>>2327>is the technology actually going to get much better?VR hardware, for all intents and purposes, has the potential to improve, but I don't believe any mass market devices --
at competitive and affordable prices -- will significantly improve within the next few years.
If I really had to guess, I would say that within the next 2-3 years it's possible that Meta will release a 2560 x 2560 per eye headset, likely running at 90Hz. 120Hz if you're optimistic. It's unlikely that it would include face- and eye-tracking since Meta has said that they cancelled their plans for a Quest Pro 2. FoV would likely remain the same as the Quest 3 at 110° HFov x 96° VFoV. Price would likely be the same as, or potentially higher than the Quest 3. General design is likely to follow the trend of the Quest having removable facial interface and headstrap, allowing for third-party accessories, as opposed to having an integrated battery counterweight headstrap (like seen on the Quest Pro). I would expect that Google/Samsung, Lenovo, and potentially Apple are all targeting similar specifications.
To understand VR development, you first have to understand that there are two philosophies of VR: mobile SoC powered standalone VR, and PC powered wired VR. The hardware exists for >=150° FoV, for >=120Hz displays, for high contrast uOLED displays, for 4K x 4K per eye resolution, for face-, eye-, and hand-tracking, inside-out tracking, and wireless VR. Putting all of that into a single headset, however, has various challenges.
The PC side of things is currently much closer to that than standalone is, but everyone can see that base station SteamVR tracking is a dead-end and that inside-out, wireless VR is the future; particularly as Valve has ceased production of SteamVR 2.0 base stations and given HTC full rights to produce them in their place. PCVR, as a result, is currently in a hardware lull, waiting for Valve to release the headset they've been working on for the last few years to see what they do. Valve, and SteamVR, is really the linchpin that PCVR relies on for hardware direction and software support, which is why the rumors about what they're doing with their next headset are so significant. Regardless, PCVR tends to appeal to more niche, high-end users so they can easily charge >=$1K for a headset by itself. The options are essentially Bigscreen Beyond ($1000, headset only, ~$600 SteamVR basestation and controllers setup required in addition), Pimax ($1000-2000 range, headset only, requires SteamVR setup), and Varjo ($1-4K, enterprise subscription required for some headsets, SteamVR setup required), and Valve Index (completely out-of-date -- nearly 6 years old -- overpriced at $1K, and now completely outclassed by the Quest 3).
Standalone, by contrast, is much more mass market focused, and price conscious. Currently, the biggest thing holding it back is mobile SoC performance: high resolution displays are available currently, but there aren't many SoC's available that can do much more than around 2160 x 2160 per eye. Apple's Vision Pro is standalone and runs at ~3660x3200 @ 100Hz, but to do so they essentially put a desktop class M2 processor and a co-processor inside and the headset itself costs $3500 -- but in typical Apple fashion, they refused to design the headset with controllers so it's unusable for gaming. As a result, if you wanted to use it for PCVR and have controllers, you could potentially integrate it into a PC SteamVR-tracked setup, but that would cost somewhere in the range of an additional $750. Standalone, wireless PCVR is also limited by WiFi development. By my hazy calculations, 2160 x 2160 @ 120 Hz per eye would require 31Gbit/s, which... That's a far cry even for WiFi 7 6GHz, which tops out at 5.8Gbit/s. So, naturally, wireless standalone PCVR is based on transcoding on the PC and decoding video on the headset. The primary app for which is Virtual Desktop, which tops out at 200Mbit/s. You can at least use HEVC 10-bit encoding, or AV1 10-bit encoding, but either way standalone PCVR streaming is certainly compressed compared to wired PCVR. It's a bit better than YouTube compression, I would say, and it's something one can get used to and ignore. At most, you may notice color banding. Wired PCVR snobs will say the compression is unbearable, but I wouldn't put much stake in their opinions. The one truth, however, is that all of that transcoding and decoding and wireless sending and receiving does add some additional latency. For a typical WiFi environment, at the maximum bitrate and resolution, it's somewhere in the 50ms range. For a very poor WiFi environment (like in a multi-story apartment with overlapping WiFi signals), and using the 2.4GHz band as opposed to the 5GHz or 6GHz band, it could lean towards 120ms of latency -- this would be an usual, worst-case scenario, however. So long as you have at least a WiFi 6 AX router, and the access point is located in the same room as you want to play VR, you should get more towards the 50ms latency range. Having an access point in the same room is very important. If you have a house, for example, and your access point is on the second floor and you want to play in the basement, that would be more akin to a worst-case scenario because of WiFi attenuation through multiple walls.
As far as the development of standalone VR itself is concerned, Meta is the biggest and most influential company in this space, but there's been a lot less VR news since the AI boom. The Quest 3 was in the pipeline for a while and released a little over a year ago, but since then there's basically been a drought in terms of new hardware news and releases, with the exception of the Quest 3s which is essentially just a Quest 2 revision. Again, lots of people are putting their faith in rumors that Valve is working on a VR headset and hoping it gets announced some time this year to introduce more interest in VR. The biggest shakeup in standalone VR is that Meta has said that they'll allow other manufacturers to use their Horizon OS, which is the OS that the Quest headsets use, for their headsets going forward. We know that Lenovo is working with Meta to release their own VR headset using their OS. Meta's Horizon OS is Android-based, but Google has previously announced that they were working with Samsung to design their own headset using their own Android-based OS. Prior to the release of the Apple Vision Pro, there were rumors that Samsung was designing an Apple Vision Pro competitor, but since the Apple Vision Pro release I haven't heard any follow-up news on this front, especially since their announcement that they were working with Google to design a headset. Presumably, they've likely scrapped their Apple Vision Pro competitor headset and are now working on a Quest 3 priced headset in the $300-500 range. Apple has also said that they're working on a lower-priced more mass market VR headset, but it's anyone's guess what that will take the form of.