Oculus Quest 2

Soooooo, my newphew is really interested in VR after testing my Reverb G2, BUT, IT IS NOT a child device.

He have some knowledge but the troubleshoting required to make the G2 work sometimes is beyond him (for now).

So i ask, the visual difference between the G2 and the Quest2 is really that bad ?

Thanks.
 
Soooooo, my newphew is really interested in VR after testing my Reverb G2, BUT, IT IS NOT a child device.

He have some knowledge but the troubleshoting required to make the G2 work sometimes is beyond him (for now).

So i ask, the visual difference between the G2 and the Quest2 is really that bad ?

Thanks.
I have both and there's definitely a difference, but honestly I dont think it matters that much. Quest 2 is great for the ease of setup and use, I let my 6 year old use it sometimes but would never let him use the G2 because of the cord and setup. It's just so much easier to strap the quest to his head and let him walk around and look at robots in VR and not have to worry about tripping on anything.
 
Soooooo, my newphew is really interested in VR after testing my Reverb G2, BUT, IT IS NOT a child device.

He have some knowledge but the troubleshoting required to make the G2 work sometimes is beyond him (for now).

So i ask, the visual difference between the G2 and the Quest2 is really that bad ?

Thanks.

I cannot compare but the clarity of Quest 2 is excellent, you really have to strain your eyes against a solid colored background to see any SDE. I could not ask for more as far as resolution goes.

Also Quest 2 is dead simple device, and quite safe because there are no cords. As long as the kid understands how the Guardian system works, everything should be golden. Only downside is the Facebook account requirement but if either of the parents has one then that is not a problem.
 
Yep, Quest 2 it is then.

I'll try to convince him to create a separated Facebook account.

Thanks guys !
 
I still have an older fake Facebook account...does Facebook now require proof of your identity when signing up?
IDK, my facebook account is 12+ years old and still doesn't have my real name. I use it just fine on Oculus.
 
I still have an older fake Facebook account...does Facebook now require proof of your identity when signing up?

Honestly, when money is involved I would not risk fake accounts. Because if you ever lose access to your games (banned by accident, etc...) then you are screwed. Games you have bought from Oculus stores are lost forever.
 
I still have an older fake Facebook account...does Facebook now require proof of your identity when signing up?

I don't know, my account is super old too.

But people have complained about their accounts getting banned and losing all their Oculus games, and all of them were because they were using facebook accounts with fake information. Making a new account just for their Quest or whatever.
And yeah, if for some reason your account get messed up you're going to have a tough time getting it back if it's fake information.
 
According to this guy on Reddit when link is used both atw and asw is done remotely on the pc rather than the headset. ATW I believe it’s the rotational reprojection to keep us from getting sick when there is higher latency for head turning.

This would make sense rather than have the headset do rotational and the pc do positional.

early reports from lucky users who have been able to upgrade to the latest quest 2 build are reporting much smoother gameplay with airlink in comparison to virtual desktop.

https://www.reddit.com/r/oculus/com...&utm_medium=ios_app&utm_name=iossmf&context=3
Just finally got the updates working today, had to reset the headset experimental features ~5 times for the option to show up. It seems pretty good so far with the default bitrate, and will probably only get better as they work on it. It seems better than VD for lag, the end scene of Alyx syncs up pretty good with sound/animation.
 
Full 6dof ASW with complete Z-buffer? (ala ASW 2.0 on PC VR play)

AFAIK, the Z-buffer is not streamed over Link.

It would look like ASW 1.0 which does not need much computing power, and could fit on an XR GPU budget. So the reprojection looks worse, but has less lag.
I just tested AirLink (great stuff!) and have found out that ATW is done locally, while ASW is done remotely.
  • The rotational reprojection (aka ATW) is done locally on Quest 2 GPU
  • The positional reprojection (aka ASW) is done remotely on PC GPU.
At least this is what seems to be happening by default.

Pretty neat GPU local/remote "co-processing".

Currently, there is zero headturn stutter, regardless of whatever framerate the VR game is streaming at, and WiFi reception blips. The stutters from AirLink transients are seen in hand movements though (reprojecting that successfully stutter-free is a GPU challenge and not yet done in this generation).

You can force artifacting on this -- you can see this in action by forcing a fixed 100-200 Mbps WiFi speed on a 2.4 Ghz WiFi network (5 Ghz temporarly turned off) -- positionals lag badly (2 second lag!), while rotationals remain lagless.
 
Last edited:
I just tested AirLink (great stuff!) and have found out that ATW is done locally, while ASW is done remotely.
  • The rotational reprojection (aka ATW) is done locally on Quest 2 GPU
  • The positional reprojection (aka ASW) is done remotely on PC GPU.
At least this is what seems to be happening by default.

Pretty neat GPU local/remote "co-processing".

Currently, there is zero headturn stutter, regardless of whatever framerate the VR game is streaming at, and WiFi reception blips. The stutters from AirLink transients are seen in hand movements though (reprojecting that successfully stutter-free is a GPU challenge and not yet done in this generation).

You can force artifacting on this -- you can see this in action by forcing a fixed 100-200 Mbps WiFi speed on a 2.4 Ghz WiFi network (5 Ghz temporarly turned off) -- positionals lag badly (2 second lag!), while rotationals remain lagless.
I'm curious about forcing asw on the headset even though I believe it's 1.0 like you pointed out. Did you not get black bars on rotational turns when it failed to reproject? I get black bars on the right and left if I put the bitrate too high and i turn my head and it doesn't have enough info to reproject.
 
I'm curious about forcing asw on the headset even though I believe it's 1.0 like you pointed out. Did you not get black bars on rotational turns when it failed to reproject? I get black bars on the right and left if I put the bitrate too high and i turn my head and it doesn't have enough info to reproject
First, let’s clarify terminology.

I currently define “failed to reproject” in the rotational sense as rotational stutters (which does not happen).

But you may be defining “failed to reproject” as the existence of black bars (which has nothing to do with rotational stutter.

So I am saying it succeeded (because it eliminated stutter) but you say it failed (because black shows up at the edges). It’s a matter of definition of what denotes a reprojection success/fail.

That said, more accurately, it’s being asked to reproject far beyond its specs — but it is doing it successfully (under the stutter definition). It’s like trying to framerate-amplify 10fps to 90fps, so there is a lot of opportunity for black areas to appear.

The ideal scenario is it renders a bit beyond the viewport, to give extra room for rotational reprojection (e.g. 45fps to 90fps). But in this case, we’ve put some really intense duress, where the framerate is forced down to about 10fps, but it’s still successfully rotationally reprojecting whatever it can to full frame rate rotationally (despite missing image data at edges, the appearance of black areas).

A more everyday reprojection scenario would not be such intense duress on VR frame rate (i.e. trying to force it to cram 200 megabits per second over 2.4 gigahertz WiFi), and thus the black areas won’t appear.

However, it was useful to show how reprojection still de-stutters (ignore the black, focus on rotational stuttering temporarily — not the actual framerate inside the viewport, not the frame rate of the stuttering of the edges of boundaries, not the framerate of hand stuttering — but the frame rate of the image shifting around is staying at full frame rate within. The edges lag, but the position of the image data in the center of your view, stays consistently lagless, e.g. like an object in the far distance in the center of your vision).

This is easiest to understand with a completely static room — e.g. your Home area (avoiding animated menus or animated objects) where you can better focus on whether there’s rotational stuttering for stuff in the exact middle of your view — there’s simply none, despite the lagging/stuttering in the black-edge periphery — caused from the lagged transmission and the low frame rate.

Then again, in the real world we aren’t trying to cram 200 Mbps over 2.4 GHz wifi, to create an unusual artifacting situation that reveals reprojection limitations (continues to de-stutter successfully, but fails to hide the black areas that reveals) — we’re just doing that to force behavior that are easily observable.

As a related topic, I wonder how much extra area (beyond VR headset visible view port) can be rendered/transmitted to reduce incidence of black areas at edges during reprojection-in-duress scenarios.
 
Last edited:
The air link thing seems to work really well.
It really does as long as you have the proper setup and are on a 5ghz channel that isn't being used up and you are fairly close to your router/access point. I'm able to play in my room that has a wall between my router with my pc downstairs hooked up through Ethernet via a moca adapter and it works very well. I share channels 36-48 (using 80 mhz width) with my sister in law and her husband that live next door and have a nighthawk router that they blast at full power with only their printer being wired and still don't get enough interference to mess up my gameplay. I checked and using airlink takes up about 35-45 % airtime channel usage by itself.
 
Last edited:
It really does as long as you have the proper setup and are on a 5ghz channel that isn't being used up and you are fairly close to your router/access point. I'm able to play in my room that has a wall between my router with my pc downstairs hooked up through Ethernet via a moca adapter and it works very well. I share channels 36-48 (using 80 mhz width) with my sister in law and her husband that live next door and have a nighthawk router that they blast at full power with only their printer being wired and still don't get enough interference to mess up my gameplay. I checked and using airlink takes up about 35-45 % airtime channel usage by itself.
My router is right in the next room. Yeah, I just bought the new 5ghz 5.4gb whatever router. Stupid thing was expensive.
 
Definitely don't buy that for the Quest 2. That's only AC, AX is better.

If you're only going to use the router for the Quest 2 and air link the best deal is a $70 TP-Link AX router.
It's a good move to get AX, but a superior AC can exceed an inferior AX, latency-wise and stutters-wise. The published VR benchmarks show this, too.

My experience is some $300 AC routers with 8-antennas can -- in several cases -- outperform a $70 AX router with only 3-antennas.

The multiple antennas seem to catch WiFi signal better during head rotations and wild movements -- preventing those single-stutters caused by momentary WiFi latency spikes. The extra antennas (and a lightly WiFi router CPU) is more important than AC versus AX.

As you play crazy fast movements in VR-optimized FPS shooter game where you often rotate your head very fast to look behind you -- the extra antennas seem to help with those momentary WiFi occulsion moments, eliminating a lot more of those 1-frame stutters.

That said, try not to use your primary router if possible. Someone suddenly starting/stopping Netflix, or a download elsewhere in the house, can create a few occasional stutters in your VR experience because of either airwaves congestion or router CPU loading.

Even a 5ms latency spike inside a router can push a VR frame to the next VR refresh cycle and fail to be fully projected successfully -- those single-frame stutters you occasionally see on AirLink. Even those speed changes from 864 Mbps to 432 Mbps WiFi can create a temporary latency spike (speed shifting on WiFi) caused by your headset blocked by your body when you face away from a single-antenna WiFi unit. Those tiny latency spikes create single-frame stutters!

So if you are stutter-averse and have a big budget, have a separate 8-antenna powerful-CPU gaming router for your VR playspace.

By having extra antennas to catch stronger WiFi signals from your blocked head turns better -- you may reduce your "occasional" 10 single-frame microstutters per minute to "virtually nonexistent" 0 or 1 stutter per minute during aggressive VR movements this way.

Keeping a perma-fixed-bandwidth 5 GHz WiFi connection continuously, even when you happen to be randomly facing away from the router kneeling over (with your whole body blocking your headset from line-of-sight from router in a room with walls that poorly reflects 5 GHz signals).

Locking the WiFi speed (e.g. 432 Mbps vs 864 Mbps rather than letting it be dynamic) to prevent microstutter-inducing WiFi modem speed shifting can also be an additional tactic to get a stronger signal too, but if you're firehosing the full 200 Mbps data rate over AirLink, you may need to get the most stable 864 Mbps WiFi 11ac/11ax signal you can muster.

Ask yourself how much head/room movements you plan to make. Ask yourself how picky you are about one single-frame microstutter every 5 seconds.

If you're a sitdown gamer playing sim racing, tele-prescence, side-scrollers (VR versions of sierra quest style games), this is less important -- as long as your VR-specific WiFi router has decent unloaded CPU not shared by other Internet users. With minimal head movements and fewer chances for your head to be blocked by your body -- you may even be able to make do with just a two-antenna router, builtin WiFi, or piggybacking off a low-end ISP's router.
 
Last edited:
If it's a router dedicated for the Quest 2, then yeah, it's good enough. Complete overkill though. I wouldn't buy one just for the Quest 2. Way too expensive. An Asus RT AX-55 is quarter of the price and comes highly recommended for use with the Quest 2.
This is what I already own. So its better to have a separate router? Several other devices do use this router as well.
 
This is what I already own. So its better to have a separate router? Several other devices do use this router as well.
I made sure all of my streaming devices other than ipads and iphones are wired. I had no issues with nieces using wifi when I was playing some games on air link. My router has a pretty powerful cpu though, so that helps. If your ac or ax router has a weaker cpu then you will probably want it to be dedicated.
 
This is what I already own. So its better to have a separate router? Several other devices do use this router as well.

It might be good enough as it's a very good router. It will depend on how far away it is from where you will be using the Quest 2 and how many other devices in your home are using the 5Ghz network.

It is recommended to have a dedicated router for the Quest 2 in the same room. But, you might find that your current setup works fine.

Try the setup you have and if it doesn't work you can always buy another router later.
 
This is what I already own. So its better to have a separate router? Several other devices do use this router as well.

Just try it if you already have it. What are you asking us for?

Yes it's better to have a dedicated router, but it might be fine not to. Yes AX is better than AC but you could be fine with AC too, and a high end expensive AC router can out-perform a budget AX router.

If you already own something just try it out before buying something new.
If you don't own anything the best deal is is getting a TP-Link AX1500 for $70, wiring it directly to your PC and only using the Quest 2 wirelessly with it. It will work great.
If you don't care about a budget you can buy ultra high end AX routers from ASUS that work even better and cost way more money than your headset, and probably work fine with a bunch of devices using it all at once. You may never notice the difference over the budget option though.

I have both a higher end ASUS AC router with 6 antennas that I use for my whole house and a super cheap TP-Link AX router with 4 antennas I use just for PCVR.

I used to use the ASUS router for PCVR and it actually worked pretty good, it would just occasional get spikes because other people also use it for their devices. After I switched to a dedicated cheap TP-Link AX router I have never gotten any spikes.

These are the speeds I get with my routers:

Speeds right next to each router:
ASUS: 300/650
TP-Link: 350/850

Speeds at medium distance:
ASUS: 150/300
TP-Link: 250/75

These are just average speeds. The TP-Link router that I only use for my Quest 2 is very consistent, the ASUS router I use for my entire home isn't as consistent because it depends what all the other devices connected to it are doing. If the ASUS was dedicated to the Quest 2 it would probably be more consistent because it has more antennas and probably a better processor, but it isn't setup up to be used that way.
Since the budget router is only being used by my Quest 2 I have it positioned near where I play (or could just move it if I needed to) and always get good speed. That's a big advantage in having a dedicated router to PCVR. And at $70 it's cheaper than the official oculus link cable.
 
Again, wifi 6 router is NOT required for wireless VR. I use a a cheap 5g dual antenna router that my internet provider sold me and it is not even dedicated to VR. Not sure who made it as it is a rebrand, but it is just connected via ethernet and in line of sight and I do not have any severe latency, stutter or artifacting issues. It works flawlessly as far as I can tell.
 
The biggest improvement. I've noticed is not having multiple devices on the router while you are using vr.
 
Been playing Onward with the Quest 2. Looks great, but I really do not like the tracking. I miss my original Rift with the 4 sensors.
I do hope they add an option for the light towers. That would really be great.
 
Been playing Onward with the Quest 2. Looks great, but I really do not like the tracking. I miss my original Rift with the 4 sensors.
I do hope they add an option for the light towers. That would really be great.
I find the tracking better than the original Rift, but I do make sure there’s a few lights on in the room while I play — the tracking can improve 10x between a dim room and a well lit room.
 
I find the tracking better than the original Rift, but I do make sure there’s a few lights on in the room while I play — the tracking can improve 10x between a dim room and a well lit room.
Not sure if you are kidding or not. Full 4 sensors vs the built in on the hmd is better? Have you tried 4 sensors? It detected everything. The built on to hmd can only see so much. Light is not going to help here. There are some circumstances that it just can't see. Try looking forward and having the controller on your chest while crouched or bent over to scan areas. The only thing better than the 4 sensors are the light houses.
 
Not sure if you are kidding or not. Full 4 sensors vs the built in on the hmd is better? Have you tried 4 sensors? It detected everything. The built on to hmd can only see so much. Light is not going to help here. There are some circumstances that it just can't see. Try looking forward and having the controller on your chest while crouched or bent over to scan areas. The only thing better than the 4 sensors are the light houses.
Just to be clear:

I'm not talking about range of tracking.
I'm talking about stability of tracking (jitter-free hands).

e.g. playing a surgical-manipulation game with your hands always in front of you.

Yes, Oculus Rift is superior for range of controller tracking (i.e. controllers behind back).
Yes, Oculus Rift is superior for dark-room tracking (i.e. totally dark rooms).
Yes, Oculus Rift has more room-lighting flexibility (i.e. doesn't care as much about your room lighting)

But....have you played REALLY surgical manipulation tasks with 0-millimeter-league precision?

Oculus Quest 2 often becomes superior in stability for hands-in-front surgeon manipulation (millimeter-perfect zero-jitter surgical tracking in a brightly lit room). e.g. ~4000 lumens of PWM-free LED lighting in the room. In a really brightly lit room, Quest 2 blows away a large 10x10 Rift RoomScale in this specific surgeon type task. Tracking is rock stable if you like to play "surgeon-style" games where your controllers are always in front of you -- zero millimeter jitter, with uncannily accurate 1:1 sync between VR and real life. While I only have 3 sensors, there can still be jitter issues even with 4 in a large room. Where all the sensors have to be far away to create a large RoomScale (i.e. 10x10 or bigger)

What kind of games do you play?
Are you merely blasting away in games like Half Life Alyx?
Do you play gamers that require reaching behind you?
Or do you play games that requires millimeter surgeon-style manipulation in front of you?

So, because of that, Quest 2 and Rift certainly have their have superiority pros/cons in their tracking. Oculus Rift definitely is not universally spuerior in tracking.

We're not talking about those Nerfed manipulation stuff (big buttons, big handholds, etc) or big magic marker drawings -- I'm referring to real-life-sized tiny VR objects, tiny buttons, and other tiny maipulatable things. That really stands-out the hands-in-front tracking capabilities, where you have to be really precise with your movements.

It's possible if you use a very tiny room with all 3-4 sensors close together, that the hands-in-front jitterfree precision becomes similar to Quest 2 brightly-lit-hands-in-front (for millimeter surgical manipulation). However, the bigger the room, the less accurate Rift tracking becomes as all the cameras become farther and farther away from your hands. Whereas, with Quest 2, you've got room-size flexibility with similar hand accuracy with similar artificial lighting.

With a large RoomScale setup (sensors in four corners of room), do these two tests:

Test #1: Here's another test -- with both Rift and Quest 2 -- touch two controllers together at random locations (top-bottom, side-side, bottom-side, loop-to-loop, etc). With Quest 2, they touch together perfectly zero-millimeter no matter how I touch the two controllers together. No airspace and no intersects (not even a single millimeter!). Good for VR games that requires surgical manipulation of objects. It's hard to achieve that with an original Oculus Rift with sensors far apart in a bigger room -- you might be using an adequately small room.

Test #2: Another good test is handwriting tiny text with thin lines using a VR pen on a VR notepad (not marker-thick lines, actual thin lines). Easier to do with a Quest 2 in a well lit room -- more perfect handwriting with smoother cursive in a brightly lit room. With a Rift, the lines sometimes becomes as if you're trying to handwrite on a notepad in the backseat of a moving car. The tracking accuracy can vary a lot for sure, and a Rift might get similar if you're putting two of the sensors on your desk, pointing at your hand controllers -- but that, of course, sometimes precludes combining large roomscale simultaneously with surgeonesque hands-in-front tracking.

Also, if you hate Quest 2 controllers "disappearing" after exactly 15 seconds of keeping hands still, that's a totally different subject to whatwever we're debating -- that is the automatic battery power management that also happens to be adjustable if you are annoyed by that (there's settings to adjust).

Yes, in a dim room where Quest 2 still works but jitters bad, the Rift definitely tracks better.
Yes, one can keep moving your Rift sensors (move them closer together to play the most surgical-manipulation game, move them farther apart to play Half Life: Alyx).
Yes, most VR games don't need the precision of Quest 2 in a brightly-artificially-lit-room after all, so you might not have noticed the excellent hands-in-front controller tracking precision the Quest 2 can scale to with sufficient lighting.
But the point remains: The uncanny zero-millimeter-accuracy hands-in-front tracking capability optionally exists for Quest 2 if you decide to optiimze room lighting -- and RoomScale size has zero effect on hands-in-front jitter.

When you're staring at a surgical-manipulation task, 4 cameras only inches away from both of your controllers, tends to produce submillimeter surgical precision when there's plenty of lumens of artificial lighting. Nothing beats that -- 4 sensors 10 feet apart in a 10x10 room is not going to accomplish that. That's why I want Valve Index 2 to have optional simultaneous Lighthouses AND inside-out tracking.

Yeah, it all depends on your controller priorities. The venn diagram of line-item superiority definitely overlaps between Rift and Quest 2 hand controllers.
 
Last edited:
I had a Rift CV1 3-sensor setup for about 2+ years and agree that the Quest 2's inside-out tracking is way more stable/less jittery for me. Yes, the range of motion is inferior, particularly directly below or behind the HMD, but accuracy and jitter is where it counts for me. Honestly expected that the CV1 would have the Quest 2 beat in this regard, but the Quest 2 instantly converted me within minutes of first using it.
 
I haven't had any tracking issues on my quest 2.
On the original vive with base station. I would sometimes get jitter and have funky stuff going on or just have them not be detected.
 
The more light the better the tracking. Remember the headset uses cameras for the tracking. Cameras work much, much better with good lighting.

People think, oh, I should turn off the lights so I get less light leak through the headset and get a more immersive experience. Or they're playing at night and don't turn on the lights because they don't know how the tracking works. Then they get poor tracking and think there is something wrong with the headset or that the tracking just sucks for all Quests.

They should display some sort of message or guardian type indication when it detects low light levels that negatively affect tracking and tell users, then explain the problem when they select it or something. Then they would know all you had to do is turn on the lights to fix it.
 
I wonder if I need more light? I have 2 bulb on and the sliding door blinds open. I am used to my Rift that did not care either way.
 
Light and contrast is important for the Quest 2. The cameras map out your surroundings by placing virtual markers on areas of contrast to aid in tracking your position. Playing a Quest 2 in an empty black/white room with perfect lighting might sound like an ideal VR space, but without contrast, the HMD will struggle to place useful tracking markers.
 
Back
Top