4 ms·

My understanding is that this wasn't particularly the issue, the main problem for PC use is that you'd need a VirtualLink adaptor, and that will hold back peopl

My understanding is that this wasn't particularly the issue, the main problem for PC use is that you'd need a VirtualLink adaptor, and that will hold back people doing much work on this.

scheeseman486 3y ago
VirtualLink isn't that exotic, just another USB-C Alt mode that has 4xDP lanes with USB3 signaling pushed over the USB2 pins. There's likely plenty of ICs on the market that could do it, just few products that implement the feature because there's zero demand. Given someone gets the PSVR2 working on PC there would be incentive among enthusiasts to create hobbyist-grade hardware at least.
Nursie 3y ago
Apparently the radeon RX6xxx series has them, and nvidia 2xxx cards with USB-C ports. Though yes, it seems unlikely there will be continued support.
- ChuckNorris89 3y ago
  Yeah, GPU manufacturers seem to have given up on supporting VR headsets.
  - jsheard 3y ago
    There's nothing to support, the VirtualLink consortium was dissolved and it's now a zombie standard with nobody at the wheel. The VR headset makers that originally backed it aren't supporting it anymore either, Valve announced and then cancelled a VirtualLink cable for the Index, and HTC/Oculus/Microsoft never made it as far as announcing any products.
    - ChuckNorris89 3y ago
      Ah, damn, I had no idea
    - wolrah 3y ago
      For anyone interested in why VirtualLink didn't go anywhere, one of the limiting factors was that USB 3 and above is entirely separate from USB 2 and below. Normal USB-C DP Alt Mode delivers either four lanes of DP and USB 2 or two lanes of DP and both varieties of USB. VirtualLink takes the 4xDP variant and upgrades the USB pairs to support 3.x modes, but that means if you plugged a USB 2 device in to the headset itself there'd be nowhere to route that traffic. A chip was developed to bridge the gap and allow USB 2 devices to connect to the USB 3 bus, but there were quirks and it ended up not being ready in time to be useful to the VR market. Valve eventually cancelled their adapter plans and that was the end of it. This Hackaday article has a lot of useful links including an open source breakout board design for the chip in question: https://hackaday.com/2022/03/07/a-chip-to-address-the-fundamental-usb-3-0-deficiency/ https://hackaday.com/2022/03/07/a-chip-to-address-the-fundam...
      - jsheard 3y ago
        I suspect the rise of good-enough wireless PCVR over commodity WiFi hardware didn't help matters either - being able to tether a headset over a single USB-C cable is mildly convenient but nowhere near as convenient as having no cable at all. Oculus and Pico both went down that route and it looks like Valve are also going to with their next headset.
        snvzz 3y ago
        >nowhere near as convenient as having no cable at all. AIUI the tech just isn't there for this. There's latency and packet loss to deal with, whenever there's a radio in the path.
        jsheard 3y ago
        The tech has been there for years, if you look at discussions of using the Quest for PCVR the default recommendation is to use WiFi streaming rather than the tethered USB mode. The trick is that the reprojection still runs on the headset itself so the critical part for avoiding motion sickness has minimal latency regardless of the signal quality.
  - pjmlp 3y ago
    The first time I saw someone using a VR headset was in 1994, the technology might have improved, but the adoption issues hardly changed.
    - MikusR 3y ago
      There are more propper vr headsets sold than Xbox series consoles.
      - pjmlp 3y ago
        That still doesn't make VR a success at scale. How many of them are actually being used and not stored in some closet?
        mrtranscendence 3y ago
        It might not make VR a success per se, but objectively there are more VR headsets per capita sold today than 1984.
        pjmlp 3y ago
        Sure, if we ignore that after all these decades it is as much of a success as 3D TVs.
jsheard 3y ago
That and the fact that the PSVR2 uses camera-based SLAM tracking, similar to the Oculus Quest, and reimplementing that from scratch is non-trivial to say the least.
- ChuckNorris89 3y ago
  Isn't the SLAM tracking processing done on the headset itself and then the computed movement data sent to the PC/PS5 via USB?
  - jsheard 3y ago
    Is it? I haven't seen it definitively confirmed either way, but intuitively it would make more sense to do SLAM on the PS5s main processor rather than driving up the PSVR2s BOM with an on-board processor fast enough to do it. They have to stream the camera frames back to the PS5 anyway for the passthrough feature to work.
    - ChuckNorris89 3y ago
      >driving up the PSVR2s BOM with an on-board processor fast enough to do it The Quest 2 can do SLAM on the head set plus the 2*2K resolution 3D rendering in real time on a 4 year old QUALCOMM chip running Android OS on top and it costs less than the PSVR2. So Sony can definitely afford to put a processor in the headset just for the SLAM compute alone since the 3D rendering is done on the PS5. The chip in any optical mouse is a low resolution camera and DSP computing SLAM algos at hundreds of times a second and costs peanuts. VR headsets have more resolution to process and account for the third dimension in space but they also don't have to process the entire picture but just the differential movement of the LEDs captured in their blanking intervals, processed as white dots in B/W images, further simplifying things. So pretty sure the PSVR2 SLAM compute can be done on an ARM chip or FPGA worth ~ 10 USD nowadays.
- makomk 3y ago
  There's already a mostly-working open source SLAM implementation that people have been using on a few of the existing headsets. What complicates matters here is that at least feature extraction seems to run on the PSVR2 headset itself so the existing code can't just be dropped in as-is. (Also, whilst the controllers work in basically the same way as many other VR controllers using a constellation of IR LEDs, there's not really good open source support for any of those controllers either.)
makomk 3y ago
The PSVR2 turned out not to actually use VirtualLink, just more-or-less standard DisplayPort + USB 3 + USB PD over USB-C. Most of the compatibility issues seem to be from it requiring parts of the spec that aren't widely implemented at the same time, like 12V support and display compression. The main open question is how to actually switch it into VR mode so that it's possible to send a stereo image instead of 2D virtual cinema mode, and I don't think anyone is quite sure whether this helps with that at the moment.
- MuffinFlavored 3y ago
  > USB PD USB Power Delivery (USB PD) In case anybody else also wasn't familiar

My understanding is that this wasn't particularly the issue, the main problem for PC use is that you'd need a VirtualLink adaptor, and that will hold back people doing much work on this.

scheeseman486 3y ago

VirtualLink isn't that exotic, just another USB-C Alt mode that has 4xDP lanes with USB3 signaling pushed over the USB2 pins. There's likely plenty of ICs on the market that could do it, just few products that implement the feature because there's zero demand. Given someone gets the PSVR2 working on PC there would be incentive among enthusiasts to create hobbyist-grade hardware at least.

Nursie 3y ago

Apparently the radeon RX6xxx series has them, and nvidia 2xxx cards with USB-C ports. Though yes, it seems unlikely there will be continued support.

ChuckNorris89 3y ago
Yeah, GPU manufacturers seem to have given up on supporting VR headsets.
- jsheard 3y ago
  There's nothing to support, the VirtualLink consortium was dissolved and it's now a zombie standard with nobody at the wheel. The VR headset makers that originally backed it aren't supporting it anymore either, Valve announced and then cancelled a VirtualLink cable for the Index, and HTC/Oculus/Microsoft never made it as far as announcing any products.
  - ChuckNorris89 3y ago
    Ah, damn, I had no idea
  - wolrah 3y ago
    For anyone interested in why VirtualLink didn't go anywhere, one of the limiting factors was that USB 3 and above is entirely separate from USB 2 and below. Normal USB-C DP Alt Mode delivers either four lanes of DP and USB 2 or two lanes of DP and both varieties of USB. VirtualLink takes the 4xDP variant and upgrades the USB pairs to support 3.x modes, but that means if you plugged a USB 2 device in to the headset itself there'd be nowhere to route that traffic. A chip was developed to bridge the gap and allow USB 2 devices to connect to the USB 3 bus, but there were quirks and it ended up not being ready in time to be useful to the VR market. Valve eventually cancelled their adapter plans and that was the end of it. This Hackaday article has a lot of useful links including an open source breakout board design for the chip in question: https://hackaday.com/2022/03/07/a-chip-to-address-the-fundamental-usb-3-0-deficiency/ https://hackaday.com/2022/03/07/a-chip-to-address-the-fundam...
    - jsheard 3y ago
      I suspect the rise of good-enough wireless PCVR over commodity WiFi hardware didn't help matters either - being able to tether a headset over a single USB-C cable is mildly convenient but nowhere near as convenient as having no cable at all. Oculus and Pico both went down that route and it looks like Valve are also going to with their next headset.
      - snvzz 3y ago
        >nowhere near as convenient as having no cable at all. AIUI the tech just isn't there for this. There's latency and packet loss to deal with, whenever there's a radio in the path.
        jsheard 3y ago
        The tech has been there for years, if you look at discussions of using the Quest for PCVR the default recommendation is to use WiFi streaming rather than the tethered USB mode. The trick is that the reprojection still runs on the headset itself so the critical part for avoiding motion sickness has minimal latency regardless of the signal quality.
- pjmlp 3y ago
  The first time I saw someone using a VR headset was in 1994, the technology might have improved, but the adoption issues hardly changed.
  - MikusR 3y ago
    There are more propper vr headsets sold than Xbox series consoles.
    - pjmlp 3y ago
      That still doesn't make VR a success at scale. How many of them are actually being used and not stored in some closet?
      - mrtranscendence 3y ago
        It might not make VR a success per se, but objectively there are more VR headsets per capita sold today than 1984.
        pjmlp 3y ago
        Sure, if we ignore that after all these decades it is as much of a success as 3D TVs.

jsheard 3y ago

That and the fact that the PSVR2 uses camera-based SLAM tracking, similar to the Oculus Quest, and reimplementing that from scratch is non-trivial to say the least.

ChuckNorris89 3y ago
Isn't the SLAM tracking processing done on the headset itself and then the computed movement data sent to the PC/PS5 via USB?
- jsheard 3y ago
  Is it? I haven't seen it definitively confirmed either way, but intuitively it would make more sense to do SLAM on the PS5s main processor rather than driving up the PSVR2s BOM with an on-board processor fast enough to do it. They have to stream the camera frames back to the PS5 anyway for the passthrough feature to work.
  - ChuckNorris89 3y ago
    >driving up the PSVR2s BOM with an on-board processor fast enough to do it The Quest 2 can do SLAM on the head set plus the 2*2K resolution 3D rendering in real time on a 4 year old QUALCOMM chip running Android OS on top and it costs less than the PSVR2. So Sony can definitely afford to put a processor in the headset just for the SLAM compute alone since the 3D rendering is done on the PS5. The chip in any optical mouse is a low resolution camera and DSP computing SLAM algos at hundreds of times a second and costs peanuts. VR headsets have more resolution to process and account for the third dimension in space but they also don't have to process the entire picture but just the differential movement of the LEDs captured in their blanking intervals, processed as white dots in B/W images, further simplifying things. So pretty sure the PSVR2 SLAM compute can be done on an ARM chip or FPGA worth ~ 10 USD nowadays.
makomk 3y ago
There's already a mostly-working open source SLAM implementation that people have been using on a few of the existing headsets. What complicates matters here is that at least feature extraction seems to run on the PSVR2 headset itself so the existing code can't just be dropped in as-is. (Also, whilst the controllers work in basically the same way as many other VR controllers using a constellation of IR LEDs, there's not really good open source support for any of those controllers either.)

makomk 3y ago

The PSVR2 turned out not to actually use VirtualLink, just more-or-less standard DisplayPort + USB 3 + USB PD over USB-C. Most of the compatibility issues seem to be from it requiring parts of the spec that aren't widely implemented at the same time, like 12V support and display compression. The main open question is how to actually switch it into VR mode so that it's possible to send a stereo image instead of 2D virtual cinema mode, and I don't think anyone is quite sure whether this helps with that at the moment.

MuffinFlavored 3y ago
> USB PD USB Power Delivery (USB PD) In case anybody else also wasn't familiar