TPG CEO answers more questions on Totem VR. Says “We would have never gone with Facebook”6 Apr 2014
TPG‘s CEO Bertrand Nepveu hit the Internet this evening to answer more questions about their upcoming Totem VR headset. The questions were part of a AMA to clarify more details about the headset which is due at the end of this year in the form of a first dev kit.
There were some decent questions asked and there were some new bits and pieces. You can also read our interview with Bertrand earlier this week which covers some of the topics).
Bertrand elaborates a little more on TPG’s plans for positional tracking and he says that if Facebook had approached them for acquisition he would have turned it down. If it had been Valve that had approached them on the other hand then that would be a different story.
Of course it’s easy to say you would have turned down an offer from Facebook after the event but perhaps he really thinks Facebook is not a good fit for VR and his company.
There’s still a lot of unanswered questions but this is due to the fact TPG are keeping some things close to their chest and they’ve still to demo the device which is planned for E3.
Are you worried about violating anyone else’s patents? Competing high tech hardware manufacturers can get litigious.
We have applied for two patents of our own and we take IP seriously. The worst thing that could happen to VR is a patent war. This is not part of our business plan.
How are we expected to be interested if you have not shown a working prototype so far all I have seen is a rendered image.
More info on our mailing list and a Demo at E3.
Programs for the PC that allow you to use existing games with hmds do so by modifying the shaders to change the fov, warp and create the stereo view. How can you achieve an immersive experience with consoles, when you only have the output from the hdmi?
I agree that this is not optimal and will depend of the game you try, but this open also the door to new VR indie games on the console.
What realistic methods do you have planned for positional tracking? Using the front mounted cameras for this poses several big challenges. Doing so would mean that users would have to mount several IR LEDs around their walls and ceiling to make it reliable. You also would have to remove any IR filtration from the cameras, which would degrade the picture quality when used as a pass-through or for augmented reality.
Our Consumer Version goal is to do positional tracking without any use of any external markers. We have lots of developments in that regards but we don’t want to reveal it right now… you know, keeping an edge on the competition
Based on what the company has said in the past it seems like they want to handle positional tracking similar to how an optical mouse works. They want the cameras to take pictures of your surroundings, then measure the changes between images to calculate movement.
Yes, this is our plan.
Especially when playing with my Rift, I don’t care about room lighting and when the sun goes down I lift my HMD and it’s gone dark – bad for vision based tracking without it’s own illumination. Do you have plans for this?
IR can overcome that.
Would you say that you are focused more on AR than your competition? Or do you view TPG as a VR-first company?
We are definitely a VR company, but you can also go cool things with AR. It will be to the Devs to experiment on new concepts. Cameras will be use for Pass-Through vision and Positional Tracking.
How will they help with positional tracking?
For the answer, you can check how Valve is going their positional tracking.
Do you expect consumers to take their headsets to valves demo room, or create their own? What if they don’t have a room with the same dimensions?
No, we want to do tracking without markers and use your normal gaming environment.
Building hardware is an expensive business and a lot of people did not like Oculus solving the money problem by taking 2bi from FB. Would you try to get capital which limits your control over the company (e.g. venture capital or a buy out offer)?
We want to have as much control over our vision as possible. We would only go with a partner that would keep our HMD as open as possible.
Everyone who invests want’s some degree of control. Oculus claims that FB enables them to build a open product with all the freedom they want and people are still afraid of the deal – so you would basically go for a similar way?
No, we would have never gone with Facebook but maybe with Valve
Microsoft or Sony?
Good question, I really don’t know. It would depends of a lot of things. We want to be on PC and be cross-platform. Would they want that?
Can the Totem be USB powered or will it need a power brick (considering it has to drive the screen, two cameras and an FPGA)? What framerate will the cameras have? Will the vision based tracking with the cameras and the sensor fusion with the internal accelerometers/gyros be done in the FPGA (I guess so as it would otherwise not work on consoles)? What FPGA are you considering that can handle all that stuff + the lens distortion?
Are you working for Sony/Microsoft/Oculus/Intel ? It will not be USB powered.
When do you think the Totem will be available for public retail, and how much do you expect it to cost people to buy?
We are aiming to deliver Dev kits for Xmas. The consumer version will be in 2015 if we can get 1440p oled panels. The price a little more expensive than the OR because we have more features and hardware.
Can you please elaborate? What features/hardware will this have beyond the Oculus Rift?
Onboard Cameras, Console compatibility, Surrond Sound and Hardware Acceleration.
The console compatibility and the hardware acceleration are basically the same thing though, as the first depends on the latter. (I do applaud it however) Latching the headtracking to joystick input is nice, but doesn’t take a lot of effort and will most likely be nauseating. The onboard cameras don’t cost a lot – the trick will most likely be to eliminate the problems that come with such a passthrough – there will be a quite some.
Hardware acceleration is also onboard pre lens distortion!
What else would it be?
Sensor fusion and positional tracking.
Will you guys incorporate your own motion controllers? It would make you stand out from the competition!
Lots of other people are working on this and this is not a priority in the short term.
The website says it will be compatible for ps3/xbox 360. I assume only for movies ?
No, for games too, we emulate the thumbstick for headtracking. While not optimal, it will enable you to try 3D games you own in VR mode.
How do you handle roll/positional tracking when emulating controllers?
New games would have to be developed with that in mind. This enable also indie game dev to go on the console market and develop cool VR games!
Dragonbox Pyra (an ultramobile PC) developers have talked about the troubles in finding a landscape LCD that would work for their purposes. Are there any issues with the LCD planned for use in your devkit? Is it natively portrait-mode? Rolling shutter, latency between eye updates, etc…?
I agree that current hardware is not designed with VR in mind. With the acquisition of Oculus by Facebook, it is bringing VR in the spotlight, which will be great for VR manufacturer (panel manufacturer will listen more and understand the market potential). With the help of our onboard FPGA, we can compensate for the shortcoming of current mobile technologies.
Are you aiming at a seated experienced as well like the Oculus guys? Or are you going to give better compatibility to devices like the Omni?
I ordered an Omni and can’t wait to receive it. For us, it’s part of the future of a great VR experience.
If you can achieve great positional tracking with a camera attached to the device, definitely it makes it much better than having one in front of you that would lose tracking when losing sight.
The thing is, how is the camera going to know the positioning. I guess you would have to stick some stuff on the walls as reference or something like that (Dunno if there is another way, but that would be cool with most people I think)
We have ideas that we can’t share right now
Do you consider adding a latency tester? Low latency seems to be one major goal for immersion.
Yes, we will add a latency tester in our FPGA.
I could probably look this up online, but I haven’t checked. Are you working on this in Canada, or did you move to the States? Is Canada effectively on the VR map?
If Oculus and Omni and Valve etc had never started reviving the VR scene, do you think you’d still be working towards a consumer product, or do you feel the commercial viability would have been much more questionable and you would have kept it as a hobby?
We are based in Montreal and it’s a great place to be for games. We also need to have some East Coast love. We only announced because of the Facebook acquisition which took us by surprise and I couldn’t understand why Facebook!
We did this for 9 years and it as nothing to do with Valve, Omni and Oculus. It was never a hobby for us. Our competitive advantage is the flexibility of our FPG.
After you saw the Oculus Rift DK1 at E3 you then decided to scrap your current project and instead started working on something that’s strikingly similar to the Rift. I’m happy that you’re still passionate about virtual reality and that there will be competition in this field, but at the same time the Rift and Valves research has obviously had some influence.
Yes, it surely had some influence on the new design but we reused 75% of our code and hardware for Totem. We had cameras and FPGA on our earlier prototype. We just changed the microdisplays for a mobile screen. Doing 90FOV with microdisplays was impossible.
Could you share your vision about VR? Where do you want to get with the totem now and how do you see it going for the next 10 – 20 years? Do you think you are different enough from the oculus rift to make a different impact on the VR world?
We want to provide a device that enables people to create what their minds can imagine. You will be able to do whatever you want in VR. It will be up the the users to decide what they want to experience. With great power comes great responsibilities! I am a hacker and I want other people to hack our device and do things we didn’t think about. We will try to keep our HMD as open as possible.
Being that you really can’t deny billions of dollars (in the name of researching and resources) how far are you guys willing to go with funding or investors, beyond the upcoming kick starter?
In hardware, building things is expensive. Of course we will need good funding if we want to bring to market the best possible VR experience. We have been approached by VCs and this will enable us to go forward. There is also some partnership in the talks that would make VR a better experience. We are at the beginning of something big and we need the collaboration of as much companies as possible.
Will you be going the same route as oculus did with their DevKits? Where from do you expect to have the money from to preproduce 10k units and ship them worldwide? Kickstarter? What if you want to ramp up production to lets say 200k units? Will you be taking VC?
We have different options on the table right now: private investors, crowdfunding, VCs, etc The goal is to deliver a great product while keeping control on our destiny. We will take the option that enables us to do that.
One of the problems the rift has is weight. You are adding even more features into the totem adding even more weight. Is that a problem you still need to tackle or do you feel it won’t be that big of deal?
I personally don’t have a problem with the weight of the DK. Most people I know either. I guess it is a personal preference. But we do have ideas in the pipeline to reduce the weight.
Just a suggestion: implement head tracking for the Xbox input, so that the screen isn’t stitched to your eyes, but rather fixed in worldspace. As an option of course.
Yes, this is possible.
Do you have any plans on connecting with mobile devices? If so, what is your strategy on that – overcoming the problems like using MHL and lack of USB ports? I think it might be very interesting because of the hardware acceleration for the warping on board – it frees up significant processing power.
For our Consumer Version, we are also planing to have android running in the HMD.
What kind of input are you planning to use for the onboard Android, considering a touchscreen is out of the question?
Nothing is out of the question, we want to Devs to experiment and give us feedback for the CV. Don’t forget that our DK will have 3 onboard buttons.
So its going to be a standalone device to develop for as well as a peripheral?
If there is Devs that want it!
Has price been announced and a giveaway for us poor folks?
If you buy it first on Kickstarter, you will have a early bird rebate Kickstarter this summer!
I’ve heard some negative expectations of the audio quality of the headphones, what would you have to say to settle these worries?
You can use your own high quality headphone since we use a standard stereo 3.5mm jack.
From your Twitter, you said that you are open to any and all crowd-funding sources, but have you considered crowd-investing? After the whole Oculus-Facebook debacle, a lot of people have become unhappy with crowd-sourcing, and would prefer to use crowd-investing so they could either prevent another thing like that sale, or at the very least, profit from it.
Unfortunately Crowd-Investing is illegal in Canada.
What is your stance on the implementation of advertisements and identity datamining into your product? Do you agree that “everyone has a price” in terms of acquisition by a large corporation? Do you believe in an “open” VR community? What does “open” (or not “open”) mean to you?
We want our device to be cross-platform, unbound to any specific marketplace, but it’s important that the VR experience is perfect. SteamVR can provide that and also console. There will be a lot more devices suited for VR and we will make sure that the VR experience is great on them too. We hate data mining as bad as most people do!
Could you share your vision about VR? Where do you want to get with the totem now and how do you see it going for the next 10 – 20 years?
Do you think you are different enough from the oculus rift to make a different impact on the VR world?
We want to provide a device that enables people to create what their minds can imagine. You will be able to do whatever you want in VR. It will be up the the users to decide what they want to experience. With great power comes great responsibilities!
For example, we are now a member of the Immersive Technology Association (ITA3D.com) and we want to collaborate with other companies from the industry. As a comparison, Oculus was a member but got out recently.
Do you agree that “everyone has a price” in terms of acquisition by a large corporation?
I personally don’t think it’s a price thing, it’s more about: can the other company that acquires us bring technical knowledge that we do not have and that will helps us make the best possible VR experience?
What is True Player Gear, and how will it change everything?
We want to be part on the future of mandkind.