Oh nice! I just bought Quest over the weekend from a guy off craigslist who was selling them at cost and still wrapped in plastic.
Fantastically impressed with things. Totally worth the $500. The games are amazing, but what is really impressing me is all the vr video content. The ability to "stand" on a stage while the orchestra plays Mozart all around you is mind blowing.
It is neat to see them pushing the boundaries with updates.
The Quest is probably the first thing I've bought in a long time that has given me that "Christmas morning" feeling. It's such an impressive piece of tech, and VR without wires is so much more immersive, even if it limits the graphical capabilities
Wait really?! I was wondering why I would want to fork over so much money for the app, but having this as a possibility changes everything. Thanks for the heads up :D
To wirelessly stream PC virtual reality games to the Oculus Quest you wouldn't just need Virtual Desktop but also a 5GHz wifi router.
You may be interested in the alternative (official) solution, which is not wireless but allows streaming PC VR games using "Oculus Link" using a compatible USBC cable. The cable that comes in the box is unfortunately USB2.0 (so not compatible), your best bet is to checkout this page: https://support.oculus.com/444256562873335/. The official Oculus cable has fibre-optic transceivers built in so is very expensive. Oculus lists some third-party cables on Amazon by Anker, which used by a lot of people (10 foot recommended, but the tiny 3 foot one is really cheap).
I haven't yet tried Virtual Desktop 5GHz streaming, but Oculus Link with a 3 foot cable streams perfectly (good for seated VR, which many PC games assume). I do want to try 5GHz Wifi streaming though! Untethered play is the best!
Small update for you: I did this on my AmpliFi HD mesh network and it worked pretty flawlessly! I tried it in multiple places around my house and didn't experience lag, exposure of apps I tried was super limited cause I got distracted by other things, but it's really nice! I also tried the Oculus Link method, and honestly the Virtual Desktop experience was surprisingly better? Maybe it's all in my head because it's all wireless and feels like magic haha
Loading up the SideQuest APK was super easy and has now led me down the awesome rabbit hole of poking at Oculus dev :D
As mentioned, definitely make sure you're connected to the 5GHz hotspot if it's available -- it makes a big difference in my experience streaming a regular desktop with VirtualDesktop. Also as you might be aware, unlike regular 2.6GHz wifi, please remember that 5GHz signals don't penetrate objects (including walls, furniture) so your best bet to reduce potential lag (and thus nausea) is to make sure your Quest always line-of-sight to the router.
Note that to get that functionality, you'll need to sideload a small .apk (Quest is built on Android). You can use plain adb to do this, or something like SideQuest if you want a GUI.
Unfortunately it doesn't work well on my Wifi with more graphically intensive games. It's playable but there's enough jitter than I get a bit nauseous after a while
Yea, I was surprised. He had them in boxes that were clearly shipped from Best Buy. I asked if he got them cheaper and he said no. Didn't ask why he was selling them at cost, but I thought it was nice.
or simply doesn't know he could sell higher. Like I just have the (often wrong) mentality that electronic gadgets just deprecate right after you bought them. Quite surprised when learned that some of them worth more than retail price at secondhand market.
I recall reading about a scam, that went something like this.
You buy something online, and it seems to be a reasonable price. Maybe 40% less than the listed price. So instead of paying $1000, you pay $600. Nice, you think. You scored a great bargain. You buy it, and the product gets shipped to you, all new and everything.
Fantastic! You think. You scored a great deal.
But behind the scenes, what happened is that seller bought it with a stolen credit card, at normal prices. Then sold it to you at the cheaper price. He pockets the $600. The vendor gets paid the $1000. But the guy whose stolen credit card got used, now gets hit with a mysterious $1000 bill. Poor guy. Now he has to file a police report, and claim that he didn’t make this fraudulent purchase, and turn his life upside down for a few months.
My Oculus Quest is the best toy that I have bought for myself in years, even more enjoyable since I did VR for SAIC and Disney about 20+ years ago, so super exciting to see such a fun low cost consumer VR device. I can’t wait to try a game/experience using hand tracking. Lot’s of great content, my favorites being the Darth Vader trilogy and the RX racket ball and ping pong games.
I really want to know why this is feasible for the Quest but not the Rift S, given that both have Inside-Out tracking (ie. IR cameras mounted on the headset). In fact, the Rift S has 5 cameras to the Quests's 4 so it should be at least as capable shouldn't it?
It's a quirk of the Qualcomm chipset they used in the Quest. It has an extra DSP accelerator that was originally unused, but they managed to repurpose it for hand tracking. It was an opportunistic play, as I understand it.
Could you link to a source about that dsp being specifically used for hand-tracking? My quick google didn't found a result that could confirm that and it would explain a bit. Thank you!
It's in here somewhere. It's also extensively covered in the hour breakout session by the folks who implemented it, which is uploaded somewhere though I can't find the link atm.
Rift S could do hand tracking better than Quest, but it requires a separate effort that isn't currently a high priority. Source: I worked on it at Oculus.
I would be suprised if it's primarily a technological reason.
My thinking is: The Rift S is 99.99% used for "serious" gaming and PCVR will always be that way. For serious gaming you'll probably need some kind of controllers for the near future, since you want that fidelity and haptic feedback. Now, controllers could be replaced some day by braceletes or what not, but this is not going to happen next year.
The Quest and its successors (i.e. stand-alone VR), on the other hand, is and will be more and more used for lighter gaming, video, browsing, productivity, fitness, social etc. Often times, hand-tracking will be enough without the need for controllers.
Therefore, they focus their resources on bringing hand-tracking only to the Quest.
My guess is it's partly that but also mixed in with aligning to their mass market consumer ambitions.
I've given my Quest to countless people to try out and while every single person is blown away, many / most of them are challenged to use the controllers as anything more than simple pointing devices without significant training. With hand tracking that all goes away. You just use your hands how you normally use them and it "just works". It's similar to how the original iPhone captured everybody's imaginations because it offered such intuitive gestures to control its UI.
If you see the Quest as their "mass market" device and then you consider that their top priority is to knock over every barrier that stops it spreading like wildfire, then it's a no brainer to eliminate the controllers.
+1. Also worth noting, the implementation would be fundamentally different. On the Quest it's running on-device on specific chips, whereas on the Rift S this processing would be done on the user's PC.
It's doable, but you probably can't just drop the implementation for hand-tracking on one platform and have it work on the other.
The Quest can link to your computer and essentially act as a Rift using a USB cable (needed to be USB 3.0, a recent beta update knocked that down to USB 2). I can easily see the Rift being removed from the product line in the future.
Does the Rift S send its camera feeds to the PC for controller tracking or does it do it onboard and send the tracked controller locations?
Perhaps the onboard hardware is capable of controller tracking as sold, but not powerful enough to enable Quest-style hand tracking. What if in addition to that, the camera data for one reason or another can't be piped over USB to the computer, either because it's missing the hardware that would do the encoding, or because there isn't enough bandwidth on the cable?
As I recall, the original Rift's outside-in tracking was extremely picky about having separate USB 3 ports for each of its tracking cameras, to the point where if you had a three-camera setup it didn't even want them on ports handled by the same USB controller on the motherboard.
Now we're saying "The Rift-S has five cameras and it's connected to the computer, surely it can just let the computer process that data." I doubt it's that easy.
Consider the fact that the Rift-S has been out of stock for months. Oculus continues to re-stock the Quest (which sells out within 24h), but we haven't seen a restock of the Rift-S in probably 3 months?
Their "Del Mar" which is coming up, is very likely not going to be a tethered headset. My money is it will be standalone like the Quest.
Oculus is trying to get the masses to adopt VR and I salute them for it. There will always be headsets like the Valve index for enthusiasts.
I think it's a smart play. Standalone VR makes it incredibly more accessible. Everyone with a Quest does the same thing, they cart it around to parties and gatherings to let people experience VR and it's a blast. Even if this wasn't intentional marketing it's the best strategy for getting the word out. I'm hoping the Del Mar is a big upgrade, if it can bring better processing and better resolution to the table hopefully with a killer feature like adaptive focus and gaze tracking then I think it'll just knock it out of the park. Regardless though it's only a matter of time until all our screens are virtual.
I mean, going forward, given the existence of Link (and that Link surprisingly actually works), there is very little reason for the Rift S to exist at all; it sucks that they sold it to people and then so quickly obsoleted it by another product that came out at the same time and cost the same amount, but other than the feeling of responsibility for those customers I can't imagine any reason for them to spend any time at all on that device line: that use case is now Quest Link; if they built and sold more of them they would just be digging themselves a deeper liability hole of more limited devices they don't want to support.
Linus Tech Tips did a pretty damn good review of the Quest + Link solution. At the time he did the review, it looked like it had some very strange black bar artifacts when turning your head quickly, which the Rift-S did not. He also described it having a slight lag in the controls.
These kinds of things are likely non-issues for your average gamer who isn't moving really fast, and isn't requiring ultra precise controls. For some people though, it's a huge deal breaker.
I'm personally very grateful for both Valve and Oculus. Oculus is serving the mass market, and doing a great job of it. I just hope in the process, we don't lose the high end consumer gear like Valve Index.
How long ago was that? The Quest link cable stuff is still in flux. Not only did they just announce a few days ago that any USB 2.0 cable should now work[1], but they note in that article that Carmack is hoping to add a new mode to take advantage of the higher bandwidth of USB 3.1.
It's entirely possible if that Linux Tech Tips review was more than a month or two ago, things might look considerably different now.
I know, it sure seems like he has a lot of skill and institutional knowledge to contribute. That said, I'm not sure what his current involvement is, I just know they referenced him in that article.
By that argument aren't the controllers experiencing more latency on RiftS than Quest already? I realize there is more post-processing with hand tracking, but given that it already experiences the latency of having to process the control actions... not really sure how this could affect it that much.
Also, you could argue the hand tracking calculations could be done faster on a PC, so even if there is some perceived latency it could balance out or be better... would really need numbers for all of this, but it does just seem very suspect that they are trying to push more things on the Quest intentionally..
Internal company politics and prioritization of Quest. It's possible that hand tracking will come to Rift S, but it might be a side effect of wanting to support it with the Quest link cable.
Just the reality that headsets like Quest are the true future of VR, PCVR is an evolutionary dead end a very small percentage of people who can afford them want a gaming PC tower in their home, PCVR is a fraction of a fraction of percentage.
I own both and demoing my Rift on a high spec system my friends were blown away and the first question was "Can it run on my macbook?" Then thats where the idea of owning VR died for them.
Demoed my Quest and one person bought one within 2 minutes of taking the headset off, took his to a family BBQ and two more people there bought it after trying it for a few minutes.
I still enjoy many PCVR games but this is just reality.
> PCVR is an evolutionary dead end a very small percentage of people who can afford them want a gaming PC tower in their home, PCVR is a fraction of a fraction of percentage.
PCVR is no more niche than the demographics for Far Cry 5. If you have the hardware to run a modern AAA game on PC, all you're really missing at that point is the headset. And I don't see Ubisoft refusing to make games for PC because there are too few people with powerful enough PCs. As GPU tech continues to develop and higher performance reaches lower price points, it'll also become more accessible to more people. Though you can already pay 200-250 Euros for a decent GPU now and it'll be more than sufficient for VR gaming.
PCVR capabilities are too high for it to be a dead end. More niche, absolutely, because mobile VR/AR has somewhere from console to smartphone-level growth potential--but a smaller, higher-paying market for premium experiences is going to continue.
It's unlikely that PC will ever be mass market, but when/if VR makes more in roads in productivity applications PC will be key. There is a lot of promise in art creation, game development and 3D content creation, and of course HN's traditional programmer dream of infinite terminals, but the real killer app is telepresence for remote meetings. All of those use cases need better software and hardware
>There is a lot of promise in art creation, game development and 3D content creation
Absolutely agree with this, literally anyone who can use the headset is able to sculpt a 3D model in Oculus Medium that resembles what they want. Same can't be said for any non-vr 3D modelling software, huge potential in the creative industries.
Not exactly as powerful but something similar and that can run on a browser, opening up to a lot of creative applications, handpose powered by tf.js(similar model exists for tflite)
With the worry about Coronavirus, hand tracking with cameras or something like Google Soli[1] might be useful in ATM’s, elevators, kiosks, etc so people can avoid physically touching the devices.
Apple's first published ML paper[1] described how they trained a neural network to estimate eye gaze. To augment the limited data set of labeled eye photos, they rendered synthetic eyes and then used adversarial training to make them match the real-life data set.
This paper came out several months before the iPhone X introduced Face ID with eye gaze tracking, so it can be seen as a hint of things that were to come.
What's intriguing to me is that the paper also talks about doing the same with synthetic depth maps of hands, adding details such as occlusion and imperfectly-captured edges. With Apple investing heavily in ARKit, and rumors of an AR headset, it seems like hand tracking is inevitable.
Does the Hand Tracking API allow for tracking other objects? In particular, could a long straight object be tracked as a pointer based on tracking two points along its axis? This seems like it would be computationally easier than tracking hands, and would open up a lot of possibilities for using real world objects as props within the VR environment.
I imagine some kind of VR tracker for the Quest will come out eventually. Something like the Vive trackers but smaller. Agree, it would open up all kinds of interesting applications.
They could be wildly successful if instead of $100 pucks they could just sell $1 stickers that you could attach to whatever you want. With enough of them you could probably use it to aid scanning in geometry of real world objects
Not at this time, hands only. I imagine it might be difficult to distinguish a pointer from other straight lines in the room, hands are a little more distinctive.
With hand tracking they mean finger tracking, seeing you 'click' with your index finger and thumb. The Oculus has hand tracking today via controllers already.
They mean handtracking. Oculus touch is a tracked controller, which means the controller is tracked and the tracking system has no concept of a hand. You could put the controller on your foot, or the floor or on your dog and it would track in the same way as your hand. Hand tracking uses computer vision to recognize hands, for most humans fingers are included in that
And Quest has that proper hand tracking already (had it for something like 5 months now), it's just been an experimental feature until now. I haven't used it much; for some reason the tracked hands seem to be placed lower than the fake hands when I'm using controllers, which makes the latter a better experience for me.
Increased latency due to using complex, deep learning based, entirely computer vision driven hand tracking vs. tracked controllers which partially use accelerometer data to correct drift and account for latency, and use a much faster classic computer vision algorithm for position tracking (https://developer.oculus.com/blog/tracking-technology-explai...)
Do Deep Learning-based methods introduce all that much latency? My understanding of neural network is that the training is expensive, but using them is essentially free - you have to push your pixels through a few layers of matrix multiplications and you have the results; it's just tuning those matrices that takes time.
VR latency is measured in milliseconds, the operation you described can easily take multiple frames to complete even with a very fast, optimized model. Machine learning inference is never free, and often isn't real-time. Training can take hours to weeks depending on the model.
At the risk of being a downer: I'll be so happy once we move past the "preschool playset" era of hand-tracked software.
I wonder what it is that's keeping us from moving past these dull tech demos. Is the issue the interaction design being too simplistic ("Look, I can pick [thing] up!")? Or perhaps the opposite: that, without suitable haptics, too much energy has to be devoted to a very unnatural way of interacting (essentially with phantoms). It seems that games like BeatSaber and apps like TiltBrush have cracked the code, but you have to wonder how.
Fantastically impressed with things. Totally worth the $500. The games are amazing, but what is really impressing me is all the vr video content. The ability to "stand" on a stage while the orchestra plays Mozart all around you is mind blowing.
It is neat to see them pushing the boundaries with updates.