Etee Finger-tracking VR Controller Kickstarter Concludes with Over $110K in Funding

145

London-based hardware startup TG0 launched a Kickstarter campaign earlier last month for the Etee dev kit, a VR controller with integrated SteamVR Tracking. TG0 positions the controller’s button-less design and finger tracking as its main attractions.

Update (May 11th, 2020): The Etee Kickstarter has concluded today, boasting 338 backers pledging a total of £90,650 ($112,000).

Surpassing £90K has unlocked the stretch goal for a formalized developer program, something creator Mick Lin says will include an event where devs will help the company expand Etee’s usecases.

The final funding amount fell just short of the £100K mark, which would have allowed the team to work on Android support.


Update (May 6th, 2020): The Etee Kickstarter campaign for a finger-tracking VR controller with SteamVR Tracking is now 153% funded with £70,450 (~$87,500), having well surpassed its goal of £45,900 (~$56,000). With four days left, the project is approaching the £80,000 (~$100K) mark.

TG0, the creators of the project, recently shared new gameplay footage with Half-Life: Alyx and Boneworks to show how the button-less controllers handle common gaming inputs like shooting guns and moving around the environment:

The project has reached two stretch goals so far—open-source CAD files for the controller’s handle, and an ‘EteeRoom’ demo game—and could reach its third stretch goal at £80,000 which will see the accelerated development of an Unreal Engine SDK for the controllers.


Update (April 14th, 2020): The Etee Kickstarter campaign has exceeded its goal of £45,900 (~$56,000) with nearly four weeks left. Among 190 backers for the controllers, 87% have backed the SteamVR version of the controllers, which is understandable as the base controller doesn’t offer integrated 6DOF tracking. Upon reaching the campaign goal, the creators revealed the first of several stretch goals which will be unlocked as more money is raised.

For the first stretch goal the company has announced that the LED surface on the controllers has two separate sensors which can serve as two touch-sensitive buttons, or a continuous thumb slider. It isn’t clear if this was something the company has added for the stretch goal or is merely explaining in more detail.

For the second stretch goal at £60,000, the company says it will make the available a 3D model of the controller’s handle so that makers and modders can create different handle shapes and accessories.

Four additional stretch goals, each £10,000 higher than the previous, have yet to be revealed.

Earlier this week the company showed off Half-Life: Alyx gameplay with the controllers and shared a Q&A.


Update (April 9th, 2020): One week after the launch of the Etee Kickstarter campaign, the project is nearing its £45,900 (~$56,000) funding goal, currently at 81%. TG0, the company behind the controllers, also announced that it has opened up the campaign to worldwide shipping, allowing developers from all over to back the project.

While the campaign looks on track to reach and exceed its goal, $50,000 doesn’t seem like much for a hardware project of this scope, which likely requires bespoke parts and precise manufacturing. Still, we hope to see the campaign succeed, as more controller options means more choice for developers and potentially for consumers further down the road.

Original Article (April 2nd, 2020): TG0 has launched the Etee Kickstarter campaign with the hopes of raising £45,900 (~$56,000) for its VR controller dev kit. The Etee dev kit will come in two versions, one with 3DOF tracking starting at £200 (~$265) for a pair and another with 6DOF SteamVR Tracking starting at £240 (~$315) for a pair. There’s a range of higher tiers available with more unique rewards, including a version with advanced haptics and another with a cool transparent shell.

Based on a thesis that sounds… downright wrong to the ears of any VR gamer, TG0 says that “buttons are way out of date,” and touts the Etee controller’s button-less, trigger-less, and joystick-less design as its major selling point, alongside finger-sensing, which the company says detects proximity, touch, and pressure.

This is in contrast to the rest of the VR industry which has steadily coalesced around VR controllers and games which make use of use of buttons, triggers, and joysticks for key gaming interactions. TG0 says that Etee supports gestures which can be used in place of buttons.

While removing the reliance on binary controls sounds great on paper, in practice it has proven difficult in the VR gaming space because of the need for precise and highly reliable inputs.

But VR gaming isn’t the only use-case the company is touting for the Etee controllers. As a dev kit, they could of course be used for any application where motion input is useful. Indeed, VR content that doesn’t demand the binary precision of hardcore game experiences—like training, art, therapy, social, remote control, and more—could definitely leverage Etee as a more intuitive means of input than a VR controller covered in unfamiliar buttons, triggers, and sticks. We’ve made a similar point about Oculus Quest’s experimental hand-tracking feature (which of course also lacks buttons, triggers, and sticks).

Image courtesy TG0

Though the controllers are a dev kit at this stage, thanks to integration with the SteamVR Input system, the Etee controllers should be technically compatible with SteamVR games out of the box, though we’d expect the need to experiment with custom bindings for many games to reach a point where things are truly playable with the controllers given the need to remap buttons and other controls to Etee’s unique inputs.

Image courtesy TG0

TG0 says developers can expect battery life up to 6 hours of continuous use and 14 hours of standby, and that the 3DOF version of the controller weighs 75 grams and the SteamVR Tracking version weighs 120 grams.

The Etee Kickstarter campaign runs until May 11th and the company expects the first ‘Early Bird’ controllers will begin shipping in December 2020.

This article may contain affiliate links. If you click an affiliate link and buy a product we may receive a small commission which helps support the publication. See here for more information.


  • manhunt0r

    sigh, another genius project that assumes gaming can ever work well without the physical feedback of buttons and thumbsticks. im sure This time it will totally work.

    • Jan Ciger

      Well, there are plenty of non-gaming apps where this could be useful but otherwise I do agree with you.

      This design is just silly – “so we give you a controller that you have to hold in your hand but you can’t really use it as a controller …” all the while the competing designs either have both controller and (some) finger tracking (Index, Touch) or do finger tracking using cameras so that you have free hands (Quest, several Vive versions, Leap Motion …).

      This is literally combining the worst of both input systems – obstructed hands and lack of buttons/joysticks that will make mainly navigation difficult and non-intuitive. Humans aren’t used to moving using waving their hands and fingers and walking often isn’t a practical option either due to the distances that need to be covered or because the user doesn’t have so much space available.

      That doesn’t mean that we can’t make finger-tracked hands work (heck, data gloves have been in use much longer than gamepad-style controllers with joysticks) but for something that aspires to be a consumer product it is a silly choice to declare that “buttons are obsolete” when 90% of the apps on the market aren’t designed to have menus and buttons poked with bare fingers … Oculus and others know well why they have labeled the finger tracking as experimental.

      • Immersive Computing

        I’m very interested to try this new input device, finger tracking has been very cool so far…

        https://uploads.disquscdn.com/images/5a931d45cbc14a7042c9a398d3f7b579c0ddce0af48412bd53906511f01d7eef.jpg

        • Jerald Doerr

          Sweeviver 2.0!!

        • mepy

          Have you tried typing on a virtual keyboard? If so how is the experience? Difficult to type in the open air? Have you tried typing on a surface, soft or hard, with a virtual keyboard overlay?

          • Andrew Jakobs

            I already hate typing on my mobile onscreen keyboard which had no tactile feedback.

          • Zantetsu

            Do you prefer having no typing option altogether?

            Sounds like that’s what you are advocating.

          • Andrew Jakobs

            What kind of stupid question is that? the onscreen keyboard is a necessity, but would be much better if it had decent tactile feedback. Why create a controller which doesn’t have tactile feedback to replace something that does?

          • Zantetsu

            It’s not a stupid question. How are you going to provide tactile feedback while virtually typing? Buzzing the controller?

            I think you will hate that much more than you hate typing on your mobile onscreen keyboard because at least with the keyboard you have direct physical feedback of touching the surface.

            Basically you will always be unhappy with any virtual keyboard given what you have said.

            Therefore your criticism that this particular implementation would be unsatisfying to you is meaningless as this would be true of all solutions aside from actual keyboards that could be represented in VR.

            So if you want a keyboard with tracking, that’s fine. But that’s like not even in the universe of talking about normal controllers, which is what this discussion is about.

      • Jistuce

        Oculus and Valve have also both reviewed usage and determined that buttons and thumbsticks were essential for actions that can’t be intuitively mapped to natural hand gestures.

        A lot of thought and usage study went into the original Touch controllers, and a lot of the same thought and study went into the Index controllers and arrived at the same conclusions. The conventional controls aren’t actually there because “it is a dualshock, but now with motion tracking”, no matter how hard the “natural hand movements only” crowd insists otherwise. They’re there because… conventional controls do things that finger gestures don’t.

        • Jakub Kamecki

          I know it’s hard to give the benefit of the doubt to an unknown startup. Consider that Valve Index and Oculus have been in development for what 4-5 years at least. It’s not outside the realm of possibility that there’s a lot of technological debt there.

          Our company was founder in 2015. And we took this long to bring the technology to a wider audience. Capacitive sensing and flex based sensing, thumbsticks was all the world knew back then.

          Who is to say that you won’t see this approach pop up in other places as we establish ourselves more? You already have Sony patenting something very similar – although with much worse design and likely not as robust finger sensing as ours.

          https://www.dailymail.co.uk/sciencetech/article-8038731/Sony-patents-new-motion-sensing-controller-hints-new-PlayStation-VR-headset.html

          • Zantetsu

            Your patience with these negative nancys is commendable.

      • Jakub Kamecki

        Have you seen the movie Moneyball? There’s a perfect quote there that describes the situation at hand (no pun intended).
        https://www.youtube.com/watch?v=D59Oh8sQFSM

        We are trying something new and it’s different. We are inviting the community to join us and help us improve the product. There are some thing that we are not ready to do just now – but we listen.

        For instance we wont add a stick to this iteration. But we may split the LED touch lens into two for one more additional control surface. We are also considering to add an additional electrode to the side of the tube where your thumb can rest – so that you have one more new place to form a fist.

        If we are successful with this version we can always build a gaming dedicated version.

    • mirak

      thumbstick are over rated

      • manhunt0r

        nah, tp movement is utter garbage and what held alyx back, among other outdated trash for vr freshblood.

        • anarfox

          You do know Alyx have smooth locomotion?

          • manhunt0r

            have you played the game? i finished it. it was 100% designed around tp movement, you quite literally can not finish the game without teleporting.

    • mepy

      I don’t think they are assuming their devices will be primarily used for gaming.

      • Jerald Doerr

        You don’t? I would think so because 85% of the video consists of showing the device used for gaming. Guess their marketing department has no idea what there doing?

        • Jakub Kamecki

          We see ourselves as a general purpose VR controller that removes distraction, increases immersion and is a more accessible experience.

          We are using the kickstarter to gather feedback, judge product market fit and ask the community to join us. We went with gaming because it’s easy to showcase and VR gamers are by definition technology early adopters.

          Only recently we hired a marketing manager and she’s helping us clean up the messaging. Up till like two weeks ago I was the sole marketing person. We built the campaign essentially completely in house. We are a small team and want to put as much of our limited resources into the product.

          Thank you for your consideration and support.

    • Jakub Kamecki

      It could. :)

      This is not our first product. We are building devices for our b2b partners using our patented smart materials. We would be out of business right quick if the technology doesn’t sell.

      etee is our self initiated project because we wanted to get closer to the customer and not only work from the back stage on other peoples products.

  • Tony_Neville

    I’d like to see if a user can easily walk/rrun in a figure 8 in virtual space using these.

    • Jakub Kamecki

      Can you suggest a game that would you’d be happy with? I can ask the team to put this together.

      • Tony_Neville

        I’d be happy with Fallout 4 VR. I have an HTC Vive and use the left touchpad to strafe/move forward and backward. The right touchpad is used to rotate. It’s all very smooth. There is no teleporting. I love that the movement can be setup this way.

      • Tony_Neville

        I forgot to ask, how well do the Etee controllers work with Half-LIfe: Alyx?
        I know Etee must have tried them on that game as the temptation would be far beyond what any mortal can bare.

        • Jakub Kamecki

          very good. :) we have already two episodes of content recorded as we go through the tutorial – im actually now watching our implementation of the grabbity gloves. i hope to start releasing them this friday.

          our backers have early access to the content on our backer discord.

  • Jistuce

    This looks like a device designed to separate fools from their money. Promise something amazing, poopoo established technology as inferior by virtue of existing, gloss over the fact that you use decades-old technology too, and make sure your prototype looks like it just came off the Apple assembly line.

    Also, it looks like they decided “let’s clone Knuckles, but do it in a sexier and less functional package”, and skipped buttons for aesthetics and then justified it later by claiming they were old and outdated and it takes courage to remove buttons from the design.
    That’s really the biggest problem here, that these look like a budget clone of the Index controllers. I grudgingly admit that they actually MAKE SENSE in a world where the Index controllers don’t exist. But we live in a world where they do, so the Etees are just a less ergonomic set of Index controllers with more limited functionality.

    • Jakub Kamecki

      Thanks for the feedback. We are not criticising existing technology just because it’s existing. Rather we try to compare ourselves to existing solutions that people know. We just strongly believe we are better. Hope we can change your mind.

      I have the deepest respect for Valve and what they are doing and how they are creating and opening up their ecosystem to smallish players like us.

      If you follow the conversation we have the opposite problem – because we are a small company, completely challenging how things are done we are being dismissed because we don’t conform to the status quo and don’t have a massive marketing budget to drastically shift the zeitgeist on controllers.

      • Jistuce

        I believe you are mistaken, but look forward to being proven wrong.

        • Jakub Kamecki

          Cheers, brother! That’s all I can ask for. An open mind. We’ll do our best to meet and exceed your expectations.

      • Charles

        Why not VR gloves? They’re the natural and inevitable solution. I tried a pair in a robotics lab 11 years ago and they worked perfectly. Maybe just add low-profile buttons and joysticks in strategic spots and make them visible in-game.

        • Jakub Kamecki

          Hi Charles, in my view gloves have serious limitations.

          * not one size fits all
          * prone to tearing and breakage
          * not washable
          * cant be used in environment with gloves already on – cold climates, military
          * enterprise level – very expensive
          * uncomfortable to be used in hot climates
          * difficult to put on and off – cant switch between users easily
          * hygiene issues between users

          Some of these are less of an issue if you are a single user at home. etee solves a lot of these and most importantly is much more accessible from a price perspective.

        • Products are only vaiable if there is certain amount of users willing to pay for it. Now VR is small, so every company that would target consumers wanting gloves would fail. We have to wait untill VR is mainstream, so there will be more users overall including users wanting gloves.

      • Dan DeMontmorency

        I think a good compromise would be to use an extensible design to have a trigger add on idea similar to idea of Psmove/wii and Vive trackers ie the vr rifle. Nice to see some great ideas to push the limits outside of the standard boxes. Keep up the good work.

        If you would consider join the community.openmr forum. Heliosurge

    • Zantetsu

      I don’t like my index controllers. I find them uncomfortable and awkward, and the finger tracking is marginal at best. I would be interested i these Etee controllers if they are more comfortable than index controllers, although I am a little skeptical of controllers without triggers as I find games involving shooting in VR generally the most fun and triggers are super important there.

  • Matteo Valles

    Looks like it would be good for guitar hero?

    • Jakub Kamecki

      We are hearing good feedback / interest from Beat Saber community. Social experiences like NeosVR VRChat are also great with etee – you can dance, freely interact with people using gestures. They are light an un-intrusive.

      You are also seeing an explosion in VR telepresence, conferencing, remote work. These solutions require ease of use. You cant use a keyboard or a mouse with current controllers. With etee you can.

  • Andrew Jakobs

    Nice, for some type of games, but I just like tactical feel, so real buttons etc. I just don’t like not feeling a trigger when I’m shooting. As I said it can be great for some type of games, just not all..

    • Jakub Kamecki

      Appreciate your feedback. And you are right. We see ourselves as a more general and accessible controller. That removes distraction from VR. We certainly acknowledge that there may be issues with this approach, especially for backwards compatibility because existing games have been designed for a different interaction paradigm.

      But we are also working on providing our own bindings for those games to make the experience as seamless and true to our vision of immersion as possible. We will also continue working with the community on incorporating their feedback and developers to provide native support.

      If the campaign is successful it will be much easier for us to build that.

  • I have a devkit that should review on my blog next week. Until I try them, I don’t express an opinion… for sure the fact that are 3dof is their limit, IMHO

    • Jakub Kamecki

      Looking forward to the review, Tony. We acknowledge that limitation fully, that’s why we are kickstarting a fully SteamVR compatible version.

  • Jerald Doerr

    That’s pretty cool… Just missing analog stick where the thumb is and PS 4 type L 1 and 2 analog triggers… Would be awesome!!

    • Jakub Kamecki

      Our thumbpad fully emulates a stick. The pressure sensing technology we offer is fully analog so in effect you have 8 LR triggers, 4 on each hand. if we exclude the index finger as a proper trigger with a set of already existing interactions you are left with 6 triggers to be used for bindings and interaction.

      • Jerald Doerr

        Yeah, I could see that… But I’m way over trackpads as a thumbstick alternative.

  • Jerald Doerr

    Lol!!! That’s funny! To top it off 3DOF doesn’t drift? Hahahaha

    • Jakub Kamecki

      Of course it drifts. That’s why we are releasing a 6DOF SteamVR compatible controller with our own designer tracking module. Appreciate your support.

  • Jistuce

    Valve bought a bad batch of thumbsticks, therefore all thumbsticks are bad. Any perceived advantages of a thumbstick are actually detriments.

    The past is holding us back from reinventing the wheel, only maybe with corners and flat surfaces so it is easier to manufacture.

    • Jakub Kamecki

      Has Nintendo Switch also bought a bad batch? Thumbsticks are a notoriously hard piece of component to get right. In gaming especially they undergo massive stress and are prone to breaking.

      • Andrew Jakobs

        Yeah, but that’s due to the nature of the beast, it isn’t actually that hard to do good thumbsticks but those cost a lot more.. valve tried it with their steam controller, but it wasn’t a succes. People just want tactile feedback. As i mentioned before i also like to feel an actual trigger when i pull one, not just act in the air like i’m pulling an imaginary trigger.

        • Jakub Kamecki

          Fair enough. I just ask to keep an open mind about us. :) Appreciate your consideration.

      • Jistuce

        I believe the answer is actually “yes, Nintendo bought a bad batch.”
        Or possibly “Nintendo once again special-ordered a fundamentally-flawed design” like with the N64 stick module(where they took a very robust design, and created a remarkably unreliable device through a change of materials, and thousands of failures that could’ve been mitigated with the slightest application of lubricant).

        I’ve had very few thumbstick failures over the twenty years since they became popular. I believe I can score it at a grand total of one, actually. PSP, and the analog slider module physically disconnected from main board, but the actual control was undamaged. Hell, my PS1 DualShock from 1999 still works perfectly despite a dog trying to eat it.

        If we include devices which would NOW be recognized as thumbsticks, but weren’t in their own time(the term not yet existing)… my Vectrex’s mini-joystick from 1982 still works fine. I’ve had to open the controller to work on the buttons, but the analog stick module has endured almost forty years maintenance-free.

        I’ve actually seen more failure of solid-state components than thumbsticks.
        This seems to fly in the face of the argument that they are fundamentally unreliable, fragile, and prone to failure.

        • Jakub Kamecki

          Some thumbsticks are fine. But if you have a chance to talk with anyone who designs controllers for the big three and ask them whats the most single point of failure. It’s going to be the stick.

          Regarding how our technology is different. We transform everyday polymers into capacitive surfaces via carbon black additives. We then through machine learning, signal processing and pattern recognition identify the change in the signal and interpret the output. It’s not a traditional approach to the problem as you know from Vive or Steam Controller trackpads.

          Not only is the tech cheaper because only plastic is used, it’s more sustainable and lighter, more sweat resistant.

          But onto the really exciting bit. Because it’s fundamentally an analog technology we can do very cool things with it – dynamic baselining and calibration of sensitivity for one. We can account for changes in temperature, humidity or even damage to the surface either accidental or through extended use. Another thing is that the underlying algorithm can always be updated as more people use the device and the data set grows.

          Why this is better than a stick? If a stick fails you are SOL very few people can service it themselves. Our technology provides same level of performance (plus Z pressure sensitivity) but is much more robust and long lasting.

      • Ace of Spades

        Im 40 years old, been gaming on consoles and PC all my life and still do, I never and I mean not a single time in my life had Thumbstick problems with Branded gamepads, i never buy others so i dont know whats going on there.

        But lets says all the x360 gamepads that I had xbox one, Dual Shock 3 and Dual Shock 4, even GameSir [a quality Chinese gamepads brand], I never had no issues with thumbstikcs or buttons.

        Playing a game without holding thumbsticks and pressing buttons is unrealistic and will be a torture.

        Look at Rift S and Valve controllers, IMO thats how VR gamepads should be, there is no need to fix whats not broken.
        As far as Finger gestures it can be added to the existing controller, and its not really needed.

        When I used the new Rift S controllers i never said to myself that I want to cut them down or add something that they dont already have for VR.
        As far as feeling and immersion I would like to see what Sony did with new PS5 Dual Sense gamepad: Force Resistive Triggers and Haptic Feedback.
        These 2 will make it perfect.

        • Jakub Kamecki

          Currently both Switch and Index have stick issues. Previously Dreamcast and N64 had stick issues. I’m sure I can look for other examples. I believe either the Vita or PSP had issues, with a similar technology. It’s a notoriously hard engineering problem to solve in gaming.

          And if all you do is game that’s fine – go for the Index. With etee we are aiming for other use cases too. We want to bring VR to a broader audience than just gamers.

        • Your experience with good thumbsticks means really nothing because it’s an nanecdotal argument. This controllers aren’t designed only for you, they are designed for wide range of users, and I can ensure you, not everyone is delicate with thier devices. It’s an interesting thought to do a research on it, but when comparing big companies constantly failing to design good thumbstick to your anecdocal argument… Also, in electronics world counterfits are really a big problem in supply chain, it’s complicated.

  • mepy

    Sigh, I miss punctuation and capital letters.

    • Jakub Kamecki

      My apologies. You are right. I will try to do better. :)

  • mepy

    For virtual desktops it makes sense to have something light and robust that uses our more intuitive movements with hand gestures. Controllers are better suited for gaming. Eventually though we will get good ergonomic haptic feedback gloves such as something like what HaptX is developing.

    • Jerald Doerr

      mepy, I mean Mepy. Ever use an Air Mouse? Works perfect and $20.

      • mepy

        Interesting, the Etee (and other finger tracking VR controllers) could be adapted to be used as Smart-TV remote controllers also.

        • Jakub Kamecki

          Yes! you get it. :)

          We had to focus on one message for the kickstarter. But there is a reward tier in there for a one hand or a two hand package without SteamVR. It can be used to control 2D games, drones, IoT and as you say smart home devices. Perhaps some sort of game with Alexa?

    • Jakub Kamecki

      Not only virtual desktops. We see etee as a general purpose controller that can broaden VR/AR use to non-gamers, a more casual non-techy crowd.

      Regarding gloves. They have their own issues. Hard to make one size fits all. Difficult to take on and off between users. Not possible to use in an environment where you are wearing gloves already (hot, cold climates, military use for example). Hygiene issues due to sweat and hard to wash. Prone to breakage and tearing. Tend to be quite expensive from what I’ve seen.

      We are trying to address these glove limitations with our form factor.

  • Jakub Kamecki

    Is it not valid? The Valve Index stick issues are fairly well established.

  • Frank Chevalier

    No stick and trigger, im out for this one. No future in
    competitive/very competitive VR shooter. I much prefer a physical
    feeling than tactile for immersion and precision

    – fire: A weapon need a trigger for fire, so a trigger on a controller

    -moving: stick is a lot more predictable and accurate. When you are
    under stress, you dont play smooth so you’ll probably click or do some
    bad movement accidently with tactile

    – reloading: rejecting mag on a gun is done by pressing a physical button

    A stick can be reliable if done correctly. Not like the valve index
    controller but more like the CV1 controller. My stick still pretty good
    after 500 hours of hard play

    If you make a controller for shooter, it will fit on all games. And that will attract more players. Its a fact

    • Jakub Kamecki

      Competitive scene was never our target audience. I’ve addressed the other points in other places.

      Summing up we will consider releasing a more gamer specific version if this does reasonably well.

      • Frank Chevalier

        What the point to focus non-gamer audience when the main one is gamer?

        You did good work and we feel your motivation. I just want to understand

        • Jakub Kamecki

          Here in lays our strategy. We want to address a market beyond gamers. We are not really competing with the Index. For gaming it’s a pretty well put together controller, barring the stick issues. But for a more casual user ours is the better choice.

          Also competing with Valve with our scale? it’s a recipe for disaster. Better to carve out a niche and be excellent at it.

          • Frank Chevalier

            The next guys who will do a no faulty index controller like will be millionnaire. Put a reliable stick, change a bit the design for make thumb to fall better on stick and buttons(like the booster tweak do), switch the stick in the middle and trackpad at the right and that will be perfect! Or just making a cv1 touch controller like can be perfect too. We just need something like that compatible with lighthouse sensor

            In short, good luck jakub!

  • kontis

    When I first saw these controllers, the first potential issue that came to my mind was that curling fingers will make the software think that you are twisting the wrist, because contrary to Knuckles this one sits on your fingers.

    Funny thing, the first shot with controller and screen shows that exact problem I thought of… ;)

    I’m afraid the trade-offs they chose might not be the optimal ones. They will benefit from simplicity and maybe lower friction of use, but perhaps at too high cost of functionality.

    • Jakub Kamecki

      Interesting observation. That’s not an issue I’ve personally observed with the controllers. But yes, you are right they sit differently than the index on your hand.

  • Trenix

    For gaming, we need buttons. Unless you can feel the objects and the push back, it wont work. Might be neat for a few games and software, but not for most.

    • Jakub Kamecki

      Technology progresses and changes, things are invented – some are good others are bad and fade into obscurity.

      Naturally we believe that our solution has the potential to be the future of tactile interaction, and we are asking the community to judge if we are right or wrong by supporting our campaign.

      You may disagree and that’s your right, but let’s see who will be proven right by history.

      Once we needed buttons for phones, too. Earlier than that a rotary dial. In the near future you won’t even have a phone, but you will conduct your business via glasses, wearables or other implants with time.

      Computer gaming is a fairly new invention, VR even newer. The existing fascination with buttons is technological baggage since at least the 80s or even earlier.

      Personally I’ve played on all major gaming systems starting with the Atari, C64 and Amiga. My first own console was the Dreamcast, but I spent a lot of time on my friends NES and later SNES. The fundamental concept of these controllers have not changed.

      Is it not time for something new? At least an attempt at something new and better.

      • Trenix

        Just because it’s new, doesn’t mean it’s better. We went from indestructible flip phones to ones that shatter in a matter of seconds and can’t even read our inputs correctly. Hand gesture suck, plenty of TVs had them which resulted in accidental readings. Kinect has also failed for Xbox. Voice recognition is also a mess. I’m not saying to not move forward, but first work with what you have and improve what’s missing.

        Plenty of VR games have it where you grab a certain area in your body and end up having something else pull up. That’s how gestures work for you. They simply don’t, unless you can “feel” what you’re doing, it wont ever work as intended. The vast majority of games won’t support a device that relies on hand gestures, be my guest and prove me wrong.

        • Jakub Kamecki

          And we are now back to flip phones again, but now with screens. :)

          Our technology has been proven with our b2b customers. We are now bringing it to the b2c customer. We are confident that it will work. But as with any invention there are bugs that we will work to iron out.

          I’ve always said that a huge risk is developer support. We are not Valve or Oculus, don’t have that scale. We depend on the community to gather around us and join us on this journey.

          • Trenix

            So listen to your community, buttons are required. I at least give you a bit more feedback, maybe once you provide some push back, would these kind of devices shine.

          • Jakub Kamecki

            Unless we met at CES, EGX or CEATEC recently one hasn’t had a chance to check out etee in their own hands. Very few people have a legitimate basis for comparison how etee feels vs traditional button-based controllers.

            We always listen to the community, and when the product releases we can update it later based on informed feedback.

            What you are saying now is essentially after looking at the first iPod – “click wheel? pff no buttons like the walkman, too small, won’t buy” I know it’s silly because hindsight is 20-20 but you are witnessing a point in history where a new control paradigm may be created.

            You may not believe me that we will be successful and that’s fine. There are users that believe in what we are trying to do and are prepared to give us a chance – if we create a community around these early adopters the second wave will become more easily convinced with actual humans using them in real life.

      • Blaexe

        Once we needed buttons for phones, too.

        And still, serious gaming done on mobile screens similar to phones uses physical buttons. The Switch is basically smartphone with…buttons.

        I can see your kind of controller being used in some special scenarios, but not gaming.

        Using a smartphone is not the same as gaming with a smartphone, we have to differentiate here.

        • Jakub Kamecki

          etee is fundamentally the same freeing experience as the Oculus camera hand tracking with the benefit of full freedom of motion outside of the camera FOV enabling fitness uses, dancing, social VR experiences and other. Plus additionally the benefits of haptics and pressure sensing, which is not possible with camera based tech yet.

          Have you heard anyone say that camera tracking is not suitable for gaming because of lack of buttons? I haven’t. You know why? because they are Facebook and have millions of dollars in R&D spend and a track record, backed by further millions in marketing spend and can push the narrative that camera tracking is the end all be all of AR/VR interactions. Which it clearly isn’t as it has serious limitations.

          Can camera tracking be used in games? Of course it can. Perhaps not in hardcore competitive scenarios like Pavlov but it can and is daily, in games and other apps.

          We do the same basic thing but better and yet the community can’t accept that. Do you think that this is a reasonable and fair attitude to a startup that is trying to innovate with a fraction of the resources of Facebook?

          • Blaexe

            Have you heard anyone say that camera tracking is not suitable for gaming because of lack of buttons?

            Seriously? Yes! That’s 100% consensus within the community. Quests camera based hand tracking is interesting and nice for UI interaction and social apps but definitely not for gaming. Not even for casual gaming.

            Your view of the community is fundamentally flawed here and that is honestly surprising and a bit worrying.

          • Jakub Kamecki

            And yet they are games and experiences being developed for that interaction paradigm daily with reddit and social media being filled with developers pushing the boundaries of hand interactions showcasing their work.

            etee offers a bridge between button based controllers and the completely lifeless, tactile-less, disembodied experience of camera tracking. We are giving the “button” community a way to experience extremely satisfying hand tracking with button-like functionality. What’s not to like?

            Games will catch up as developers will start to understand better how to develop games for this new experience paradigm.

          • Blaexe

            And yet they are games and experiences being developed for that interaction paradigm daily with reddit and social media being filled with developers pushing the boundaries of hand interactions showcasing their work.

            Because it’s the “new and exciting shit” and makes sense in some scenarios. Notice I did say the same about your controllers.

            What’s not to like?

            The lack of haptic feedback. You’ve heard the feedback of the gaming community loud and clear when looking at the comments.

            I wish you all the best, but I see zero chance of success within the gaming (!) community and zero chance for widespread developer support.

          • Jakub Kamecki

            Fair enough. I think we agree to a certain point.

            We do offer haptics btw, just not in the form of actual buttons. We have a haptic engine and a patterned surface that informs your finger movement.

          • mepy

            The Quest finger tracking is done with the headset cameras, not from the controllers. You are comparing apples and bananas.

          • Jakub Kamecki

            Tell me about it! :)

            That’s not my comparison. If you follow the conversation about etee on social media one of the two comments is “no better than index” and “quest already does. this useless. would not buy.”

          • Blaexe

            It wasn’t me who came up with the comparison, it was literally the developer of the controllers. So maybe tell him instead?

          • Jakub Kamecki

            I was addressing the common consensus amongst from the critical voices. Not one that I myself created or even agree with in the first place.

            Besides the VR space will explode in the next couple of years and will be big enough for different input methods.

          • Blaexe

            I don’t want to go into this discussion again as I don’t think it’s useful, but you did mention the Quests hand tracking first.

            etee is fundamentally the same freeing experience as the Oculus camera hand tracking with the benefit of full freedom of motion

            I didn’t mention Quest at all beforehand as I don’t think it’s comparable in the first place. Some haptic feedback (like in your case) is better than not haptic feedback after all. And I don’t mean HD vibration.

          • Jakub Kamecki

            And I’m agreeing with you. It’s my bad. I’ve been dealing with the same questions over and over and it tends to blur together. I shouldn’t have brought it into the discussion, if you have not mentioned it.

  • Jakub Kamecki

    You can do sign language at your TV with just the one hand version starting at 129 GBP.

  • CQCoder

    Solution in search of a problem. Doesn’t appear to bring anything new to the table. The idea that buttons/thumbsticks are ‘bad’ is…silly. Pretty sure the thumb sticks on my Xbox Controller work just fine (and I love my Index Controllers) No physical feedback..that sounds fantastic.

    • Jakub Kamecki

      Innovation often feels silly until it doesnt. There are real reasons why buttons are a distraction for VR and inhibit full immersion.

      • Ace of Spades

        But a solution to the problem you described is Hand Gesture controls, is to do everything without holding anything, its not adding another stick in peoples hands.
        Is for cameras to see your hands and copy your movements to the virtual world

        • Jakub Kamecki

          Cameras are limited to camera FOV. Current implementations require you to have your hands in front of your face. Which is immersion breaking.

          Most of the arguments against etee is that we lack buttons and a trigger. Cameras still have that problem.

          etee actually offers you a physical object that you can interact with. provides full spherical freedom of movement enabling dancing, sports, presentation and social interaction beyond just looking at your hands.

          Cameras also don’t provide haptic feedback.

  • Oh where did all of the touchpad fan-boys go?

    Back in 2016, I said Valve’s decision to use touchpads instead of joysticks was a stupid idea… and, on this very site, I was ganged up by touchpad fanboys telling me how very wrong I was. They said touchpads were more natural and gave greater controller flexibility.

    I told them you can’t SEE your hands. You need joysticks and buttons because they give real feedback, before and after you use them. We use those controls because, out of decades of evolution, duel thumbsticks and buttons have proven themselves to be THE BEST means to control a game. Touchpads will never touch that ease of use or dependability.

    Thank god Lucky Palmer felt the same. Gabe Newell was an idiot for pushing them on every Steam device he could. Touchpads in VR suck! They always have and always will.

    • Jakub Kamecki

      I’ve never been so conflicted about a reply before. :D You make some good points, that support my view, but then make also ones that go counter to it.

      In regard to touchpads – you will always get supporters of big companies, it’s easier to believe that they know what they are doing, because they have more resource. But also then they can’t really push the envelope too much. Look at the recent announcements from PS or Xbox – the controllers are unchanged for generation. Perhaps it’s because we reached peak controller, and there’s no improvement to be made. But I kind of doubt it. Thats just me though.

      Coming over to the seeing hands part – if the surface of the touch pad is patterned and gives tactile feedback, like it is in our implementation that argument goes out the window. You can easily find our touch pad and navigate with it.

      The very principle of not being able to see our hands has been at the core of the design direction for etee. When you cant see your hands buttons become confusing and detract from the immersive experience that should be the ultimate goal in VR.

      Coming back to innovation and thumbstick being the superior control option. I just disagree, plainly, can’t disagree more actually. Our implementation of the thumbpad is far superior to any stick or pad you used up till today. It’s more robust, does not have drift issues and can be mapped to other interactions than just simple movement.

      • Ace of Spades

        Im sorry but why the obsession of changing what works?
        Lets replace steering wheels with sticks like in Tanks?

        Look at PS5 Dual Sense, what they did is what I wanted from them to do, build on top of what already works and make it better, Resistive Triggers and Haptic Feedback will make games more fun and feel immersive.
        Next Step [expensive right now] is Resistive/Force Feedback Thumbsticks, like on car simulator steering wheels.
        Also ill take that Sony touch pad and make it OLED screen, even B&W if colored costs a lot, will be nice for messages, get notifications and changing game options without quitting gameplay, and adding virtual buttons, right now its swipes that you need to remember but if they had a screen they can just show on screen, Map, Settings, whatever.
        Another thing they can do [maybe they already did] is using Bluetooth with aptX HD or WiFi for connection, to make the sound from controller headphone port higher quality and lower latency.
        All the rest like RGB buttons for people that need to see at night and paddles on the pack for extra buttons are nice to have.

        • Jakub Kamecki

          > Lets replace steering wheels with sticks like in Tanks?

          Steering wheels are being replaced in cars as we speak, removed even.

          Do we really need to be putting more stuff on the landfills that has outdated it’s usefulness if there’s better technology available?

          > Also ill take that Sony touch pad and make it OLED screen, even B&W if colored costs a lot, will be nice for messages, get notifications and changing game options without quitting gameplay, and adding virtual buttons, right now its swipes that you need to remember but if they had a screen they can just show on screen

          Do you remember what was the initial reaction about the Touch pad on those controllers? It was universally criticized and very rarely used in any meaningful way except by 1st party developers.

          What we are offering is an innovation that used by developers could open up new interactions in games, new ways of playing. Something that is not possible in this current form factor. Is that really such a bad thing? There’s no telling what a talented developer can do with the pressure sensing we offer or individual finger sensing.

    • NooYawker

      Philosophically i thought touchpads would be more immersive for VR but the reality is it just sucks. It was a huge difference upgrading from the Vive controllers to the Index controllers.

  • anonmon

    I know this is beating the point over the head by this point, but you need sticks.

    It’s not “innovation” given ho many people before have done the same, it’s not “something to get used to” given how many times they failed, it’s pure practical realism. Beyond omnidirectional treadmills there aren’t any real practical ways to move around a virtual space without some input method for it, and no matter how much “innovators” try to push it (This is not the first time, and won’t be the last), analog sticks are the best way to do it.

    Why? The user knows exactly what input they’re putting into the system, at all times. With any touch surface of any kind, there’s ambiguity with what exactly you’re telling the computer, because there’s zero resistance or feedback. (And no, buzzing the surface with vibration “engines” or whatnot is in absolutely NO way any kind of replacement, these are purely dev defined and have zero bearing on what is physically ACTUALLY happening, even if it wasn’t absolutely arbitrary) You know, at all times, what input you’re giving the computer because you physically know at what angle that stick is pointing. And just as importantly, you can let go of the stick and it goes right back to neutral, so even if and when in the event the user looses track of what the stick is doing, they can let go and reset with ease. The closest you can do with a touch surface is to have a nub of some kind in the middle of the sensed area, which I shouldn’t have to state why having to touch the surface that’s actively tracking where it’s being touched to find neutral is a bad idea.

    Same with buttons, you can either press it or not press it, you know exactly when you press this button, and the button itself tells you whether its giving a input or not because of the difference in resistance against that finger changing between on/off. Zero ambiguity.

    You know why, despite how much Valve love their touchpads and had no intention of putting one on the Steam Controller originally put one in it anyway despite touting how much their touchpads “Could replicate the feel of sticks through the touchpads and rumble features”? (I own one, it does not in the slightest, as much as I love using the right touchpad for mouse duties, which I still have issues with when I need to know absolute position because, what do you know there’s no real way to know where your thumb is on the pad unless you hit the edge)

    You know why, despite being the company in gaming known for innovative input schemes, even at the expense of user usability at times, Nintendo ALWAYS has physical buttons, and sticks when applicable, alongside touchscreens, and screen pointers? Because they over decades of R&D know about as well as anyone can, there is no replacement for physical buttons for actually making a definitive input.

    You know why phones suck for games that require much more physical interaction than peggle and people go to great lengths to attach controllers to them for emulators and such? Because no buttons.

    You know why absolute every game for the Kinect was about as shallow as a kiddie pool with the interactions you had with any Kinect games, despite how much Microsoft was pushing it during the latter half of the Xbox 360’s lifetime? Because you straight up CAN’T. Every menu required you to hold a giant bubble square for X amount on time to make a decision, because there was no way for the system to know what the use wanted to do otherwise because of the massive ambiguity.

    You know why Power Glove like VR controllers weren’t the first thing to hit the modern VR market in 2016, despite that always being the Minority Report “VR dream”? Because it turns out that only waving your hands about in the air isn’t great for what you absolutely need to make an input into a computer because, oh look at how repetitive this is getting, there is too much ambiguity when you aren’t directly interacting with a physical object.

    Stuff like the Knuckles are absolutely wonderful for when you want to track your fingers and have interactions that don’t require the user to make a definitive input. But you can not have that at all times. You need menus. You need in game buttons. And as stated above, you need a way to move around in a virtual space. Anything that is not a real physical object moving in real physical space is too ambiguous to make as a reliable input method.

    I’m sure this has had plenty of R&D put into it to make it work as the people developing this want it to, and that’s great. Thing is, R&D around the world from the start of gaming have also asked the same questions this has as it’s been applicable, and people have always come to the same conclusion: You want definitive input? There is nothing better than a button.

    If you want to have experiences with these things that are largely mindless and don’t really ask much of the end user, then they’re great. Anything that requires definitive input, unless you want to laser-point everything, which gets clunky fast and not exactly practical for obvious reasons, you need something tangible.

    • Jakub Kamecki

      holy mother wall of text. nice write up!

      i think it’s fair to say that you thought about it and there’s nothing I can say to convince you. if we do our job right and the solution is what i say it is perhaps you will come about to our way of thinking.

      • anonmon

        Huh, guess “Too much text = spam” here.
        tl:dr, You need thumbsticks and buttons because there’s no ambiguity with what you’re telling the computer what to do, with a myriad of historic examples of instances where the same sort of thing has been tried and failed (Steam Controller, Kinect), or instances where it’s clear that despite a willingness to try the sort of thing these are doing, which at the end of the day sticks and buttons are clearly superior (Anything Nintendo has put out, the fact we didn’t get VR gloves in 2016 when modern VR kicked off).

        • Jakub Kamecki

          Not sure why it was removed. :( It was a good response, thought out and you put effort into it.

          Thumbsticks are only familiar to people that have been playing games forever. My pitch is that our approach doesn’t assume prior knowledge of thumbsticks to be used effective. In essence we aim for the casual user, non-gamer demographic.

          • mirak

            Thumbsticks of the Index are not even better than the touchpad of the Vive.
            I think valve fuckedup here by caving.

  • Ace of Spades

    Useless for most games, maybe good for some 3D work, but not for games.

    No Thumbsticks ,no Bottoms = NO GAME.
    Remind of the horrible Vive Shovels

    • Jakub Kamecki

      We are aiming to be a broad use case controller that is accessible to everyone. Gaming is just a part of the VR landscape. And we are banking that this will rapidly change and VR will expand into other areas.

      • Ace of Spades

        I wish you luck actually, I think your scene is more business oriented and 3D simulation.

        • Jakub Kamecki

          Thanks Ace, yeah, that very well could be. I think there’s a space for us too in the consumer market but for more immersive experiences like dance, fitness, tele-presence where ultra precise movements via buttons is not needed and it’s more about expressing the human form.

  • Rosko

    No easy task to pull off a new vr controller, i wish them great success.

    • Jakub Kamecki

      Thanks. Appreciate it.

  • brandon9271

    buttons and thumb sticks are necessary for gaming. Look at the Knuckles controllers. They have everything. Why but give folks options? I love my Steam controller but a Dpad is sorely missed for certain games.

    • Jakub Kamecki

      Our thumbpad emulates a dpad if you configure it that way. We will be releasing HLA videos tomorrow and on Sunday showing of that functionality.

      • brandon9271

        Hopefully it’s better than the Steam controller dpad emulation.. because it sucks. :) Not sure that a Dpad is particularly useful for VR anyhow.

    • mirak

      I still prefer the touchpad of the Vive than the joystick of the knuckle, because it’s easier to keep the finger position if you move the hand and your body.
      The joystick will receive too many undesired movements

    • mirak

      A patent from Valve has been spotted a few days ago about a Steam controller with swappable controls.

  • Jakub Kamecki

    Completely outside of the current argument – you have 2-3, Maybe 4 major players in the current market. Valve, Oculus, Microsoft and PSVR. They follow a similar path of innovation.

    We don’t have that baggage of massive install bases of users that would flip if they innovate too hard. We can push a bit further and yes, maybe create a bit of controversy in the process. Who knows maybe the industry will come out better off at the end of all of this.

    • brandon9271

      Well, I wasn’t just talking about the VR space. I was talking about gaming in general. Nintendo had the Wii, Sony have the PSmove and Playstation Eye and Microsoft did the Kinect. They were all innovative but they all went back to gamepads because that’s why the users want. VR is mostly the same I think. You can innovate but users like what they like. You have an uphill battle convincing anyone otherwise.

  • Ace of Spades

    BTW, I would like to thank Jakub Kamecki, for actually coming here and interacting with people, no many devs do it.
    I hope others will take note.

    • Jakub Kamecki

      Thanks Ace, means a lot. We stand and fall by the community. Although you may not agree with the direction we are taking we want to build something new and exciting and bring it to our consumers.

  • ¥DK¥

    Useless product. Offers nothing new over existing competitors :|
    There’s absolutely no reason to get this over Index Knuckles

    • Jakub Kamecki

      Smaller, less clunky, lighter and better finger tracking would be some of the key advantages that come to mind.

      • Zantetsu

        Those are ALL valid points Jakub and they are exactly the problems I have with my Index controllers.

        However – I also agree with people who want physical buttons/triggers for some types of experiences.

        I wonder — have you considered making some kind of physical buttons/triggers that could snap on to your controllers? Kind of like you have the base station tracker mounted on there. It could get kind of frankenstein but creative solutions to these types of problems is what makes for innovative products.

        Just an idea!

        • Jakub Kamecki

          It’s a valid idea. And thanks for your support!

          Priority for now is getting this version to be a success, once we can show that we delivered against our promises in this campaign more options will open up and we can look to other form factors and functionalities in future versions.

          If you see potential here, please consider supporting the campaign. We are very close to our goal and you will have a bigger voice in directing the development of the product.

          The lack of a physical trigger is not that big of a deal. It’s actually quite nice once you get past the initial difference.

          https://www.youtube.com/watch?v=f_oBVWuyMDM

    • mirak

      Knuckles are far from perfect.

      I believe there should have been two positions on the index, with one that don’t require your finger to rest on a trigger.

      And also a position where your thumb can pinch your index.

      Not all games require a trigger, and you can’t have your index tracked without having it on the trigger.

  • NooYawker

    It looks like they glued a vive tracker on top of a leap motion.

  • Mul

    what I am curious is how locomotion and buttons simulation will work with this. In their video it is not really clear.

    • Jakub Kamecki

      we’ve recently released HLA videos that explain it a bit better. tg0.eu/youtube

      • anonmon

        Watching this video, how do you deal with having to hover your thumb over the trackpad for extended periods of time? Doing that at length gets fatiguing pretty hard and fast. Sure, you could have a dead zone in the middle of the track pad, but between significantly reducing the effective tracking surface and making it basically impossible to hard define the hard line between “this is the safe area” and “You’re giving input now”, that doesn’t seem like a ideal way to do it.
        Plus hovering the thumb over the input means you don’t know absolutely where your thumb will hit unless it’s absolutely straight up and down from its current position. Raw muscle memory can only help so much when there’s no way to tell where exactly you’re touching without looking aside from hitting the edge.
        I presume the problem of the last point is what the dip in the middle of the trackpads are supposed to alleviate to some degree, but that doesn’t really help if you need to readjust to where you actually want to put your thumb every time you place your thumb on it, creating unwanted input in the process. You can put your thumb wherever you want on a thumbstick and it’ll only do its thing when you move the thumbstick.

        • Jakub Kamecki

          the interaction is the same as you imagine it. there’s a deadzone in the middle and due to the controller slant it’s a very natural position to be in – you keep your finger in the dip. not hovering. once you move from the middle to the sides you get inputs – either as Mick is doing it here DPAD like functionality or thumbstick like functionality.

          regarding muscle memory. there’s a dip in the middle and a pattern on the rim. we never had an issue in testing with finger position. it’s much better than smooth trackpads in that way.

          • anonmon

            So the tracked surface is so small you’re always either hitting the edge or the dip? Sounds better than most any touch surface I’ve seen in that way, though it does sound positively tiny to make that possible in use.
            How do you deal with super granular desired input when every slight rock and twitch of the thumb can create a different center point of touch? Just dampen and/or filter input to the point there’s some amount of delay to everything?
            Is the deadzone a hardware or software thing? I’d of guessed the latter, which I’ve never been the biggest fan of personally as there’s no universal standard “This is what you have to do to make it work” when it’s up to the system settings or dev as to exactly where such things are.
            Deadzones on thumbsticks are usually more to compensate for the slight play in the physical components than for someone to actively tune to a liking, given the more degrees of input you can have the better is pretty universal.

          • Jakub Kamecki

            https://imgur.com/gallery/R7bp1XO

            check these gifs out. they show how the top is structured.

            all of our technology is fundamentally software based, with thresholds, calibration and base lines of detection. based on feedback we can modify how it reacts to touch. so you could set the deadzone size, sensitivity, behaviour and additional interactions. it’s very much work in progress so everything is up for debate.

          • anonmon

            Ooh, the entire intended “active” area was supposed to be ribbed like the rest of the thing the entire time. Between all the marketing material and especially the videos, the small size, usually being obscured by something, and especially the video compression abound completely taking fine detail out of footage and screengrabs I’ve had the mental image that the entire thing was smooth aside from the very edge of the surface where the outside ribbed texture met the top.
            Though that does bring in questions of how these will last in the long term, as I can say basically all thumbsticks I’ve ever owned or seen other people own do and will eventually have any raised surfaces or patterns removed through continual use. Which with the etee will eventually make those touch pad surfaces smooth along the top like I thought they were initially. Even with hard plastic, which while it will take longer and is extremely dependent on how raised the patterning is, will be smoothed over time eventually with continued use.
            Before that happens though, I can absolutely see having a constant texture that radiates from the center like that would make it possible to know where your thumb is with the change in angles of the ribs beneath. For as long as those ribs last anyway.
            Still doesn’t do anything for the “continually giving input when you may or may not want to” thing given the software bound nature of the input method, but it certainly fixes most of the “What direction am I even hitting?” potential issues.

          • Jakub Kamecki

            https://imgur.com/gallery/pb770KE

            two things here. the device is very elegantly built providing for the ability to the replace the rubber quite easily. the rubberized surface is not the sensing part. the sensors are on the inside (the middle in the picture – like a fish bone skeleton)

            but even if the rubber gets worn out over time because we do dynamic thresholding and calibration we can account for changes in the surface, humidity, temperature, different hand sizes or capacitance and not affect the sensing at all.

            when i say this is the better approach to sticks it is. and i’m not just blowing smoke up everyone’s ass. it’s fundamentally more robust and provides the same functionality.

          • anonmon

            It’s good to see you’re not going full apple with how you design the hardware and think about (user) replaceable parts, and the wearing down over time won’t effect physical performance too much.
            It’s then really only a question of how long spare parts will last for when things do wear and break, which is a question with anything that’s niche and likely won’t have a ton of parts machines to pull from, or be popular enough that knockoff parts will start to be produced, unless printing new parts to the standards of the originals becomes practical.
            Frankly this is a issue with all of VR right now with replacement CV1 cables being practically impossible to get, or really getting any support with the myriad of WMR headsets that were only popular once they were being sold at clearance prices.
            Heck, being able to get SteamVR dongles after the Steam Controller went out of production basically died, with the only replacement seemingly being with getting them from someone selling them intended to be part of a developmental setup for lighthouse tracking.

  • Lucidfeuer

    I see it as a non-practical but interesting experiment. What would controllers be without the need of buttons and instead have functions being tied directly to the movement of each finger and articulations of the hand (if I understand the concept well). It’s just a shame it needs those damn Vive trackers and doesn’t solve that as well.

    • Jakub Kamecki

      We are developing our own trackers.

      • Lucidfeuer

        Nice. How far from the stick/How wide on the stick does it detects input from each individual fingers? In fact am I write to assume that the concept of the eTee, instead of relying on fixed individual buttons relies on each fingers as directly as inputs replacing buttons?

        • Jakub Kamecki

          the etee not having “buttons” is a bit of marketing. there is control parity with the HTC wands.

          you have 4 finger sensing on the body, now 2 sensors in the led light, trackpad that either does a stick or dpad with click functionality. plus three pre-baked gestures and gyro support. this all can be further combined and customized in engine to provide quite advanced and fluid controls.

          https://imgur.com/a/jo3L4YT

          • Lucidfeuer

            I have trouble rationalising the value proposition (or simply concept) of the eTee in a key sentence. This is a buttonless equivalent of the wand with added preprogrammed gesture interactions, of which the goal is more intuitive/direct handling?

          • Jakub Kamecki

            yes, buttons are a distraction. an abstraction layer that takes away mental capacity that can be used for other stuff.

            so by removing buttons we remove distraction and streamline the VR experience. making it more accessible and immersive so that more people can enjoy VR without a steep learning curve or a background in gaming.

  • Clarity Solutions FZE(+971-50-

    Very Informative, thank you

    • Jakub Kamecki

      Looking forward to a professional evaluation of the ergonomics! :D

  • JB1968

    Guys, all these amateur companies(including Oculus and Valve) controllers sucked so don’t have big expectations here.

    Note the big players here(Sony, Microsoft, Nintendo and others…) have put years of research and tons of money into this area. Don’t expect to get premium, durable product from a companies with nearly zero experience with massive production of physical devices.

    As for the buttons vs. gestures discussion:
    Remember when you wake up hearing your mobile phone alarm… Do you hate the moment you have to open your eyes first to do the proper touch gesture to turn the alarm off? Thats exactly the same feeling when you want to play VR games with gestures.

  • Keep an eye on Skarredghost for my”experiments with etee”…