We're testing Oculus Quest's two craziest new VR features.



[ad_1]

SAN JOSE, Calif. – Oculus is clearly optimistic for its wireless RV system, the Oculus Quest, which means this week's Oculus Connect conference is replete with $ 400 headphones. Unsurprisingly, the biggest queues of the show were dedicated to Quest and the surprise pair of surprise headphones "coming", announced Tuesday morning.

Quest's focus is largely on promising sales figures: "more than $ 20 million worth of games and apps have been sold in Quest's digital market since its launch in May," said Oculus on Tuesday. "More than $ 80 million in Rift-specific software" since the launch of this platform in March 2016. Four months versus three years and more? We do not need a graphing calculator to determine the platform that generates the most software sales.

Keeping this momentum in mind, I cut a few lines to see the two intriguing features planned for the near future of Quest: a PC-VR wired connection, launched in November, and a fully launched manual tracking API, "start 2020 '.

Oculus Quest Link: A deadly advantage, not a substitute for the Rift S

The comment sections of each Oculus Quest article at Ars have included a common chorus: wireless virtual reality is cool, certainly, but is there a way to plug it into a more powerful computer? To get a two-in-one device?

On Wednesday, Oculus unveiled the new "Quest Link" feature via a firmware update in November. This allows anyone to connect a Quest headset to a game-grade PC via any type C wired USB cable (provided it is rated high enough for data transfer), and then Use the Quest as if it was the Oculus Rift. Meaning: It will load Oculus Home and Oculus Dash interfaces, exclusive to computers, and support "all" Oculus Rift games and applications with Quest's Oculus Touch coupled controllers.

Hours after the announcement, Oculus researchers organized a roundtable to explain Why the feature was not ready for launch. In short, the data rate of the USB Type-C may be high, but it's not high enough to meet the high-resolution, low-latency video visuals hanging from the face of the camera. 39, a person. Without optimization, any VR video sent to a Quest via Oculus Link had two major problems: obvious visual artifacts (pixelization, noise) and increased latency between your actual movements and what appears in virtual reality.

Oculus admits that the USB Type C pipeline is not fast enough for a pure 1: 1 VR video signal. The team therefore had to develop a process to compress and code the virtual reality action from a PC, and then decode this information on a quest. The most important compression point the team could work with was not a reduction in overall scene resolution, as it would introduce an obvious pixel blur into a virtual reality game or application. Instead, Oculus has opted for a custom version of sophisticated rendering. The central part of any Oculus Link video signal will be closer to the 1: 1 resolution compared to the PC version, says Oculus, while the outer radius is reduced on the PC side with fewer pixels. and a fish-eye effect. When decoded on the quest side, this information is resized and resized.

Shortly after the presentation, I got attached to an Oculus Quest connected to a PC via Oculus Link and I played at the upcoming adventure game. The anger of Asgard for 10 minutes. I've spent most of my demo time moving wildly around the head and hands while squinting the outermost pixels of the Quest's outer radius. I went to look for obvious artifacts in the area where I was told that it was the most compressed. I spent my time in a colorful hub restaurant full of great characters and light effects, and honestly, I could not perceive any significant issues related to corner radius pixels.

This may be because the Oculus official line might underestimate the amount of compression applied to all Quest panel, not just its outer radius. Although Quest offers a high-resolution OLED screen with a resolution of 2,880 x 1,600 pixels, any image driven by Link seems to have a smooth but noticeable smoothing effect. In the end, it seemed more detailed than Quest's native content, but I could confirm anecdotally that a nearby game station, Oculus Rift S, looked much sharper, even though its LED screen is rated at 2 560 x 1440.

This feeling of improved smoothness could be explained by the fact that Rift S operates at a refresh rate slightly higher than 80Hz, compared to the refresh maximum of 72Hz Quest. There is also the issue of movement latency, which has been confirmed by Oculus Link researchers before Link can be put into production. The researchers' solution was to break down the transfer of coding and decoding required from one computer to another, according to a process called "slice image transfer". Basically, each image of visual data is sent as horizontal bands, one above the next, to be decoded and displayed, instead of following this process for one discrete image at a time.

Smooth: that's how we do it

This sounds like a recipe for "tearing the screen", in which different parts of the screen image fit into loose strips. But I did not see anything like that in my demo. However, when using Quest Link's test version this week, there are clearly some extra latency images between when I wave or push a button and when it happens. in the world of virtual reality. This irregularity may be equivalent to the latency found in wireless systems such as the HTC Vive wireless adapter or perhaps a hair faster than those. I will need more time to test before feeling confident in this call.

But as I have already pointed out, Oculus Quest 's built – in sensor range is solid and the natural movements of my head and hands have not been lost just because I' ve gone out of my way. a Rift S to a Quest. This is promising.

I also did my test with the official Oculus Link cable (visible in a gallery above), an optical fiber five meters long. I agree with the call of Oculus to say that this cord is specifically designed to distribute its extra volume of plastic in a way that seems barely perceptible, as opposed to a standard cable-type C sticky-right that you could buy at Monoprice. But Oculus has not announced any price for this cable, which does not inspire confidence. (Hopefully Oculus did not hire anyone from Monster Cable to pick a prize.)

The presentation of Quest Link ended without questions and answers, so I could not ask about compatibility with other VR software platforms, such as SteamVR. The good news is that the Quest Link connection fully emulates a standard Rift S headset and the hardware already works seamlessly with existing SteamVR software. So I am optimistic.

There is always a chance that my test experience this week will be replaced by other adjustments and updates from Quest Link before it's launched in November. Despite everything, I left my demo confident in two things. First: Oculus obviously had engineering problems to make it work without additional hardware. This is great news for current and future owners of Quest, with regards to the durability of its device.

Secondly, Quest Link introduces obvious and acceptable tradeoffs, which means that anyone who prefers PC-VR and ponei for a dedicated computer VR system should let go of his buyer's remorse. Quest Link is cool to the limit, not the ultimate PC-VR option.

Tracking hands looking: good as free update, not yet a game changer

Oculus also invited me to test two implementations of its first-ever manual tracking solution, which will arrive at Oculus Quest in early 2020 without additional hardware. That's the good way: Quest's version of tracking real hands in virtual reality is not good enough to recommend paying extra money. But as a totally free option, it's damn good.

The best news regarding manual hand tracking of Oculus Quest is that when it works, it is almost instantaneous. Hold your hands, rotate them, wave your fingers and make common hand gestures – and they will appear as remarkably accurate representations in the world of virtual reality. As long as your hands are not touching, Oculus Quest's Oculus Insight cameras are very well modeled.

My first test was to turn the bird around, which the Quest modeled well. (This means that Oculus does not actively censor your most idiotic gestures.) But I was more intrigued by the way Quest had modeled my fingers when they were pinching or squeezing each other. This type of gesture, in which the fingers meet like a bird's beak and faces turned outward, might be difficult to perceive for Quest's built-in cameras. But Quest never stumbled in this respect; its sensors have always understood my desire to take things up. I could never say the same thing about the LeapMotion add-on for standard VR headsets.

The catch? None of the applications I've tested have allowed me to pick up, transport, or drop objects in virtual reality. One of them was an incredibly simple training application bearing the Farmers Insurance brand. She had just pointed out to me the damage done by the water damage in the kitchens and determined if they looked like something to repair or clean. It was a hands-on experience as gentle as they come. (I guess it throws more evidence on the pile of research that video games are actively training our kids to become … adjusters.)

The second application, Elixir, was even more impressive, even though it did not include real "physical" interactions with virtual reality objects either. Instead, I could put my hands in different puddles of water, lava, poison or electricity, and then my virtual hands would change with strange mutations. The best one made my hands look like those of the Wolverine of X-Men; when I clenched my fists into a ball, the claws escaped from the "fwwwwsh" style, out of my fingers. When my hands mutated, they exploded with visual effects and makes convincing sound effects. Without the haptic growl of a controller, this audio-visual feedback was crucial for selling hand tracking as an immersive virtual reality experience.

But, again, I could not gather my hands. It is at this point that they disappear. The tracking of the hands of the quest constantly searches for some kind of skeletal pattern, and it is likely that the riddle of touching hands is lightened to lighten its processing load. I am happy to admit it for the sake of ease of use of virtual reality, but the most important problem at the moment is that the Oculus Insight tracking system is struggling to keep the trace of his hands outside a central point of visibility. Take one of your hands out of your cone of vision towards the front and it disappears fast enough. worse, this hand needs almost half a second to reappear in VR once it has returned to the cone.

If Quest's manual tracking continues to suffer from this vision problem, it will probably no longer be a reliable control option for active video games. (It's a good thing that Oculus Touch works so well, in terms of tracking and speed.) But I hope the features that make From the beginning of the work – tracking the masked finger, quick hands recognition, grip and pinch – you benefit from the benefits of applications on the fronts of education, training, communication and accessibility, especially as a default Quest value. (Say it with me: sign language as an integrated VR standard.)

Oculus registration image

[ad_2]

Source link