal how many units were sold in that first batch, though did promise another batch is coming and that the company will be demoing at CES next year. TPCAST is planning to ship the kit in China in Q1 of 2017, and it’s not possible to pre-order it internationally just yet, though Liu said it is in the certification process for the FCC right now.

As for a launch in the US, all Liu would offer is “sometime next year.” A US price for the kit has not been set.

Liu also told us that TPCAST has support for other headsets on its road map, though it was “hard to say” when it could turn its attention to those other products. TPCAST is working on a version of the Vive kit that outputs in 4K too, which is likely to future-proof the device for when new headsets support that resolution.

Staff Writer Jamie Feltham and Senior Editor Ian Hamilton contributed to this article.

Tagged with: , , ,

  • Robbie Cartwright

    Wow! Good to know that the wireless sender doesn’t seem to add any lag to the experience. Thinking of buying a Vive to cross-develop between it and Rift, and a cabeless solution sounds great!

  • AcroYogi

    This is truly an epic step forward for the industry. still requires an immense amount of hardware, but for many market verticals, this will be an absolute must-have. expands HTC market to much larger segments: industrial, medical, training, aerospace. impressive accomplishment.

  • Petr Legkov

    the fact you actually have been searching for ninja back flipping people suddenly makes a lot of sense xD

  • Nicholas

    Want to know more about those compression artefacts, but this looks very promising indeed!

    • unreal_ed

      My guess would be that they do some compression on the more peripheral pixels. You can look at them but you usually spend more time looking more-or-less straight ahead. In that scenario you wouldn’t notice compression artifacts.

      Plus, once eye-tracking becomes a thing, you’ll be able to compress the video data even more, at least theoretically

      • Nicholas

        I’m hoping they output the full raw HDMI stream, and only resort to compression on a lower band if the 60GHz signal is lost.

  • bibberbiber

    did anyone ask about the health issues a 60ghz signal could provide, when wearing such a transmitter on your head for aprox. 10 ours a week.
    didn’t read the whole articel but that’s what less people consider by offering such devices.

    • Mourz

      Yes, it is in the article and they say that it is less dangerous than a cell phone.

    • Lachlan Sean C

      Still non ionizing radio waves. Shouldn’t be any worse than getting infrared blasted at you (ie not dangerous)

      • silvaring

        And i suppose we have data from a variety of individuals with years of exposure near the brain in that freq. spectrum to come to such conclusions?

        • Lachlan Sean C

          Hey I’ll volunteer if you want to sit this one out for a few years

        • John Nemesh

          We DO have years of data from reliable scientific papers that show that the human body is totally unaffected by electromagnetic energy in this spectrum. Google it.

    • Michael Gossett

      I implore you to learn English.

    • WiredEarp

      it would be truly terrible if it caused spelling and reading levels to decline.

      • lightsteed

        lol

      • Ryan DeLuca

        I literally created a Disqus account just to upvote this. lol!

  • Tom Daigon

    I find it odd that there was no mention on the image resolution / quality. How does it look as compared to the HDMI signal thru the cable?

    • Tom Daigon

      Liu says in interview its 2K at 90 hertz.

    • Mourz

      It is in the article, it says that there was no decernable difference, but they will do further testing once they get the commercial product and can do a side by side comparison.

      • JustNiz

        You’re totally ignoring the part where It says they noticed some artifacts.:

        “There were, however, instances where we noticed artifacts in the virtual scene, which momentarily reminded us that the headset was indeed wireless.”

  • Badelhas

    Great article and I am very happy to know from this preview that it really works. Can’t wait to see the first final consumer version reviews and the final price. If they keep it under 200 bucks I’m getting one for sure.

    Cheers

    • Gav

      Based on the statement this is “60GHz WiGi”, this is WiGig and currently a Dell WiGig dock is $269 on Dells website. I’d expect a price a little higher than that considering this is two devices, kind of.

      • Badelhas

        That’s a pity. I already spent 900 euros on the Vive and 500 on the GPU and I don’t use it that much, at least for now. I’ll guess I’ll stick with the cables for now, then.

        • hyperskyper

          I believe it was around $200-$250 for the preorder on the Chinese site.

      • John Nemesh

        No it is NOT “WiGig” of whatever you want to call it. 60Ghz wireless technology has actually been available for YEARS now for video applications, Gefen was actually selling a wireless HDMI kit using 60Ghz tech…but it isn’t the WiGig 802.11ad tech that is coming out on the market…it’s using the same spectrum, but it’s not the same technology and could be using any proprietary protocol they want to.

    • MR Not Dux OSAR

      Definitely looks like a must buy. Cables are a drag… literally.

  • Sleeping Lesson

    I’d like to know if it supports Super Sampling resolutions, as SS has turned out to be super beneficial to VR games.

    Also, I don’t like the pocket battery. Wouldn’t it be more ideal to just be able to swap out the 2-hour battery?

    • Az Balabanian

      Unfortunately, that’s the one thing I wanted to test but had run up on time. The developers said that they hadn’t tried it either, but I wouldn’t expect for it to not work.

      • Lou Cipher

        once the SS is done on your GPU/CPU, you’re left with the same resolution images at the same bandwidth as if you did no SS at all. it’s all upstream on the PC and is irrelevant when considering HDMI or wireless specifications with connections to the HMD in mind

        • Ugur

          You are totally right regarding the super sampling having no impact on the resolution of the final image delivered to the headset.
          Regarding why they chose a lossy image compression, well, likely because it leads to much smaller filesize/lower bandwidth needs, so that makes sense to do.
          It seems like it was only very rarely in noticeable way though so i wonder if it only kicks to more lossy level when the transmission is not ideal.
          In an idealised implementation it would compress the picture heavier towards the edges of the screen more than towards the less positions, not sure if they already do that. And in yet another improved version (but that would require eye tracking hardware in the headset so can only come in later headset iterations) it could then compress everything more than what one is really looking at right now.
          I’m curious to see if it does any such more targeted area specific compression in different levels at all already or it’s with same compression level on the whole picture, but yeah, we’ll see once it’s out.
          In either case, the article and video makes it seem like not an issue during most of the play time, so that sounds good =)

        • DougP

          Re: ” i do wonder why they opted for a lossy image compression algorithm tho”
          I understand the sentiment, but the likes of DVD, bluray & bluray UHD (4K) also opted for lossy image compression.

          Further still, consumers seem fine with even lower quality (higher compression/more artefacts) in their video – see Netflix/youtube popularity for watching shows/movies.

          Re: “prefer to wait for a lossless option”
          Unfortunately, if you wait, I suspect you’ll be waiting indefinitely.
          Compression will become even more necessary as HMD resolution increases.
          Even the likes of a Steam link box, over a physical wired network connection use compression at 1080p.

          It’s not exactly analogous or a precise comparison, but think of it this way, just watching a 3D bluray with high-quality (& mutli-channel) audio – the vast majority of internet connections can’t stream that much data over a wire (cable modem/fiber) & that’s compressed source material already. That’s 48fps at 1080p & a Vive runs at 90fps (close to double the framerate).

          Resolution & framerates for VR will only increase… so low-latency wireless bandwidth will not be able to keep up w/o compression.

          My motto –
          jst sy no to lssy cmprssn!

    • DoubleD

      SS happens completely on the PC.
      It sends the same amount of pixels to the headset.

    • Ugur

      This should totally work because the image one gets with super sampling at the end is still the same resolution as without super sampling. What gets done there is basically the image gets rendered at higher res and then shrunken down to the final (same as without super sampling) target resolution (the resolution of the headset display) so it then looks way sharper.
      So the image transmitted to the headset, no matter if with or without tpcast has the same resolution with it as without it.

      • Reuben Ahmed

        I have a gTX 1080 and I haven’t been “super sampling” any VR games on the Vive. Any more info? I will google now…

        • Ugur

          As user you would change the values of MSAA (Anti Aliasing) or Renderscale if the game/app has those adjustable in it’s settings menu to trade better look for performance.
          In some games/apps for the Vive/Steam VR one can also just press the escape key to show a menu where one can adjust some such things.
          Besides that it’s not something one does as user but rather what developers implement.

          • Reuben Ahmed

            Thank you.

  • Unknown

    ‘in China in Q1 of 2016’
    Is this a mistake?

  • Ugur

    This is a great article and video, some of the best coverage on a VR related product so far for sure, major props for that =)
    I’m quite excited about your test results, like you i wished it would work well but was quite sceptical up front, it just sounded too good to (already) be true.
    I so hope they can deliver the same (or better) for the consumer retail version (and not do what some companies (for example MS) do, where a great well working prototype is shown and then something working way worse is sold as consumer retail version.
    (which then deservedly usually also backfires heavily, see everyone moaning about cut back hololens compared to prototypes)
    If this works as well as shown in the video with the consumer version, this will for sure be an instant purchase for me (and many others).

  • Wow, finally an answer to my question about if this kit works or not!

  • I’d wait for independent safety tests before attaching that so close to my skull. And the fact that their CEO starts his every sentence with “I think” is not particularly reassuring.

  • Bassem B.

    Do you have any info on how the Vive’s camera is handled? Is it usable without much effect on the bandwidth? Because I know that even with wired Vive the camera can cause performance issues sometimes. Did you get to try Tron mode and/or the dashboard camera view on the controller?

  • ummm…

    im sure this question has been asked. i live in new york city. what is the quickest way that i can get my hands on one of these?

    • DougP

      Re: “what is the quickest way that i can get my hands on one of these?”
      Probably ordering one online & paying for it….once they are actually produced, clear FCC, & priced for US market.

      • ummm…

        I was hoping u knew a guy in Chinatown, or a reseller. This is awful news. I wont recover.

        • John Nemesh

          Well, it does say at the end of the article that the “first batch” sold in China sold out already, and that they need FCC licensing for US sales…I don’t think it would even be possible to import one if it doesn’t have the FCC’s blessing…

          • ummm…

            oh dear. blow after blow. are you guys trying to destroy my morale?

          • John Nemesh

            It did also say that they expect to ship to the US market mid next year…patience! 🙂

          • Reuben Ahmed

            Why we have to wait so long?

          • John Nemesh

            It takes time to get the logistics in place, have distribution agreements signed with the people who will be selling them in the US, and of course, to get FCC certification completed.

          • Reuben Ahmed

            Thank you.

          • John Nemesh

            Get your wallet ready, though! Once they DO announce US availability, I am pretty sure they will sell out almost immediately! 🙂 I know I am buying one!

  • Crate – A – Day

    Pretty surprised this worked as well at it did within the same year that the Vive came out. The future will be great.

  • MELT

    Would it work with 2 Vives in one space or would the kits interfere?

  • CMcD

    I’m curious what the furthest distance a pc can be from the vr area when using this. I’m currently looking for a solution to run my wires further as I need to move my pc physically closer to play vive games. I wonder if this would solve that issue for me while also alleviating wires. Definitely looking forward to reading your full review of the consumer model

  • Markku Hänninen

    Must buy after they release it elsewhere too. I think they are using chinese as guinea pigs to trim out the bugs.

    • user

      please what? china overtook the usa as the biggest gaming market. it’s not a lab.

      • Markku Hänninen

        EU and USA combined are much bigger + they pay more for the product. But it’s difficult to say for certain as we don’t have Vive sales numbers available. I’d make an educated guess and say that EU and USA citizens have much more disposable income and as Vive setup isn’t cheap in any way, the numbers should be much higher in USA/EU. How many chinese gamers can pay the cost of the system?

        • user

          millions of gamers can afford the vive in china because they use it in internet cafés.

  • The Omnivore

    They should sell a kit to attach it to a fan. I have a fan over my play-area!

  • PhotonArena

    We were neighbors at CES!

    The number of major innovations coming out for VR in just over a year is amazing. Really makes me wonder what’s coming down the pipeline.