Tech Talk #4 : Standalone VR

Hey,

We talked about the Best Accessories for your VR Setup last week!

Thank you so much for providing your helpful advice. This will undoubtedly helps the community in enhancing the VR experience!

This week, we will be discovering :

The Standalone VR Headset

With the limited performance, how can this be improved?

Do you prefer the standalone version with a big FOV or a high resolution?

What other factors do you reckon are essential to the performance?

  1. Built-in wireless
  2. Tracking impact
  3. Software’s Platform
  4. Others

Feel free to leave a comment or suggest a topic for discussion.

Regards.

1 Like

With Pimax Developing a mobile AiO solution in the 12k pimax should work on bringing a Linux drivers and Opensource the Settings client.

It would be Awesome as well if the Team started to work on an AiO module idea that the Team mentioned in the KS as an external say belt device could also be a way to bring wireless to current pimax hmds.

Since pimax failed to make the wireless module a reality after saying in past years it would be demoed soon.

2 Likes

I believe it depends on your target groups and long term goals.
If you address the enthusiasts or industry which have different demands.
If you offer something with open standards or open connectors for the public, industry will adopt it quickly, they wait for it.
If new headset allows wireless, even on lower bandwidth / resolution but acceptable wide FOV AND AR features like camera integration, that would be a potential game changer.
On the other hand, then, we enthusiasts will not get these devices due to market demands, so drop it.
For us enthusiasts, I believe you may stay on outstanding features like wide FOV together with high resolution, that´s what I am paying for and willing to pay for, even for wireless but then with lower resolution. Maybe provide high bandwidth wireless as extra optional feature, at least allow to make use of such future devices. We will see WiFi7+ 5G connections which may be perfect for home use AND industry in the future, maybe not from PIMAX, but with PIMAX headsets.
The software platform: Linux for Industry, Windows for consumers. If you provide Linux, even steam users may profit.

2 Likes

I think what you mean here is that limited performance is a given, so what is the least painful to sacrifice compared to when it’s tethered to a PC.

For me, the priorities for any VR headset are:

Frame Rate (up to 90Hz) > FOV > Clarity > Frame rate (above 90Hz)

So I definitely think standalone should sacrifice resolution first. And that is, in fact, what I do already with my 8KX if I’m in a heavy VRChat world with 40 avatars or whatever. I’ll drop the resolution in SteamVR on the fly to keep the frame rate up.

I consider frame rate the highest priority up to 90Hz. I would like to have the option to sacrifice FOV on the fly if I need to as well. I wish this was an option on the 8KX with SteamVR, but it requires restarts. I would sacrifice resolution first, and then if that’s not enough, I’d like to be able to drop FOV, too.

Native standalone applications should be required to handle on-the-fly FOV changes. I’d also like to have this ability tethered to the PC with SteamVR/OpenXR.

#1 Wireless: I don’t really need mobile capability in 12K. It’s too large and valuable, and I have a smaller and more expendable Quest 2 for that kind of purpose.

My primary interest in standalone would be using wireless while in my playspace at home while tethered to my PC. For this purpose, wireless needs to be equal to wired. If it offers less FOV or resolution or adds latency, forget it. I’d rather deal with the cable. At least for action games.

I spend a lot of time in VRChat, and I’d be more willing to accept some limitations to latency and resolution there. VRChat with full body tracking is also where eliminating the cable would be the most useful. The cable isn’t much of a problem while I’m just standing, but in VRChat I may be dancing, sitting, laying down, etc.

So a particular feature I’d like to see is the ability to quickly and easily switch between cable and wireless. In fact, if that could be done on the fly, that would be really ideal. Then I could switch between them as the situation warrants.

One of the big values of this would be dealing with batteries. If the battery is low, you could just switch to cabled on the fly.

#2 Tracking: The Quest 2 style tracking of the hand controllers from the cameras on the headset is, to me, the biggest weakness of the Quest 2. This makes for meaningfully inferior tracking compared to basestation tracking. In gaming, it offers noticeably less accuracy when aiming guns. In VRChat, the limited tracking volume really hurts. Quest 2 players walk around with T’rex arms because even having your hands down at your sides puts them outside of the tracking FOV. If you’re using full body tracking, your hand positions are going to want to leave the tracking FOV area a lot because you’re not just limited to the typical standing and having your hands in front of you at all times. How often are your hands in front of you when you’re laying down? How often are your hands behind you while you’re dancing?

I was disappointed to see that the 12K will implement the same kind of inside-out tracking for the hand controllers as the Quest 2. I have no interest in it, and I’ll 100% replace it with the basestation face plate. If the basestation face plate is not available at launch, then I won’t be buying the 12K until it is.

Now if the 12K was offering controllers with cameras in them to do their own inside-out tracking in the manner that Project Cambria is rumored to do, I believe that this approach will rival or exceed basestation tracking. I’d be very interested in that. If it works well enough, I may be able to retire my basestations.

#3 Software platform: Since I have little need for mobile VR, I’m primarily interested in SteamVR games. However, there are games that come out exclusively for Quest 2 that I’d like to play. I have a Quest 2, but playing native games on it is meaningfully inferior to gaming on my 8KX in every way (except not having a tether cable). So I don’t really find myself actually playing these games.

If the 12K can offer these exclusive Quest 2 games on a better platform than the Quest 2 so they have wide FOV, etc that would be of interest.

Since the 12K also runs XR2, it would be really nice if it could actually emulate a Quest 2 fully such that games don’t even know they’re not running on a Quest 2. It would be nice to be able to give Meta the finger with running their Quest 2 exclusives on a different (and better) platform. I feel dirty whenever I use my Quest 2, and I’d rather avoid rewarding Meta for its anticompetitive practices as much as possible.

#4 Passthrough cameras: Going from the Index to the 8KX, I definitely missed having passthrough cameras and being able to overlay real world and virtual world. I needed this primarily when I move around in RL because I’m picking up something like a water bottle. Or if I’m transitioning from standing to laying down or vice versa (full body tracking). In the 8KX I end up having to do this blind, which is a little dangerous.

I imagine the 12K will have this ability with its tracking cameras. However, I won’t be using that face plate. I want to use the basestation tracking version. I’d like this version to have passthrough cameras. I’m willing to pay extra for that.

5 Likes

This is interesting. In theory if they do there Inside-Out tracking so-to-speak right the SteamVR tracking plate shouldn’t be necessary to use Index/Pimax/Other SteamVR tracked controllers. As they could do “good” hybrid tracking similar to how folks have used Index Controllers on non SteamVR tracked headsets. And the Bonus of course is what you mentioned with having the passthrough cams.

Now I would like to see there controllers follow a similar idea of Etee with the Tracking ring cam be swapped out with a SteamVR Tracker or other potential tracking solution. An idea that was discussed during the Pimax KS.


Now on a Seperate note as I have seen others comment before use more standardized non propietary connectors on the hmd side. Use things like Usb-C and Mini DP.

I would also like to see something like a separate micro SD for Firmware so easy backup Firmware. With Watchman Dongles also being swappable for easy repair and/or upgrade.

1 Like

that´s what I mean with AR, exactly. even if it´s just an option. THIS is what industry waits for, too.

1 Like

+1 for passthrough cameras, needs to become standard on all headsets.

1 Like

There is a current issue with the 12k Reality design as far as Simmers. Simmers often have comfortable Sim Chair setups. Am Often Complaint is the Ridgit Headatraps. Both the Vive Pro and the MAS based on the Vive Pro are bulky on the back that interferes with the comfort of having a headrest.

The 12k will be even worst as this is where the battery pack inserts.

So for these folks will require a different head mount option that if the Battery pack is needed as mentioned with one of the current prototypes will need to be able to place the Battery pack maybe on a tether to be placed elsewhere or an external power adapter.

The question too will be is the current proposed headstrap accomodate headphones better for those whom want to use there own audio headset?

2 Likes

That’s a point. However, mixing inside-out tracking with basestation tracking like that is a hack, not a real product feature. It requires 3rd party software to work (playspace calibrator), and also manual edits to SteamVR files. Calibration is not perfect and drifts over time.

I used to do this when I used FBT with an Oculus Rift CV1. I was really happy to not have to deal with it anymore when I got my Valve Index. I’m not eager to get back into using this hack since it’s not the kind of thing you only setup once and never have to deal with it again. It requires ongoing effort.

Also, I think Pimax needs to get away from any of this kind of thinking entirely. It’s products shouldn’t require hacks and tinkering in any way. As long as that kind of thing is incorporated into their planning and product vision, their customer base will be limited to a very narrow market.

So I would like to see pass through cameras on the basestation version of the front plate so that it’s a real product feature with no hacking required.

2 Likes

The team and have been working on this for a long time and we hope to have an announcement soon

1 Like

Main Topic - High Fidelity (Both High Resolution and FOV), primarily FOV depending on the game… creates a more immersive experience!

Just as important to me is Great nighttime inside-out (camera) tracked performance so, I can use tracked VR controllers and not worry about Light Bleed (from Outside Headset).

Also, Long battery life, or actual Quick-change Battery swap ability – Extra Battery out of the box for said Quick-change battery function. On that note, “standard” off the shelf battery replaceable VR controller batteries are a must for quick-swapping especially since Sword Controllers (and new Pimax REALITY controllers) most likely both do not include extra batteries in the box.

AIO (All-In-One)

-Included Picture & Video player (Loops seamlessly, A-B Repeat, and Speed settings)+VR Video related perspective fixes similiar to DEOVR or Playa Video
-Interesting feature Seamless AIO to PC mode transition (for supported games)
-Long battery life in Wireless AIO mode (3hr+ Lowest settings)
-Lower possible Refresh Rates (Battery life)
-AIO mode allowing FOV setting dependent upon specific app(s) in use
-When AIO mode switches to PC/Standalone mode not detaching from wireless/connected wireless devices
(Ex. VRchat Standalone mode switching directly into PC App VRchat)
-Powerful fans to keep both headset itself cool and separately keep our faces comfortable
-Adjustable Fan curves & maximums per application

—On Headset Buttons for Interaction (Accept & Cancel) useful when using simple WebXR webpages, WebXR games, and Video players - especially useful if you do not have a need for two controllers wasting their battery on just watching “Lightyear” or similar CG films… obivously once it lands on blu-ray!
PLUS on headset volume buttons

3 Likes

…If what you mean is that you want the 4 inside-out tracking cameras to be on the Steam LIGHTHOUSE cover then well that is already the case… just watch the shortened video on SweViver’s Youtube channel it shows this… the 4 cameras are always visible. The Mixed Reality cover is maybe the only one that might function weirdly due to what may be a limit of the XR2 HMD Reference Design (specifically the XR2 Chipset limitation)-something about number of cameras it allows, not sure if that limit is a hard limit or just a suggestion.

1 Like

A “simpler” solution might be just getting a Neck supporting Neckrest… still a bad solution… I know the back Headstrap is a personal issue of mine when just trying to rest my head in VR… must be real bad for long hour Simmers. I remember playing Forza 4 Horizon for hours (not in VR) and that was great just not nearly as the realistic feeling helped by VR, more intense I guess.

1 Like

What are “Watchman Dongles”?

1 Like

I totally agree, I wanted to be able to do this for the longest for various reasons Early days VR videos (Videos shot in 4K and below) sorta need this to not break immersion while newer VR videos are shot in 6K and above.

Plus a per app setting could also be useful.

Additionally changing FOV on the fly in-apps could allow for dramatic theater black border film cinematics in-games!

1 Like

They are what the controllers pair to in the headset.

1 Like

Ahh… thanks!

1 Like

Overlay compatibility, for developers to add hand-tracking based keyboard, mouse, web browser, VNC, command prompt, cloud computing, etc.

Android apps basically being ‘graphical interface as is and that’s it’ is very unhelpful for developers. We need something that lets us quickly and interactively write and debug VR apps, in the headset itself.

Otherwise, the standalone capability will be limited to running just some apps that are mostly tested with PCVR, and only the features that do not depend on things like cloud computing (or virtual machines, etc) will be available, at that.

Imagine Elite Dangerous, but with desktop computers, voice commands, and and some scripting, for the ship panels, instead of the buggy menus it has now. Both Standalone and PCVR should get that kind of functionality.

I have to admit that I care 0% about any stand-alone solutions in Pimax and would vastly prefer that research and development be spent making the PC-Based experience as good as possible. With other brand stand-alone VR solutions costing so little, anyone buying a Pimax HMD would realistically be able to pick up a second cheapie for anything stand-alone. There is simply no way (currently) that any chip will be able to produce the clarity I consider decent. I vastly prefer a clean, sharp picture. I push AC to my Pimax 8K-X at over 6000 horizontal pixels per eye, for example, which requires a 3080 Ti level or better level of performance (I use SOL, Rain Effects, etc.) I just have zero desire for any low-resolution or low-detail gaming and consider efforts to implement that to be a waste of time and resources.

I also have zero interest in tracking-solutions that don’t work as well as my current room-scale setup. Even a tiny bit worse would be unacceptable to me. IMHO everything required (at the HMD end) for base-station tracking should be included with the equipment shipped with the HMD by default; we should not be forced to pay for anything simply to keep tracking precision from regressing. Let us use our current tracking setup as-is with no additional costs. As stated by someone else above: I will NOT be buying any 12K until basestation tracking is a working product, just as I didn’t buy the 8K-X until 90hz was a functioning reality.

I do see a great potential of producing an HMD with a chip good enough to do the PP transforms in the HMD rather than at the primary GPU level. I don’t know if that’s technically even possible (likely those transforms HAVE to happen early in the rendering pipeline). But imagine the performance gains we could see in titles that currently require PP to render properly. Suddenly we would have a tremendous number of games that worked at Wide FOV and high Refresh rates properly. That would be very, very cool if doable.

In short: Pimax should prioritize the HIGH-END of features for the 12K FIRST and only when those features work properly, (and if they still really want to,) work on implementing features prominent in low-end VR. People wanting those low-end features aren’t and can’t be the 12K’s target market; not by a long-shot. All Pimax will end up likely doing is building a product that alienates potential buyers in the high-end VR market.

4 Likes

By my understanding, it should occur after rendering the frames, and just before-, or integrated into- the barrel distortion pass.
Were it moved to the HMD, we would need to transfer the huge “raw” frames to it, via cable or radio, rather than just the finished, display-size, ones.

1 Like