3 ms·
Since it only had two front-facing cameras, the Mirage Solo would have required extra hardware to track any controllers. Also I don't think there was ever an ap
by ggreer 6y ago
Since it only had two front-facing cameras, the Mirage Solo would have required extra hardware to track any controllers. Also I don't think there was ever an app for it that let you walk two steps away from the origin. The only truly room scale thing I saw was a demo written by some developers. They had to put the headset in dev mode and disable a bunch of safety mechanisms. I wonder if this limitation was imposed to minimize drift.
- jayd16 6y agoI developed on it and the limitation was mostly self imposed for safety/perf reasons. Its much easier to draw a consistent roomscale boundary on the Quest with the 6DOF controller. Without that its a real UX problem. Even the Quest is constantly asking to redraw the boundary. It was one of two SOC generations (as well as VR SDK gens) earlier and it was quite hard to build out a full room without a lot of perspective tricks. There was an experimental add on for 6DOF controllers https://developers.google.com/vr/experimental/6dof-controllers https://developers.google.com/vr/experimental/6dof-controlle... It really wasn't a fundamental technical leap to go from the Mirage to a Quest. The Quest feels like a (well thought out) iteration instead of a revolution compared to the Mirage Solo.
- schrep 6y agoTracking controllers in 6DOF from headset cameras is the hardest part. Nobody has yet replicated this for a stand-alone headset.
- jayd16 6y agoWhy do you say its the hardest part? Hand gesture tracking is quite hard, but tracking IR LEDs on a controller is much easier than tracking head motion as long as you have the camera setup to keep them in view.