Slashdot Mirror
Oculus Unveils the Rift S, a Higher-Resolution VR Headset With Built-In Tracking (theverge.com)
Oculus VR unveiled the Oculus Rift S, a higher-resolution pair of virtual reality goggles that remove the need for external cameras by incorporating built-in tracking. The company partnered with Lenovo "to help it speed up manufacturing and to improve upon the design of the original Rift," reports The Verge. From the report: The result is a new VR device that is more comfortable, sports 2560 x 1440 resolution (or 1280 x 1440 per eye), and features the same inside-out tracking system that will ship on Oculus' upcoming standalone Quest headset, which the company calls Oculus Insight. That way, you won't need cumbersome cameras to enable full-body movement. In another twist, both the Quest and Rift S device will cost exactly the same at launch: $399, with the same pair of slightly modified Touch motion controllers included and the same integrated audio system (plus a headphone jack for external audio). That decision makes it clear that Oculus wants its VR platform to offer a choice not between two vastly different pieces of hardware, but by the more simple determination of whether you have the hardware to power PC-grade VR. The Rift S will support every existing and future game on the Rift platform. "The company is also enabling cross-buy and cross-play features," the report adds. "That way, you can buy a Quest and, at a later date, upgrade to a Rift S and still have your entire library intact. Additionally, multiplayer games that support both platforms will let players play one another, regardless of whether you're playing on a Quest or Rift device."
The Rift S and Quest will be shipping this spring.
The Rift S and Quest will be shipping this spring.
I've used Oculus headsets for a few short things, including some experiences at the Void.
While they were pretty cool I could tell that even higher resolution would for sure help with fidelity, as sometimes you are looking at things you can tell should have more detail but the resolution is failing you...
Hope they are finally getting around to having these headsets work on the Mac though. Ignoring a customer base with a lot of money seems like a pretty stupid play for VR makers, or at least opens up a giant door for Apple to eventually drive in through.
"There is more worth loving than we have strength to love." - Brian Jay Stanley
The screen has higher resolution, and is not pentile so it's actually even higher than the ~40% pixel increase. Interestingly it actually has less resolution than the Quest.
The screen is LCD rather than OLED, so one might think it will have poor black levels compared to the CV1. However, CV1 is often driven with compressed blacks and so doesn't get that brilliant OLED "completely off" black in practice anyway. Will need to play with it to see.
It also runs at 80hz rather than 90hz, though I suspect that may not be detectable.
Sure, it has more inside-out cameras (I think), but it no longer has mechanically adjusted PD and no ear phones.
In some ways, it's a step backward. I'd rather have seen a hybrid approach to sensors - the inside out is great, but a couple sensor pods behind wouldn't be too bad. Oculus "partnered" with Lenovo on the "S" - and they basically swallowed up the Lenovo Mixed Reality headset (which can usually be had for a lot less than $399), and passed none of the cost savings onto consumers.
Where's the wireless option? Go and Quest are standlones, and have their own issues... how about something like Vive's wireless option? Instead of pushing VR tech forward, they've just sidestepped into Microsoft's MR standard. It's not a terrible thing... the MR headsets are very good, and inside-out tracking is very slick, but it isn't a step forward.
I want no tethers and galvanic stimulation to ward off motion sickness (and feel motion). I want OLED displays. I don't like losing the earphones.
hence the "Built-In Tracking"
sports 2560 x 1440 resolution (or 1280 x 1440 per eye in Canada)
FTFY
(-1: Post disagrees with my already-settled worldview) is not a valid mod option.
Macs don't have suitable drivers for the video card and the apps/software runs on Windows.
They have capable hardware so that is not an excuse. In what way do they not have "suitable drivers"?? Are you saying I couldn't easily drive a Rift from a Thunderbolt 3 connection and an eGPU or iMac Pro GPU? Come on.
Plus, you'd need a Mac Pro,
Nope. Newer iMacs, any laptop that can support eGPU, an iMac Pro. Multiple options for how it could happen and consumer who would pay if it worked well. Very likely as large a market as those who are willing to spring for VR gear in the first place. When Apple enters the market they will dwarf all existing marketshare near instantly and show just how wrong that opinion is.
"There is more worth loving than we have strength to love." - Brian Jay Stanley
Figures. Yet another platform tracking user behavior. Is there a "Do Not Track" privacy setting, like in browsers? :-)
It must have been something you assimilated. . . .
I really hope that pure inside out tracking doesn't become the new thing unless it improves significantly. I have a lenovo explorer WMR headset, and the tracking is horrible compared to my friends vive. It's especially annoying playing beatsaber when I lose tracking on a controller while swinging behind/to the side and it teleports into a mine. The controllers also jitter horribly, to the point that shooting games are near impossible to play with any kind of precision.
Or built 802.11ad ("WiGig") into the Quest, and allowed anyone with a suitable access point (or PC via peer-to-peer) within 3-10 meters in the same room to basically stream 90fps uncompressed video directly to it with minimal latency. From what I understand, short-distance line of sight 60GHz 802.11ad is capable of easily sustaining 4-6gbps with a minimal implementation, potentially faster with a more sophisticated implementation.
Assuming my math is correct, 1440 x 1600 x 2 x 4 x 90 = 1.659gbps. With 4-6gbps to play with, there's plenty of headroom left to stream the Quest's own video and sensor data straight back to the host PC (without delay or compression).
If I understand correctly, and the Quest's own cameras can be used to capture eye-level stereo video suitable for passthrough AR, having 802.11ad onboard would also potentially enable you to do something like this:
* Grab a pair of raw video frames at the camera's fastest framerate, plus all the sensors that the host PC has subscribed to.
* Throw them across the network at the host PC without compression for it to use for things like object/gesture-recognition.
* A moment later, the next pair of frames arrives from the host PC (either without compression, or some low-latency mostly-lossless fast compression that only compresses the delta between the left and right frames), with 8-bit alpha channel. Merge them with the most recent local passthrough frames, and display them.
The advantage of doing genlock-type overlaying of PC-generated video with locally-captured passthrough video is latency... when doing passthrough VR, even a single frame of latency is painfully noticeable, and two or more is downright nausea-inducing. By keeping the passthrough-AR video's processing realtime and local, you could shave the frame or two of latency you'd otherwise end up with if you had to wait for it to get to the host PC, get edited into the next frame, then get sent back.
Remember... with VR, and even moreso with passthrough AR, latency is EVERYTHING. Anything you can do to reduce it is going to profoundly improve the user experience. If you're going to just overlay remotely-generated content into video captured locally ANYWAY, it makes no sense to add the latency of sending that local passthrough video to the remote PC just so it can be edited into the scene and transmitted right back. The Quest has more than enough horsepower to do THAT job locally.
Anyway, that's my view. 802.11ad has plenty of bandwidth to stream high-framerate video with minimal latency in both directions, so you'd get the best of both worlds... you could use the Quest's own processor to do locally things that would take longer to roundtrip, while nevertheless offloading the really heavy lifting to a PC across the room that's as powerful as it needs to be to do whatever you want.
Now, the good part. IF (big "if") the Quest actually has a USB 3 port with enough bandwidth to sustain anything close to 2gbps+, and Oculus exposes enough functionality to allow a thirdparty to pull it off, some thirdparty (or grad student looking for a cool thesis project) could conceivably make their OWN 802.11ad interface for the quest, and do exactly what I've described. Say, built around an off-the-shelf 802.11ad mPCIe module, in a belt/pocket-mounted case the approximate size of a RasPi with its own battery, connected to the Quest via USB. Ideally, the Quest supports being a USB host... but even if it doesn't, implementing the module as a USB host and connecting to the Quest via OTG isn't all that big of a hurdle. The biggest unknown is whether the Quest's USB implementation is actually fast enough to sustain the kind of transfers using an external 802.11ad card would require.
Ugh, the math actually isn't correct. I realized a half second after hitting submit that it should be 1440 x 1600 x 2 x 90 = 414,720,000 PIXELS/second. Assuming 32-bit alpha-blended color, that would be a little over 13gbps (415 million pixels x 32 bits/pixel).
That said, my idea might still be do-able... assuming the left and right frames are relatively similar, you could get lossless compression approaching 2:1 right off the top, reducing it to ~6.5gbps. Converting the video to 4:2:2 YUV with 2-4 bit RLE-compressed alpha-blending should easily chop at least a third of that away, getting us down to 4gbps. Dropping back to 4:2:0 color for the first frame after a radical scene change and referencing the previous frame can probably get us down to 2-3gbps. It would probably mean we'd have to do all the local image-recognition onboard instead of streaming the onboard cameras directly to the remote host since we'd already be maxing out the best-case USB bandwidth, but it should still be do-able since that would be just about the only thing (besides rendering the remotely-streamed video directly to the screen in the Quest, possibly after overlaying it on the local passthrough video in a genlock-like manner) we'd have to DO on the Quest itself.
So... I think my idea might still work, but only if the Quest's USB interface really, truly can sustain 2gbps at the bare minimum, and would be risky unless it could reliably sustain 4gbps.Any lower bitrate, and we'd probably have to forget about uncompressed (or lightly compressed) video from the host PC (or at least, settle for major visual artifacts whenever there's a radical scene change, since we'd only be able to reference past frames without adding latency and any completely new frame would be limited to the data we can fully transfer within 1/90th of a second)
I think as soon as you throw compression into the mix you introduce huge latency issues.
The sad fact seems to be that we just don't yet have the wireless bandwidth to stream a premium VR experience.
Unless perhaps you can operate multiple WiGig channels simultaneously without interference - 4 channels delivering a combined 20+Gbps could do the job nicely.
Or, a personal favorite - what if you took something like the Quest, and wirelessly fed it pre-transformed and optimized geometry so that it only had to perform the final-stage pixel rendering?
--- Most topics have many sides worth arguing, allow me to take one opposite you.
Clearly the Quest has better resolution and the better colors of OLED.
I'm open to trying a LCD VR display.
With OLED dark areas especially in space SIMS have annoying lag to them. Worse is the burn-in dirty sheen that gets worse as display elements age drifting further out of calibration. Not all that impressive to me either.
Keep in mind that OLED usually has a pentile sub-pixel layout that only includes two "color dots" per pixel instead of LCDs 3.
The actual number of individually subpixel elements is then:
LCD: 2,560 × 1,440 x 3 = 11.06 million
OLED: 2,880 × 1,600 x 2 = 9.22 million
Despite the apparently lower resolution, the LCD actually has 20% more "color dot" subpixels, and will thus deliver an overall sharper image.
--- Most topics have many sides worth arguing, allow me to take one opposite you.
Didn't Oculus claim gaze tracking was the next big step here after the last headset was released over a year ago? (having trouble finding any article now)
And my reaction is "Lame"...I've seen a competing device with these resolution specs (1440p) and built in gaze tracking. Out for over a couple years now already. I'd been waiting on higher FPS and maybe from a bigger company.
https://www.getfove.com/
Interesting note... the Quest uses a Snapdragon 835 SoC, which was designed to cheaply support 60ghz 802.11ad with the addition of a single chip and some passive components. While that chip obviously isn't in the Quest, I can't help but wonder whether Carmack might have been able to pull some strings and find a way to expose the pins needed to interface with that chip so it could be implemented as an external add-on.
In most Android devices, the USB port is actually connected to the USB root hub through a crossbar chip that enables the port's literal pins to be connected to multiple different circuits inside the phone, and used for purposes that have nothing whatsoever to do with USB. For example... JTAG, I2S digital audio, analog audio, a TTL serial port, whatever.
I don't have an Adreno 835 datasheet in front of me, nor do I have any idea how the Quest is wired up internally. But... given how badly Carmack WANTED the Quest to support Rift-like usage, I'm going to go on a limb and theorize that if there's any way the engineers could have exposed the necessary pins to connect the Qualcomm 802.11ad chip to the 835 through the crossbar and USB-C port without raising the manufacturing cost... they almost certainly DID. Carmack is someone who absolutely HATES to slam the door on a future possibility if there's any possible way he can leave it propped open for free. He might have lost the battle to add the extra hardware to support something like DisplayPort-over-USB-C, but exposing a data bus through a crossbar that's already "there" using lines that would otherwise have gone unused is another matter ENTIRELY.
Another possibility: if at least a few of the pins on the USB-C interface are connected to pins on the 835 that can be tri-stated, and the pins on the 835 that are needed to connect to Qualcomm's 802.11ad chip can be tri-stated, Oculus could have conceivably dispensed with the crossbar entirely and just connected the two sets of pins in parallel. It would complicate the software initialization slightly (if something caused the 802.11ad-interface pins to go high or low, like a firmware bug, it could cause the USB interface to malfunction), but in hardware terms, it would basically be free. Worst-case, they might have to add some pull-up or pull-down resistors that they might have gotten away with omitting if the lines were single-use, but officially are required for standard-compliance ANYWAY.
On the downside, I'm not sure someone actually COULD viably sell such an 802.11ad interface, even IF the signals were exposed as described. The problem is FCC certification. If someone built a module that used an already-certified mPCIe 802.11ad module and connected to the Quest via literal USB 3.1, getting it past the FCC would be easy... they'd just have to prove it didn't emit excessive RFI, and that the 802.11ad module itself was already certified. In contrast, if they tried to roll THEIR OWN 802.11ad module using Qualcomm's bare chip, they'd have to go through a much more rigorous and expensive certification and approval process that would likely result in an 802.11ad add-on interface that cost $25 to manufacture and had to sell for $600 to cover the cost of getting it certified in the first place.
Compressing with something like h.264/265 that depends upon having a few frames to reference adds latency. Doing things like HuffyUV... YPbPr with 4:2:2, RLE, etc, is fast, because it doesn't depend upon knowledge of anything besides the current frame.
The big delay with most codecs isn't literal calculation time, it's the need to wait until you have at least 2 or 3 frames in the pipeline before you can even START compression. For a modern GPU, matrix transformations between RGB & YPrPb or between 4:2:2 and 4:4:4, are practically instantaneous.
The key to making something like this work with a 1-2gbps link is to forget about codecs designed to be maximally-efficient, and instead look at codecs designed to be fast & "efficient ENOUGH".
I have a Vive pro with a wireless modules which uses WiGig. And it works perfectly. The delay is negligible and you don't see the difference with a tethered headset. The only drawbacks are that WiGig doesn't penetrate anything so the screen goes grey when you block the line of sight your body and the battery only lasts about two hours so if you play a lot you need two and be prepared for an interruption after about two hours.
Yeah, that's what the world needs, people who pay even less attention to their surroundings.
We used to have a Bill of Rights. Now, with the rights gone, all we have left is the bill.
I'm fairly sure that any and all activities of the player will be tracked to the utmost detail.
It's Facebook, after all.
We used to have a Bill of Rights. Now, with the rights gone, all we have left is the bill.
It would have been a better product if there was only one product to begin with. A single VR headset that can operate in a portable and tethered mode. Something that allows people to play PC games, but also play games on the go. Having multiple products with different specs and meaningless monikers just confuses consumers and fragments the player base.
The Odyssey is actually lower resolution than most MS MR headsets, with the same "2880x1600" OLED resolution as the Occulus Quest, so it will also have a lower quality than the Rift S. In order of decreasing number of subpixels:
From what I can find:
MS MR Acer, LCD: 2880x1440x3 = 12.4 million
Rift Quest LCD: 2,560 × 1,440 x 3 = 11.06 million
Rift S, Vive PRO, Odyssey, OLED: 2x2880x1600 = 9.22 million
Original Rift and VIVE, pentile OLED, so 2160x1200x2 = 5.2 million
And of course there's lots of variation in comfort and adjustability, "halo" versus straps, lens quality, and the fact that some of them lack a mechanical interpupillary distance adjustment, when means the optics will mis-aligned for virtually everyone, and there's only so much you can compensate for that with rendering tricks. And of course the LCD-vs-OLED comparison should also consider LCDs general inability to display anything close to true black.
As for why anybody is bothering with Occulus - I've only used the VIVE, but from most everything I've heard the rift is delivering the more polished and pleasant to use product. Not the most technically capable perhaps - but like a keyboard, the ergonomics and feel of anything you physically interact with are some of the most important aspects.
--- Most topics have many sides worth arguing, allow me to take one opposite you.
Truly? I suspect that varies a bit from person to person, but you say no problems?
That bore investigation, and you might be interested that one of the things I came across was a comment about high-capacity replacement batteries that extend the life to 4-5 hours
It really does demand two extra QC 3.0 batteries if you're going to use it for any extended period. It draws 18 watts (12v @ 1.5 amps), which means a 20k mAh / 72 kWh battery lasts about 4-5 hours. The included battery's life is too short to be usable
That's all the info they offered, but it sounds like it might be worth investigating.
--- Most topics have many sides worth arguing, allow me to take one opposite you.
The main issue with 802.11ay is that 802.11ad exists today at a cost that's within the realm of 'sane', even though it's still pretty niche. 802.11ay, not so much. Not everything involving 802.11 turns to gold. Just look at 802.11y (basically, 802.11n with tweaks & extensions to operate in the lightly-licensed 3.6ghz band).
Back when AT&T still had a 1gb monthly data cap, I would have totally bought a home 802.11y access point & license, and a pocket-sized access point that used 802.11y for the backhaul & 802.11n for connections from my phone, because it would have enabled me to have free ~256kbps(ish) data within ~10km of my house for approximately what a year of monthly overage fees from AT&T (if I'd used it without restraint) would have cost, then paid for itself thereafter.
The only problem was, as far as I can tell, 802.11y-2008 gear never actually EXISTED as off-the-shelf products. The closest I found was a mPCIe module (by Microtik, I think) that *might* have been usable to make my own access points... but even now, I'm not 100% sure the module was 802.11y (the description circa 2012 mentioned wi-fi and 3.6ghz, but never came out and specifically said, "802.11y"). Until an 802.11* standard is at least mature enough that someone can choose between at least two off-the-shelf APs, and has at least one or two USB, mPCIe interfaces available, it's too risky to depend upon.
That said, I *am* kind of surprised that 60ghz is literally SO line of sight that it can't even work if you're 10 feet from the AP, but blocking the direct signal path with your body. I know it has no ability to PENETRATE anything, but I expected that it could at least REFLECT off of walls, the floor, the ceiling, etc.
Does the Vive just have a really, really shitty implementation (poor antenna, less than max legal transmit power, poor RF design vulnerable to noise or overloading, etc)? Or is that how bad 60ghz 802.11ad is in general?
Update... just did some more research about 802.11ad. According to what I've read,
* 60GHz can't penetrate anything (including fleshy lifeforms), but absolutely DOES reflect from most surfaces. Its max distance is approximately 100 feet, measured along the signal's path (including any reflections necessary to reach the receiving antenna).
* In an area like an enclosed conference room that's approximately 12x20 feet, with the AP itself located in a spot that's sane (ie, not inside a cabinet or deliberately hidden behind/below something), but not otherwise chosen with any particular care, you can realistically expect 1gbps in 99% of the room, 2gbps in 90-95% of the room, and 4gbps in 60-80% of the room.
* It appears that the real-world performance difference between two single-antenna 802.11ad network adapters directly communicating as peers is PROFOUNDLY worse than the performance you'd get with a proper access point operating in 'infrastructure' mode with beamforming, hanging from the ceiling near the center of the room. In the former case, bodily obstruction could make a HUGE difference, because the alternate paths might exceed more than 100 feet after factoring in multiple reflections... multiplied by all that potential multipath interference. In the latter case, the AP would constantly be probing around to find the ideal beam direction that maximizes signal strength and minimizes multipath interference... simultaneously increasing signal quality and minimizing echo-like background noise.
* Supposedly, 802.11ad has extraordinarily low latency compared to 802.11ac and older standards... like, 10 microseconds. Apparently, it's low enough to use as the transport layer for PCI Express as long as the transmission conditions are basically ideal (like, a wireless laptop dock that's nevertheless within a foot or two of the laptop, with clear line of sight between the antennas).
So... it looks like that with a proper access point in an intelligently-chosen location and a halfway-reasonable attempt to keep the client's antenna in a reasonably visible location, you could feel absolutely confident of sustaining 1gbps just about everywhere within 10-20 feet of the AP, 2gbps nearly all the time, and 4gbps most of the time. Of course, this all assumes that the Quest has a viable interface to the outside world, and that its USB port (for example) would not itself be the primary bottleneck.
---
Side thought... I'll be absolutely SHOCKED if a future teardown of the Quest doesn't reveal the presence of at least a few "test" pads on the circuit board that expose the signals needed to connect Qualcomm's 802.11ad chip and/or implement Rift-like HDMI connectivity, EVEN IF Oculus has literally no intention of ever officially supporting their use by end users. The fact is, it's a lot cheaper to have one of their engineers grab a production Quest and hack it into an experimental prototype than to go through the effort and expense of building one-off prototype headsets for internal use from scratch. Chances are, the Quest prototypes that were being made a few months before it went to manufacturing (and before the official specs were finalized) had circuit boards that implemented "everything" (802.11ad via Qualcomm's extra chip, a HDMI-like video link, etc), then later prototypes simply omitted the components to implement the unnecessary functionality. Even if they subsequently pruned away the component footprints, they almost certainly would have left the pads, just in case someone down the road at Oculus needed some in-house prototypes that re-implemented the desired functionality.