Our virtual reality arcade platform, VRcade, real-time demo

[JamieVRcade]

Hey everyone,

For the past three years, we have been working on a virtual reality platform that is designed to be deployed, after a successful crowd funding campaign, to brick and mortar virtual reality arcades built and designed by us, starting in Seattle, WA.

The video below demonstrates the speed and accuracy of the system. There is no skeletal or finger tracking yet, though that is coming very soon. The game engine is in Unity. We can take any current Unity project and put it in the VRcade within minutes. We are working on an indie VR development kit as well.

http://www.youtube.com/watch?v=FcJZylnM ... e=youtu.be

It's exciting stuff. We have had a WIDE range of people play this, from 5 year olds to 62 year olds and EVERYTHING in between. There is no learning curve and we have the latency so low and the 3D immersion so high that the human brain actually believes that it is somewhere else. Upon removing the HMD, testers report that it feels like they have come BACK to the testing lab, although in reality, they never left. Testers refuse to walk through walls that aren't really there and they get freaked out by simple AI seeking robots that follow them. It's an interesting, intense, and extremely enjoyable experience for everyone who has tested it. Nobody has had anything but glowing things to say about it.

If you have any more question about our efforts, you can post them here. We will be blowing the lid off of the project very soon.

Thanks!

[AntonieB]

Looks nice!! would love to see more about your project.

[JamieVRcade]

Looks nice!! would love to see more about your project.

We should be releasing a ton of information soon. In the meantime, you can ask any questions that you might have here. This may become a good resource for others with questions as there is an awful lot to talk about.

[Krenzo]

Your music reminds me of the Area 51 arcade game. It looks good. Are you using an off the shelf optical tracking system?

[JamieVRcade]

Your music reminds me of the Area 51 arcade game. It looks good. Are you using an off the shelf optical tracking system?

We are. Using the optical tracking system wasn't the difficult part. The difficult part was getting that data acquired and into the game so quickly that we minimize latency. Pretty much all of the components are off the shelf at this point. Once the arcade opens, we will have quite a lot of proprietary gear and software.

[Krenzo]

We are. Using the optical tracking system wasn't the difficult part. The difficult part was getting that data acquired and into the game so quickly that we minimize latency. Pretty much all of the components are off the shelf at this point. Once the arcade opens, we will have quite a lot of proprietary gear and software.

Do you mind saying what system it is? Optitrack?

It also looks like you're using wireless video to feed the display. I don't see a laptop or anything on the back. Can you say what you're using for that?

[JamieVRcade]

We are. Using the optical tracking system wasn't the difficult part. The difficult part was getting that data acquired and into the game so quickly that we minimize latency. Pretty much all of the components are off the shelf at this point. Once the arcade opens, we will have quite a lot of proprietary gear and software.

Do you mind saying what system it is? Optitrack?

It also looks like you're using wireless video to feed the display. I don't see a laptop or anything on the back. Can you say what you're using for that?

Yeah, it's OptiTrack. Wireless video is sent over a 60 gHz wireless device. The player is connected to a high-end PC. The only wires on the player are the ones connecting the power and data to devices worn by the player.

[Krenzo]

Wireless video is sent over a 60 gHz wireless device.

Ok, cool, have you had any issues with occlusion? Do you just have one transmitter for the video? Are there any issues with latency?

Have you thought about multiple people playing in the same space for coop games, or are you focusing on one player?

[JamieVRcade]

Wireless video is sent over a 60 gHz wireless device.

Ok, cool, have you had any issues with occlusion? Do you just have one transmitter for the video?

Have you thought about multiple people playing in the same space for coop games, or are you focusing on one player?

The entire arcade concept is built around multiple people playing within the same capture volume, whether they are playing together or separate games. Testing will begin shortly on the ability to send out 8 different AV signals wirelessly along with bluetooth traffic. Single player, co-op, and real athletic deathmatch are all targets.

We have had very few occlusion issues. In addition, the cameras in the final arcade will be MUCH more robust than what we have now, and there will be quite a lot of them, so occlusion should be even less of an issue.

We plan to track the player's entire body, including fingers, by using proprietary VRcade suits that are easy on-easy off and which fit over most clothes.

[Krenzo]

Ok, I couldn't find any info on how many channels WirelessHD supports. Is manual channel switching supported out of the box?

[JamieVRcade]

Ok, I couldn't find any info on how many channels WirelessHD supports. Is manual channel switching supported out of the box?

We are in talks with chip manufacturers to give us that flexibility. Very few allow you to choose the channel because they are consumer focused and most consumers don't ever need to worry about that. Each PC will be locked to its own channel which corresponds to the appropriate player's HMD, ideally. It's on the 60 gHz band, so there is PLENTY of room. It's simply a matter of having a way to reliably dial it in.

[WiredEarp]

Damnit. I was literally about to register VRcade.co.nz this week. Haha great minds- guess thats one of the names that comes easily to mind when thinking of VR arcades.
I think you could do well, I would be quite interested in a system like this.

It looks nice and responsive - how many trackers does it support, and how much?

[Nick3DvB]

Amazing work Jamie! Many of us here have been envisioning something like this for a long time, struggling with the tracking issues etc, but it looks like you’re pretty much ready to go!

I looked at Optitrack but the cameras seemed prohibitively expensive, did you try anything that doesn’t tie you to propriety hardware, ipi etc? I was hoping a ton of PS3 cams would have done the job, the RasPi camera module looks interesting, it might have been a good IP solution, but the lack of focusing might be an issue?

http://www.tweaktown.com/news/29201/mor ... index.html

What’s the thinking behind a Kickstarter, it seems logical to sell this into existing laser-tag sites, it would be easy to add markers to the existing kit and they really seem to be struggling here in europe. That said it's probably best not to get tied into some kind of exclusive licensing with one outfit, the more people get to experience this the better! Congratulations on everything you've achieved so far, keep us posted!

[JamieVRcade]

Damnit. I was literally about to register VRcade.co.nz this week. Haha great minds- guess thats one of the names that comes easily to mind when thinking of VR arcades.

Haha, interesting.

[JamieVRcade]

Amazing work Jamie! Many of us here have been envisioning something like this for a long time, struggling with the tracking issues etc, but it looks like you’re pretty much ready to go!

I looked at Optitrack but the cameras seemed prohibitively expensive, did you try anything that doesn’t tie you to propriety hardware, ipi etc? I was hoping a ton of PS3 cams would have done the job, the RasPi camera module looks interesting, it might have been a good IP solution, but the lack of focusing might be an issue?

http://www.tweaktown.com/news/29201/mor ... index.html

What’s the thinking behind a Kickstarter, it seems logical to sell this into existing laser-tag sites, it would be easy to add markers to the existing kit and they really seem to be struggling here in europe. That said it probably best not to get tied into some kind of exclusive licensing with one outfit, the more people get to experience this the better! Congratulations on everything you achieved so far, keep us posted!

Thanks, Nick.

The infrastructure is definitely out of the price range of the average consumer, which is why the average consumer can come to us and use it for a MUCH smaller price.

We aren't interested in creating the perfect VR experience in the home. Due to a lack of physical space, cost, and infrastructure, that's not the smart move. The smart move is to create location based amusement centers where players have games designed around MASSIVE floor space where they can run, jump, flip, duck, crawl, etc. without having to fake it or move couches out of the way. Because of that, the infrastructure needs to be flexible and powerful, which is what these cameras give us. Using any other solution wouldn't cut it for this particular application. We aren't covering a desktop or small space...we are building the world's largest optical capture volume to play in.

As for Kickstarter (we will be using IndieGogo), the applications for a system like this are incredible, even outside of gaming. In addition, you need to have space custom tailored for it. So putting it in an existing laser tag facility would only work if they got rid of all of the walls (occlusion nightmares) and offered just VRcade games, which, in my opinion, would be the smart thing to do. The amount of flexibility that you get in our system is astronomical compared to anything else out there. With that comes the need to architect around the technology, so building our OWN space to house this system is crucial. We are also able to custom tailor the experience, from entry to exit, to make it an incredible place to just BE, even without stepping into a game.

In regards to proprietary software and hardware, the VRDK will support a WIDE range of engines and cameras. For our purposes, we will be using this system, but like MiddleVR, our system will allow you to integrate our high standards into your game engine of choice. We are a completely flexible platform, even on our end. As components change, we can swap them out and replace them with better ones. Same goes for the software and high-end hardware.

Thanks!

[cybereality]

Looks awesome man!

[ChrisJD]

This is definitely the kind of group experience I could see being a lot of fun. Even if I had some kind of standing setup at home with an omni treadmill of some sort I could still see myself going out to this kind of vr arena with friends.

[JamieVRcade]

This is definitely the kind of group experience I could see being a lot of fun. Even if I had some kind of standing setup at home with an omni treadmill of some sort I could still see myself going out to this kind of vr arena with friends.

Imagine....get ready for it....Ghostbusters.

4 players. Your name on your badge on your chest. An actual Proton Pack backpack with gun and ghost trap. We can track the trap and the gun, along with your full skeleton (hands, feet, fingers, etc.). We can make each individual player hallucinate since they all have their own displays, audio, and PCs. You will have to take cover, shoot targets near and far, work together to wrangle ghosts, dodge attacks, use PK meters, everything. And of course, avoid crossing the streams.

[Nick3DvB]

Imagine....get ready for it....Ghostbusters.

Excellent! But you'd better get some feet markers sorted out, you can't kick an Ecto-trap out without them!

So you're sticking with OptiTrack long term then? I realise you need to go for an industrial solution, and I’m sure you’ve done your homework, but I'm just a bit wary of NaturalPoint in general, given the history with TrackIR etc - their consumer kit isn’t exactly consumer friendly… planned obsolesces, encrypted API issues etc. Good luck with it.

Standard laser tag sites would need adapting for sure but it still seems like a good fit, they better get on-board with VR soon or they won’t be around for long, Hasbro are starting to get the idea though. As you say, flexibility is key, keep it as open and modular as you can. It look’s like you are using DIY Rift clones in the videos, are you partnering with Oculus for the HMDs? It would be a great in terms of mutual promotion.

[Krenzo]

As for Kickstarter (we will be using IndieGogo), the applications for a system like this are incredible, even outside of gaming.

Are you aware that there was a similar project that tried to get funding via IndieGogo? (Thread 1 and Thread 2) While your demos are a lot more advanced with an actual demonstration similar to your final setup, there were a lot of good points raised before about how crowd funding does not fit with this type of business. Crowd funding is based around backers getting something in return for their money, but what you're offering is an experience that is limited by geography which limits the amount of potential backers.

[JamieVRcade]

Imagine....get ready for it....Ghostbusters.

Excellent! But you'd better get some feet markers sorted out, you can't kick an Ecto-trap out without them!

I realise you need to go for an industrial solution, and I’m sure you’ve done your homework, but I'm just a bit wary of NaturalPoint in general, given the history with TrackIR etc - their consumer kit isn’t exactly consumer friendly… planned obsolesces, encrypted API issues etc. Good luck with it.

Standard laser tag sites would need adapting for sure but it still seems like a good fit, they better get on-board with VR soon or they won’t be around for long, Hasbro are starting to get the idea though. As you say, flexibility is key, keep it as open and modular as you can. It look’s like you are using DIY Rift clones in the videos, are you partnering with Oculus for the HMDs? It would be a great in terms of mutual promotion.

As mentioned in the post, we will be tracking the full skeleton. Feet included.

We have been in contact with Palmer even before his Kickstarter campaign began, so he knows what we are up to. We intend on using all Rifts as well as selling them as merchandise. We believe that our VRcade is the perfect fit for the Rift. No question.

[JamieVRcade]

As for Kickstarter (we will be using IndieGogo), the applications for a system like this are incredible, even outside of gaming.

Are you aware that there was a similar project that tried to get funding via IndieGogo? (Thread 1 and Thread 2) While your demos are a lot more advanced with an actual demonstration similar to your final setup, there were a lot of good points raised before about how crowd funding does not fit with this type of business. Crowd funding is based around backers getting something in return for their money, but what you're offering is an experience that is limited by geography which limits the amount of potential backers.

We are aware of that limitation. The fact that only people in Seattle/Bellevue will likely contribute isn't a massive hurdle, though. This is one of the best locations on the planet to be trying this. Once you can prove a business model, developer SDK, indie VR test bed, a wide range of experiences and technology, I think the chances of us succeeding (especially with the momentum of VR and the Rift) are high. Especially once you consider the other applications outside of gaming.

Additionally, IndieGogo will let you keep whatever you raise, so the scale of the project and retail space can adjust based on what we DO raise.

[adoral84]

I love that positional tracking, it's looking great. Have you done any other props aside from the gun stuff? If you can work a lightsaber in it'd probably blow the doors off the kickstarter campaign. That's the best end to end VR demo I've seen at this point, can't wait to see where this goes. Are you going thinking of having "stage hands" like palmer was talking about at SXSW to manipulate or place physical props?

[DaveRuddell]

I guess I'll chime in, since I've been lurking for a while. We talked in great length with Palmer when he was up here for PAX, and if we can get the space we really want (150'x75' usable capture volume), we'd love to do some mixed reality experiences with some ninja stagehands. However, some experiences wouldn't require it, plus it's nice to quickly turn the space for the next player without having to clear the floor. As time goes on it'll be really important to figure out what people want to play and make sure developers are happy and don't have to worry about what objects in the level are physical mixed-reality objects and which ones aren't.

[Dantesinferno]

Looking pretty beast bro!

[TheLookingGlass]

Sorry if this has been asked already what is your HMD? Is it basically the same components that made up the prototype Oculus Rift (5.6" Display, 2 Aspherical Lenses, Hillcrest Tracker)? Or have you come up with your own design? This is looking pretty awesome! I really loved the video of the 62year old lady shooting like a pro!

[JamieVRcade]

Sorry if this has been asked already what is your HMD? Is it basically the same components that made up the prototype Oculus Rift (5.6" Display, 2 Aspherical Lenses, Hillcrest Tracker)? Or have you come up with your own design? This is looking pretty awesome! I really loved the video of the 62year old lady shooting like a pro!

Palmer was pretty open about the design of the Rift, so our software engineer basically mimicked that, creating what is equivalent to Palmer's original prototype. We should be receiving our development Rifts soon.

[JamieVRcade]

I love that positional tracking, it's looking great. Have you done any other props aside from the gun stuff? If you can work a lightsaber in it'd probably blow the doors off the kickstarter campaign. That's the best end to end VR demo I've seen at this point, can't wait to see where this goes. Are you going thinking of having "stage hands" like palmer was talking about at SXSW to manipulate or place physical props?

Lightsabers are odd. When they contact a target, they cut right through it, which is perfect, but when they contact another lightsaber, they clash. So they will need a real prop to clash into, but now you have a real prop being swung at real people wearing expensive gear.

I can tell you that we will find a solution, though. Co-op lightsaber and force powers is very doable, however.

Regarding physical props, you run the risk of occlusion with real props. Also, they would all have to stay stationary in order to line up with their in-game counterparts. Also, if they were stationary, the level could never change. Instead, all props, walls, and obstacles are virtual. Players can walk through any wall they want...they will just leave their in-game character (who must obey the laws of in-game physics) behind to get shot or killed. There is no advantage to walking through walls. In this way, having an open level is smart because with a system this immersive, people DON'T WANT TO walk through walls, even if they are virtual They will behave correctly. All 100+ test subjects have so far and I believe that trend will continue.

[Nick3DvB]

I think having physical walls / props is still very important, being able to actually reach out and touch, or lean up against, a solid wall would be a very powerful addition, having them perfectly mapped to the virtual geometry might be tough, and then there are the obvious occlusion issues with your optical tracking system, but it might just be possible with enough coverage. This seems like another good reason to target laser-tag arenas, they have the physical infrastructure in place now, and the demand is certainly there, why limit yourself to one site when you could potentially go global? If you don't then one of the established players might get in there first, I wouldn't risk loosing your first mover advantage.

[JamieVRcade]

I think having physical walls / props is still very important, being able to actually reach out and touch, or lean up against, a solid wall would be a very powerful addition, having them perfectly mapped to the virtual geometry might be tough, and then there are the obvious occlusion issues with your optical tracking system, but it might just be possible with enough coverage. This seems like another good reason to target laser-tag arenas, they have the physical infrastructure in place now, and the demand is certainly there, why limit yourself to one site when you could potentially go global? If you don't then one of the established players might get in there first, I wouldn't risk loosing your first mover advantage.

Oh don't get me wrong. We are targeting everywhere. I want VRcades in airports. Selling to fun centers and laser tag places is going to happen. On our end though, you invest hundreds of thousands of dollars into ONE level due to space and cameras if you build a physical infrastructure. You either can only ever play on that one level or you hire a crew to setup entire levels before and after every game. If you want to change a level or game, give us 30 minutes to set it up. It's not practical.

Having no obstacles gives you complete flexibility. You don't get to lean on things, but the pros greatly outweigh the cons. It would take hundreds of thousands of games played in one of those spaces (levels) to make up for the infrastructure and real estate costs of just that one stage. You also can't use that space for things like indie development or other non-game related experiences because the layout is rigid and set.

And the truth is, you may be able to track head and prop in a wooden maze MAYBE, but you won't get full body tracking. There is no way multiple ceiling mounted cameras would be able to track your feet, for example, in such a crowded space. So now, you have less immersion and less ability to punch, kick, go prone, etc.

Additionally, the way the human brain works in 3D is pretty amazing. It will believe that things are there visually, but after a few attempts at touching things that aren't there, it will learn to adapt. It will learn that it can take cover behind things, but those things can't support it's weight, as if the whole level was made of fragile paper boxes. You can hide behind them, knock them down, etc., but they just can't support your weight. So like in real life, don't put your weight on them. We have already seen people go through this cycle of wanting to lean against a crate that isn't there, stumble, laugh about it, and never do it again.

So while we will find ways to deploy to current retail attractions, they won't be as robust as the VRcade and they each will have their own challenges to overcome. You can NOT have this system be glitchy or unstable. If you are paying for a game, it needs to work perfectly. Doing anything that compromises that is an issue.

[Nick3DvB]

A dedicated VR space is obviously preferable in many ways, but there's a lot of business in retro-fitting into existing spaces, you could do a lot more than just re-skinning a static maze, a mixture of real and virtual barriers could work using different colored textures / force-field type effects etc, and you could use a few clever "tricks" like re-directed-walking or re-orientation portals (elevators etc). I wonder if optical tracking would work through transparent barriers, something like perspex panels that could be easily re-positioned, just a thought.

[JamieVRcade]

A dedicated VR space is obviously preferable in many ways, but there's a lot of business in retro-fitting into existing spaces, you could do a lot more than just re-skinning a static maze, a mixture of real and virtual barriers could work using different colored textures / force-field type effects etc, and you would have to employ a few "tricks" like re-directed-walking / re-orientation "elevators" etc. I wonder of optical tracking would work through transparent barriers, something like perspex panels that could be easily re-positioned, just a thought.

Getting it to actually work would be the challenge, yeah. The cameras shoot off ir, which is reflected off of the markers. To have any clear barrier, you start running into reflection issues. We ran into this problem while trying to decide on the materials surrounding the play space. We wanted people to be able to look in, but we didn't want the camera light to reflect or to see other light sources from beyond the glass.

[Nick3DvB]

I can see reflections being a big problem, but if your tracking system could be informed about the geometry / material properties maybe it could account for them, maybe even track off them, certainly not a trivial issue though. If only someone had a robust alternative to optical tracking...

[JamieVRcade]

I can see reflections being a big problem, but if your tracking system could be informed about the geometry / material properties maybe it could account for them, maybe even track off them, certainly not a trivial issue though. If only someone had a robust alternative to optical tracking...

I think we would have an easier time convincing laser tag to rip out their guts and offer a clean, blank room. You can re-create the original level layout in VR and create a laser tag sim that gives you all of the experiences you used to have and WAY more if you wanted to. Not having real walls to lean up against really isn't a big issue once your brain acclimates. I'm not denying it would be nice, but hoping for a miracle solution that could offer the fidelity, positional accuracy, and speed of an optical system just so you could play around ONE set of rigid barriers that you can lean against isn't worth it. You'll see what I mean when you eventually step into a VRcade. Everything just works. So far at least! I could be proven wrong in testing, but from the demos we have run, I don't see it being an issue.

[JanVR]

Jamie, this is absolutely impressive, congrats! You are building the most immersive VR experience to date.

[Nick3DvB]

I understand the technical merits of your position, I just think convincing an established bushiness to "rip out its guts" might be a tough proposition, they have a large investment in equipment (contracts that need to be maintained with suppliers etc) and a small but loyal membership that might not want to completely switch to VR. You don't have to convince me, I'm sold - just playing devils advocate here. Getting their management to realize VR is the only chance they have of saving their bushiness might be easier if you sell it as an upgrade to their existing assets.

[DaveRuddell]

I think you are right Nick, we need to have a solution for allowing continued use-cases in places like laser-tag or paintball, but build camera configuration and a level around the existing laser-tag arena to get that nice feeling of mixed-reality. Half the fun of laser-tag is running up on ramps and leaning against corners, so if we can find a way to leverage the existing physical constructs, we have an even sweeter experience.

From a business perspective, we can approach them with 2 figures and what they get for each one. Gutting the place and putting up cameras gives them flexibility in the experience, plus doesn't lock them into laser-tag, but it's more expensive for them, plus they can't use it for laser-tag anymore. Option 2 is the figure out the best camera config (definitely more cameras required to maximize play area) and do a part-time mixed reality arena, allowing them a set of "environments" to play in (alien planet, space freighter, forest, etc). Then they can charge a premium to play Lazertag-VR, or something to that effect.

[DaveRuddell]

I should also correct Jamie in his description of our HMD. It's a DIY rift that I built based on the things other people in the forum were doing, and with a little help from some of you as well. I used Rod Furlan's original posting and discoveries (from watching the KS video over and over again) to nail down the parts I needed, then I 3d-printed a few pieces from WickedAndy's awesome RA-Rift design to build a sturdy backbone for it. The last piece was my own version of the LVDS->HDMI connector that I mounted to the face of the HMD and covered with tape to conceal the ugliness within. I meant to post pictures a long time ago when I finished, so maybe I'll go do that soon.

The one difference between my DIY Rift and others is that it does not have a built-in IMU for head tracking. Instead, I am relying on the motion-capture markers to track the rigid body in 3d-space. We get a round-trip of less than 1ms into Unity by using this method and update rates of about 180fps. So the tracking is fast, maybe even faster than the Rift (definitely faster than the Rift prototypes). I think our limiting latency factor will be the panel itself. Already it's only noticeable to people who know to look. Joe-Schmoe and his mom didn't notice, but they also have to point of reference. *shrug*

I'm curious to see what latency from the tracker is like on the Rift Dev units we're getting soon. I'm hoping to build a hybrid of the 2 to ensure absolute position and orientation while leveraging the speed and flexibility outside of the capture volume from the Rift tracker. Who says we can't have the best of both worlds?

[Mystify]

You could utilize the fact that walls are non-solid.
For instance, as an explicit mechanic you can walk up to a wall, and look through. Doing so leaves your body behind and vulnerable, and you end up as a "ghost" on the other side, and can use it for quick scouting, but are unable to attack. Your ghost self may or may not be vulnerable, though it should be visible so people don't run into your real life body.
You know better than I what you are doing, and whether something like that may be appropriate, but in general I think it is best to utilize the shortcomings of your setup as features rather than them being bugs.
On the other hand, it may prevent or hinder the acclimation effect you mentioned, and consequently be a hindrance.

[JamieVRcade]

You could utilize the fact that walls are non-solid.
For instance, as an explicit mechanic you can walk up to a wall, and look through. Doing so leaves your body behind and vulnerable, and you end up as a "ghost" on the other side, and can use it for quick scouting, but are unable to attack. Your ghost self may or may not be vulnerable, though it should be visible so people don't run into your real life body.
You know better than I what you are doing, and whether something like that may be appropriate, but in general I think it is best to utilize the shortcomings of your setup as features rather than them being bugs.
On the other hand, it may prevent or hinder the acclimation effect you mentioned, and consequently be a hindrance.

The nice thing about our system is that we are making the games AND the platform, so we can design experiences so tightly around the limitations that the limitation isn't even present to the end user. A solution like yours is a perfect example of that kind of flexibility. In some games, sure, let them peek through walls. In others, don't. You can do both because there is no real wall there...you can do whatever you like.