'Pressed into your torso' sounds like it has good tracking and occlusion resistance - until you read '(about 40cm)'. That's well with the range you'd naturally hold your hands in front of your body.
-
Hey, guest user. Hope you're enjoying NeoGAF! Have you considered registering for an account? Come join us and add your take to the daily discourse.
The High-end VR Discussion Thread (HTC Vive, Oculus Rift, Playstation VR)
- Thread starter Cartman86
- Start date
'Pressed into your torso' sounds like it has good tracking and occlusion resistance - until you read '(about 40cm)'. That's well with the range you'd naturally hold your hands in front of your body.
GuitarAtomik
Member
Forgot about that game. Looks cool.
ugh, another exclusive.
StudioTan
Hold on, friend! I'd love to share with you some swell news about the Windows 8 Metro UI! Wait, where are you going?
It's being made by Oculus Studios.
Vyse The Legend
Member
Revive will save us.
StudioTan
Hold on, friend! I'd love to share with you some swell news about the Windows 8 Metro UI! Wait, where are you going?
Not sure it will with Touch games. How are they going to account for the difference in controllers?
Wrestlemania
Banned
Haven't they already got that working?
I'm not sure how to translate the resolution on these devices into practical terms. They seem to all be in the ballpark of 1080x1200 for each eye...I'm mainly interested in trying these out as a virtual workspace. But the resolution seems low for that, not to mention oddly shaped.
Can anybody give a rough figure of how much "screen real estate" you get with these, in relation to multi-monitor setups? I currently work on six 1920x1200 monitors, and I'm not expecting to get anywhere near that much, but I may be wiling to accept less if the experience is great.
Can anybody give a rough figure of how much "screen real estate" you get with these, in relation to multi-monitor setups? I currently work on six 1920x1200 monitors, and I'm not expecting to get anywhere near that much, but I may be wiling to accept less if the experience is great.
Even then, it's still bullshit. Is The Lab a Vive exclusive? Nope.(at least it wont be when Touch comes out)
Compared to Vive? Really?
Wrestlemania
Banned
Have Valve confirmed that?
StudioTan
Hold on, friend! I'd love to share with you some swell news about the Windows 8 Metro UI! Wait, where are you going?
No idea, that's why I was asking.
Yes. Valve uses OpenVR which has support for the Rift baked in. It also has basic support for Razer Hydras - Rift users have been able to play some games with them.
Valve is able to take this approach because their business is Steam. They don't really care what headset you're using as long as you're still buying your games from their store.
Wrestlemania
Banned
The reflection bit is definitely true. Not sure if it will lose tracking less while in range, but it's certainly happened to me a number of time with Vive.
Unreal Engine 4.12 Preview 1 is out, for those that want to try the VR Editor without compiling anything.
https://forums.unrealengine.com/showthread.php?108495
https://forums.unrealengine.com/showthread.php?108495
Valve has said none of the titles available on the Vive are exclusive. I dont believe they'd say that and have it not include their own games....Otherwise they'd be straight up lying.
Zalusithix
Member
There's measurements out there for the headsets' FOVs. That tells you objectively how much your view is going to be in far less ambiguous terms than how many monitors. (The monitors size, bezel, orientation, and distance all factor into how much of a user's FoV they take up.) Comparing it to a traditional monitor stack is pointless though. It's not the same as it moves with your head, and the pixel density is not uniform. That said, it's nowhere close to a traditional monitor's pixel density. You don't buy a VR headset for the visual fidelity. You buy it for the experiences you can't get on traditional monitors at all.
True that Oculus exclusives are ok, but not a deciding factor for me.
TBH the launch fumbling from Oculus along with their history is starting to identify what I believe are core problems with the company itself. Maybe it will all work out over time, but for now my confidence in them is waning. I will have to try the Vive when it get's here and decide for myself.
Your experience with both seated definitely sheds light on what others are saying and whether some of this stuff is being blown out of proportion. How would you compare the convenience of integrated headphones on the Rift compared to having a separate set on the Vive?
This is where I may be as well. Convenience is definitely a factor and I want to be able to come home, relax and throw the headset on for a bit. I do not want to be fumbling with cords, making sure things are charged/synced or worrying about whether my PC monitors are within arms reach where I can accidentally destroy one.
You can play The Lab with a rift and hydras right now if you want
I don't comment on the other things, but this is a strange thing to use. You can also put "tracking less susceptible to not working when you turn off I've if the lighthouses. Or when you let the protective film on them." You have clear instructions about how to use the device, you should not use against it a situation that it's clearly stated to create problems and being not recommended.
"Negative: the gamepad doesn't work when the batteries are dead. The keyboard is better"
Wrestlemania
Banned
It's entirely fair to point out that mirrors, an entirely normal thing you may have in your room, will have an adverse effect on one of the tracking systems but not the other.
Making space is one thing because there's no way around that whatever the technology is, but if you then have to think about taking stuff off the wall because it might interfere with the tracking then that's definite going to be a negative for some people when the alternative doesn't have that issue.
How is rift impervious to the effects of mirrors? If your body is obscuring direct line of sight to a controller, yet it is visible in a mirror, what would prevent the software from thinking that reflection is the real thing?
Edit: Even simpler, if the HMD is visible to the camera both directly and in a mirror reflection, how does the oculus software know which one is real?
Wrestlemania
Banned
I mean, I guess theoretically there's some scenario where having a mirror in view of the camera and the camera seeing both you and the mirror could lead to a tracking issue it's unable to determine which is the actual thing it should be tracking.
There's no specific guidance saying mirrors will be a problem for RIft or any reports of it being an issue though, so it's a hypothetical issue versus an actual one that's both noted by the manufacturer and reported as being a real issue by users who've experienced it.
I mean, there are Vive vs. Rift comparisons that are saying the Vive is a bit more picky about reflective surfaces. There are many pros to the Lighthouse concept, but this is one of the cons. It won't even affect everyone, and it's something you can totally work around, so I don't think there's much reason to rebel against the idea that this minor issue might exist.
Melchettsmoustache
Member
I've got an electrician pal of mine coming round when I get my Vive (fingers crossed next week) to get my VR room setup. He suggesting a "Pulley" wheel over head as a solution to make sure the cable is always out way, even if it isn't proving to be a difficulty with tripping over it. The light houses will be wall mounted and screwed in place so I'm hoping once I've got through my initial setup it will be about as plug and play as possible.
I've got Rift on order as well however I'm really toying with binning it, particularly if the mods for allowing Oculus games to run on Vive are working as well as I've seen reported.
One thing we know is that Carmack called the whole Touch project something which needed urgent attention from many engineers. Let me get the exact quote.
I think there's certainly a need to "rebel" against the idea that Rift tracking is in any way preferable.
I set up both systems exactly as you are instructed to in their respective documentation, and I've had more tracking issues with the Rift. And that despite the fact that the Vive is tasked with a far more complex tracking challenge (A HMD + controllers in an entire room in action-packed games vs. just some 360° standing laid-back experiences).
It's also less of a problem since currently most people have their rift setup at a desk, and use it in fairly close range with their body taking up a pretty big amount of the camera FOV. It seems unlikely that a mirror would be in a position to cause problems with seated/standing VR. Not to mention the touch controllers aren't out yet, and they are much smaller, will be moving more/faster, and are easier to occlude.
For someone who has issues with mirrors and vive, I don't think it would solve the problem to replace the lighthouses with rift cameras in the same spots. If at some given physical position a photon from the lighthouse laser is bouncing off a mirror and hitting a photodiode on the vive, then at the same position with rift, a photon from an LED would also bounce off the mirror and hit the rift camera in the reverse direction. Line of sight is line of sight.
They may have a better software solution for filtering out that kind of thing but fundamentally it is no better than vive at ignoring mirrors.
Wrestlemania
Banned
I would expect that the filtering could be quite a bit easier with the Rift since it's a much smaller number of IR emitters that it's tracking. If it can differentiate between the pattern of IR emitters on the headset/controllers and the reflected pattern then it shouldn't be an issue.
I'm not sure. Unless maybe the pattern of the LEDs being reversed in the mirror is checked in software?
Windows can be a problem with the Rift. I think his name is Jeremy, on the tested podcast, was losing tracking by 6 feet away from the camera due to them.
Room issues can be a problem with either setup.
Right, the exact wording was that computer vision people were "panic piled" onto the Touch project. (around 4:40)
Herr Starr
Member
And Tested, who in my experience are very thorough, directly compared the two and said that the Vive has more tracking issues, which they rarely see with the Rift.
Hooray for anecdotal evidence, I guess?
Wrestlemania
Banned
That's interesting, and no doubt there's something major they're working on that resulted in the delay, but doesn't back up that it has issues with tracking fast movements. It's been demoed and I've never seen anyone complain about that, even at it's first showing at E3 last year.
The number of emitters is much higher with rift. Vive has 4, two in each lighthouse. Rift's headset has 40. I can't find the number of sensors on the vive but it's somewhere around 20-30 I believe.
Also, the IR footage I saw of a CV1 seemed to show that all of the LEDs on the headset blink in unison, rather than individually as with DK2. I haven't seen anything concrete on CV1's LEDs, so it's possible that the LEDs blink individually with very subtly different patterns. If it works as it appears, where each device has 1 pattern for all of its LEDs, then it would not be able to rely on pattern for mirror filtering.
True, if anything it raises concerns about tracking range, not speed of movement. (Which also makes more sense if you consider how these systems actually work)
To the naked eye, the rift's emitters are symmetrical. If they are not, then yeah that would be a good way to block mirror interference.
Yeah I've wondered why it doesn't already have that.
When someone gets their hands on touch, all they have to do to answer the question is replicate this gif:
http://webmshare.com/6BRqB
I don't doubt Touch will be able to do it.
Wrestlemania
Banned
Oh, sorry, I see what you mean.
To me, the pattern looks symmetrical down the middle of the headset. If the left/right sides are the same, how will it be able to distinguish regular from mirrored?
Herr Starr
Member
Perhaps by using similar techniques to those used in multiplayer games to predict where a player would move if he suddenly suffers from a lag spike? In other words, use prediction to guess where the headset would move next based on previous frames, then discard values that are way outside of this range. Of course, this doesn't work if the mirror image is in view from frame 1.
This is an interesting topic with lots of potential answers. I hope we'll get more information about the technologies in use over time.
Wrestlemania
Banned
Oh, sorry, I see what you mean.
To me, the pattern looks symmetrical down the middle of the headset. If the left/right sides are the same, how will it be able to distinguish regular from mirrored?
Maybe comparing the data from the gyro/accelerometer with the patterns to determine the orientation of the headset to the cameras and then inferring from that which is the mirror image?
Something like that. Seems like it would be a bit easier than the issue with the Vive when the IR bouncing off the mirrors causes false positive reading. They've said they're working on that filtering though, so maybe they can improve it.
What would your opinions being about getting either a single Pascal card or going for an SLI Pascal setup for future SLI support? I'm willing to spend on the SLI setup if support for dual GPU's is going to happen sometime within the year, but if it's far enough away that the successor to Pascal would show up before any support surfaces I won't bother.
I'm almost certain I read that Rift LEDs do blink a self-identifying pattern, but I can't find a reference after a quick search.
Oculus hasn't announced any chaperone equivalent, and the Fantastic Contraption developer has said he's basically developing his own version of chaperone from scratch because of that.
Who knows, Oculus may surprise us, but I wouldn't expect them to have a chaperone system anytime in the near future. They are very much against the idea that Vive style room scale is what people want this generation of VR. I think it's a safe bet most, if not all of the cool room scale stuff for at least the next year will be found on Steam.
Oculus hasn't announced any chaperone equivalent, and the Fantastic Contraption developer has said he's basically developing his own version of chaperone from scratch because of that.
Who knows, Oculus may surprise us, but I wouldn't expect them to have a chaperone system anytime in the near future. They are very much against the idea that Vive style room scale is what people want this generation of VR. I think it's a safe bet most, if not all of the cool room scale stuff for at least the next year will be found on Steam.
Since there's been absolutely no hints at SLI support from either player, my personal prediction is that we won't see SLI support for VR until gen2.
I assume both rift and vive are doing stuff that checks "is this position reasonable?". If one sensor on the vive thinks it's 6 feet away from the rest of them because of bounced mirror light, then it is obvious you can throw out that position and wait for a more sensible reading.
Like I said before, it's possible that rift's software filtering is better at this point. It's also possible that few people if any are experiencing mirror problems with rift since the tracked volume is so small and the range of reasonable positions is easier to know. There are also fewer objects being tracked.
IR light coming from an emitter and reaching a sensor is the foundation of both systems. I don't know why you think the vive's would be more susceptible to mirrors.
Maybe someone in this thread with a CV1 could hold their rift up to a mirror and see what happens.
They blink a pattern for sure, I've seen video of it. IR footage of the DK2 shows the LEDs blinking separately, and the effect looks very similar to twinkling stars. I also saw CV1 IR footage but with that, it looked like all the HMD's blinking the same pattern. I think that may have been a decision based on how many objects they will want to track in the future. The headset has 40, each touch controller probably has 20 each. That alone is 80 LEDs that need to be identified. With one pattern per device you only need 3 distinct patterns, but with 1 pattern per LED you'd need 80 different patterns.
Wrestlemania
Banned
They could do chaperone. Mark the bounds using Touch, have them show up in game in a similar way. Of course they might not actually support games that require that migh movement, so they might not even have a chaperone system. Valve have said they'll enable it through SteamVR though.
Oh, sorry, I see what you mean.
To me, the pattern looks symmetrical down the middle of the headset. If the left/right sides are the same, how will it be able to distinguish regular from mirrored?
The pattern only looks symmetrical to the naked eye. When the camera sees it, each LED is individually labeled, like this:
That's a DK2, but by all indications CV1 uses the same technique. Details on how this works: http://doc-ok.org/?p=1095
There's no possible confusion between a regular and mirror image.
Wrestlemania
Banned
I think stacking the controllers is an even more interesting test, since it really challenges the absolute positioning capabilities, which is the crucial point. And then carry them around an entire room for bonus points. Even the Vive isn't perfect at that, though it comes pretty close. (And if you've followed attempts at absolute positioning in consumer devices pre-Lighthouse, the fact that it does come close in that test is still staggering)
Of course, the Touch controllers might not stack that well
