Support NeoGAF

[injurai]

MS spent 1/20th on not VR that FB spent on VR

 

[NullPointer]

lmao

It's like 2010 all over again.

As opposed to the 90s all over again. Take your pick ;p

I'm not specifically talking about Minority Report though (though that's a fine example too) - but anything based upon augmenting real spaces through advances in projection, holograms and haptic fields.

 

[lastflowers]

That's a strange way to look at it. MS didn't seem the least bit interested in AR until after Google unveiled Google Glass. They won't be interested in VR until after Oculus Rift or Project Morpheus become consumer products.

lol

 

[Alx]

lmao

It's like 2010 all over again.

It's the last missing piece for the "good morning batman" picture.

 

[Prince Vultan]

I'm not saying that it will be definitively like that. Hell, I don't know. All I am saying is that it could be similar for certain things. Which would be incredibly neat if it even came close.

Sorry if that came off as dickish on my part, but I can remember reading 'Just like Minority Report' after the original Kinect was announced just about everywhere I was posting at the time.

I just got a bit of a nostalgia chuckle from your post, wasn't trying to take the piss or anything.

 

[LukasTaves]

might just about be ok. Obviously not ideal due to lag, and way too slow for positional movement. Seeing your hands in front of you a bit laggy would be weird but might be ok.

Obviously I never tested it, but I have the impression that the lag perception would actually improve as you are not seeing your actual hands/arms to point out the delay on screen.

 

[Y2Kev]

To me, VR and AR are not really two technologies aiming at the same end (and that is readily apparent in the name). I think you have to stretch to compare the two, and even then it seems a bit weak.

I also could be wrong on this, but VR seems way further along. I don't think we have anything even approaching the sci-fi magic of something like Minority Report, but Occulus Rift and Morpheus seem to indicate we're not too far from very commercialized VR that creates very believable presence.

 

[DieH@rd]

I don't understand that. The kinect already handles motion tracking fine. Have you guys played kinect sports rivals preseason? The jet ski guy tracks really well. Project spark handles motion capture really good as well. Why are you saying it isn't possible when it's already doing it?

Latency of Kinect 2 is "~60 ms with processing" according to MS stats. That's OK for hand waving in front TV, but its not nearly enough for head position tracking for VR.

If they are smart, they have left the room for forcing camera to drop resolution and work at least in 60hz.

PS4 camera on the other hand seems like it was "mysteriously" built to support fast tracking:
720p @ 60hz
640x480 @ 120hz
320x240 @ 240hz

 

[BigBearJohnson]

Positional head tracking requires lower latency than the 30hz Kinect can provide.

I'm not talking about positional head tracking. The Jet ski game in KSreseason tracks your entire body, even hand. Opening and closing your hand controls the throttle, while at the same time tilting while pivoting steers, and your leg activates speed boosts and leaning forward and back executes barrel rolls on jumps in an analog fashion. It does all of that at the same time.

Not to mention the game is also two player so it's doing all thy with two separate skeletons to track. Your assuming a lot based on paper. You need to play it to see that it's more than possible.

 

[DieH@rd]

To me, VR and AR are not really two technologies aiming at the same end (and that is readily apparent in the name). I think you have to stretch to compare the two, and even then it seems a bit weak.

I also could be wrong on this, but VR seems way further along. I don't think we have anything even approaching the sci-fi magic of something like Minority Report, but Occulus Rift and Morpheus seem to indicate we're not too far from very commercialized VR that creates very believable presence.

Yes, VR is much easier.

As for Minority Report, I dont know why is everyone mentioning AR. That film used fancy holograms.

 

[Unknown Soldier]

Microsoft has that big MS Research team that has been working on many topics for years, and AR among them. I'm sure there are publications that predate the birth of Google and even of his founders.
http://research.microsoft.com/apps/...&sb=on&ps=25&t=publications&sf=&s=&r=&vr=&ra=

(the oldest go back to 1986, but I'm not quite sure they're internal work though... Or I didn't know Faugeras had worked for MS).

Well maybe I should think about the question differently. We aren't pretending that most of the research that went on at Bell Labs or something directly resulted in a product or you could claim they invented everything.

So I guess the question is, Google has produced an actual product in Google Glass which works in the real world and is available today. When did MS decide they were interested in producing a similar product and bringing it to market? Was it before or after? This is assuming they have such plans at all, but they are unlikely to be buying up these patents if they weren't.

 

[ironcreed]

Sorry if that came off as dickish on my part, but I can remember reading 'Just like Minority Report' after the original Kinect was announced just about everywhere I was posting at the time.

I just got a bit of a nostalgia chuckle from your post, wasn't trying to take the piss or anything.

Yeah, I was laughing at all of those claims back then as well. But still, Minority Report is just a good example of what awesome AR could be like and I think it would be incredibly cool to even come close to something like that.

 

[Nafai1123]

I'm not talking about positional head tracking. The Jet ski game in KSreseason tracks your entire body, even hand. Opening and closing your hand controls the throttle, while at the same time tilting while pivoting steers, and your leg activates speed boosts and leaning forward and back executes barrel rolls on jumps in an analog fashion. It does all of that at the same time.

I understand that, and I'm saying VR require positional head tracking at a lower latency than Kinect can provide. What would be their solution to this problem?

 

[jcm]

That's interesting. Let's try comparing apples to apples. So when did Google file their first AR related patent?

I don't know. But Google filed the first comprehensive patent for Glass as a product in August 2011. I don't know if they have earlier AR-related patents, and I don't know if MS does either. Probably both of have earlier stuff laying the groundwork for the actual product.

At any rate, your original statement was clearly wrong. And since no one has actually brought a product to market yet, it's premature to be awarding points for innovation anyway. Lots of companies are working on this stuff, and have been for a while. Nobody's shipped anything yet, which makes me think it's probably hard to get right.

Well maybe I should think about the question differently. We aren't pretending that most of the research that went on at Bell Labs or something directly resulted in a product or you could claim they invented everything.

So I guess the question is, Google has produced an actual product in Google Glass which works in the real world and is available today. When did MS decide they were interested in producing a similar product and bringing it to market? Was it before or after? This is assuming they have such plans at all, but they are unlikely to be buying up these patents if they weren't.

Really? Google Glass is a an actual product available today? Where do I buy it?

 

[lastflowers]

To me, VR and AR are not really two technologies aiming at the same end (and that is readily apparent in the name). I think you have to stretch to compare the two, and even then it seems a bit weak.

I also could be wrong on this, but VR seems way further along. I don't think we have anything even approaching the sci-fi magic of something like Minority Report, but Occulus Rift and Morpheus seem to indicate we're not too far from very commercialized VR that creates very believable presence.

agreed. AR doesn't seem good for gaming imo, but for enhancing our ability and access to information as we go about our daily lives.

VR seems to me to be about placing yourself in a new world to titillate different experiences.


AR enhances your daily life, VR puts you in a different world.

 

[enzo_gt]

Augmented reality has a much farther way to go in being viable as a platform outside of tech demos (or tech demoes stretched out into full games) than VR.

Yup, MS can see past VR and known AR is the better bet. Far more excited for what they plan to do than the current VR stir.

 

[BigBearJohnson]

I understand that, and I'm saying VR require positional head tracking at a lower latency than Kinect can provide. What would be their solution to this problem?

But that's it, your saying the latency is a problem based on a lack of experience with the actual setting. Latency might exist but it's hardly noticeable with kinect 2.

 

[SPDIF]

Well maybe I should think about the question differently. We aren't pretending that most of the research that went on at Bell Labs or something directly resulted in a product or you could claim they invented everything.

So I guess the question is, Google has produced an actual product in Google Glass which works in the real world and is available today. When did MS decide they were interested in producing a similar product and bringing it to market? Was it before or after? This is assuming they have such plans at all, but they are unlikely to be buying up these patents if they weren't.

Back in 2008/09 probably.

 

[pants]

I read that as "Microsoft has purchased $100-150 million worth of Augmented Reality pants"

i am a bit disappointed but it's my own fault.

I'm not that cheap

 

[watershed]

AR has been really interesting. Even at a kind of elementary way the 3ds has some pretty cool AR games/uses. When we first learned that the Wii U gamepad would have a camera I thought Nintendo was going to push even better AR stuff with their home console. Too bad that never happened.

I'd be really interested to see what one of the big 3 could do with AR games run thru a more powerful system than the 3ds.

 

[Nafai1123]

But that's it, your saying the latency is a problem based on a lack of experience with the actual setting. Latency might exist but it's hardly noticeable with kinect 2.

I'm not saying it, Oculus and Sony R&D have said it again and again. Latency perceive on a TV is not the same as latency perceived in VR. When you move your head, you expect the scenery around you to move with you. If it doesn't, you get nauseous. Both Oculus and Sony are trying to get latency below 20ms and are currently at 30-40ms. Kinect with processing is around 60ms.

 

[ricardokenshin]

But that's it, your saying the latency is a problem based on a lack of experience with the actual setting. Latency might exist but it's hardly noticeable with kinect 2.

It's clear to me you don't know why the kinect 2 camera is not viable for VR. You should read about it.

What I think would be cool, if MS decides to go with VR and ship another camera with it, is VR (with a better camera than kinect2) + kinect (with the kinect 2 camera) for some full body VR games. They might be terrible since you would only be standing around, they would have to figure out a way to add controls for movement but it would be cool.

 

[Chobel]

I'm not talking about positional head tracking. The Jet ski game in KSreseason tracks your entire body, even hand. Opening and closing your hand controls the throttle, while at the same time tilting while pivoting steers, and your leg activates speed boosts and leaning forward and back executes barrel rolls on jumps in an analog fashion. It does all of that at the same time.

Not to mention the game is also two player so it's doing all thy with two separate skeletons to track. Your assuming a lot based on paper. You need to play it to see that it's more than possible.

KSR is 30fps game so tracking with 30fps could work, however IIRC some websites mentioned that there's input lag even when it's 30fps. VR games are +60fps, so tracking with 30fps will feel weird.

 

[kodecraft]

VR is not AR.

AR is Google Glass or 3D/Vita/Smartphone demos, only MS wants to provide real world object tracking and placing of computer generated objects in our view [in addition to static Bing-powered HUD stats].

This. I think AR is the better choice over VR imo.

 

[Jomjom]

But that's it, your saying the latency is a problem based on a lack of experience with the actual setting. Latency might exist but it's hardly noticeable with kinect 2.

Hardly noticeable when you are just comparing your hand waving to your characters movements on the TV. Very noticeable when you have a VR headset showing you images that need to match up with your head movement.

OR research has already shown that that kind of latency just won't do.

 

[RamzaIsCool]

Isn't kinect in it's current form kinda useless for VR purposes? So it makes sense they try something to enhance that experience with AR, rather then to abandon Kinect and hop on the VR train.

 

[Alx]

Well maybe I should think about the question differently. We aren't pretending that most of the research that went on at Bell Labs or something directly resulted in a product or you could claim they invented everything.

So I guess the question is, Google has produced an actual product in Google Glass which works in the real world and is available today. When did MS decide they were interested in producing a similar product and bringing it to market? Was it before or after? This is assuming they have such plans at all, but they are unlikely to be buying up these patents if they weren't.

Well I suppose the question is mostly about "bringing it to the market". Just like Sony with move, I'm sure they were working on motion controls on their own, but decided to make the last step after Nintendo showed the way. Maybe even the success of Oculus Rift helped greenlight project Morpheus.
So yeah of course Google Glasses had their effect in MS strategy of doing AR. It's not an entirely reactive decision, and there's credit to give to Google for making the first step, but they're mostly the triggering event.
It's a bit like electric/hybrid cars, companies have been preparing them for decades and all had their own prototypes ready, and they were just wondering when was the right time to make the first step. Until Toyota did.

Latency of Kinect 2 is "~60 ms with processing" according to MS stats. That's OK for hand waving in front TV, but its not nearly enough for head position tracking for VR.

Head position tracking wouldn't be done by camera anyway, Oculus Rift and Morpheus use inertial sensors with a much higher data rate for that, and the camera is only there to correct drifting. Some latency is acceptable for that, and MS could have the same sensors in the glass.
The issue would be for hand and feet tracking though, where for fast motions you would see the lag between display and reality. Depending on the effect you want to produce, it could be more or less convincing.

 

[ironcreed]

I would love some streamlined glasses where menus, maps, inventories, etc. popped up right in front of you and you could manipulate them with Kinect. Then imagine also having certain effects present themselves in the room while playing. It would be immersive as hell.

Or am I off base for AR here?

 

[gamechanger87]

I think Microsoft's position from the beginning has been focused on the social aspects of gaming and its applications. AR if done correctly can provide a more social and group experience, while VR isolates the user within the gear. Everything in the initial leaked roadmap and the roll out of the XBO supports this view.

 

[LukasTaves]

I actually like Microsoft Illumiroom alot better than VR headsets.
Is it still coming out?

http://youtu.be/aoUGRGjG07E

I like that you don't have to wear anything, and adds a lot to immersion. But it's too fundamentally different to compare.

But I do think that while VR is the future, with current tech, I'd enjoy more playing with illumiroom instead of wearing a helmet.

 

[mrklaw]

Yup, MS can see past VR and known AR is the better bet. Far more excited for what they plan to do than the current VR stir.

why would you want to overlay stuff on the dirty real world, when you can escape to the imagisphere of VR?

 

[enzo_gt]

Isn't kinect in it's current form kinda useless for VR purposes? So it makes sense they try something to enhance that experience with AR, rather then to abondon Kinect and hop in the VR train.

No? Positional tracking is essential for AR. Just because the technology ATM can't sustain future developments doesn't mean it isn't worth investing in. That's why they're still pushing out Kinect models as they slowly shave the lag away between incarnations. They've opted to push it to the masses instead of cooking it for years upon years, which has worked out pretty well.

 

[Krilekk]

Yea, obviously they've been working on their Kinect plus glasses idea for a while. Time will tell what it ends up being, but really I can't think of a single situation where AR was cool.

Seriously? Remember IllumiRoom? I still think that was actually created to be able to show off AR possibilities for people not wearing the glasses. AR weather effects would be glorious.

 

[jaypah]

why would you want to overlay stuff on the dirty real world, when you can escape to the imagisphere of VR?

I actually want both which is why I've always been glad that Sony and MS weren't chasing the same rabbit. I'm going to buy both and nobody can stop me!!!

 

[ironcreed]

I actually want both which is why I've always been glad that Sony and MS weren't chasing the same rabbit. I'm going to buy both and nobody can stop me!!!

I agree. It kind of adds even more value to owning both systems. Which is what I want.

 

[DieH@rd]

Positional tracking wouldn't be done by camera anyway, Oculus Rift and Morpheus use inertial sensors with a much higher data rate for that, and the camera is only there to correct drifting. Some latency is acceptable for that.

You have no idea that you are talking, both Rift DK2 and Morpheus are using external cameras, and they will continue to use them with consumer models.

With internal sensors you can only get accurate measurements for six axis of ROTATION, which is good if your head in the game space is fixed and cant be moved anywhere. Inertial sensors are widely inaccurate. They build up drift like mad, and for VR tracking needs to be PERFECT.

 

[mrklaw]

No? Positional tracking is essential for AR. Just because the technology ATM can't sustain future developments doesn't mean it isn't worth investing in. That's why they're still pushing out Kinect models as they slowly shave the lag away between incarnations. They've opted to push it to the masses instead of cooking it for years upon years, which has worked out pretty well.

kinect is too slow. why 'slowly shave away lag' when you can use a basic webcam right now to give you 120/240Hz feedback?

I think a combination of Playstation camera and kinect would be good - you'd want the speed of the basic playstation camera for positional head tracking (that needs to be fast and accurate to avoid motion sickness), and then the more complex 3D body tracking of kinect for interaction with the world.

Still needs speeding up though..

You have no idea that you are talking, both Rift DK2 and Morpheus are using external cameras, and they will continue to use them with consumer models.

With internal sensors you can only get accurate measurements for six axis of ROTATION, which is good if your head in the game space is fixed and cant be moved anywhere. Inertial sensors are widely inaccurate. They build up drift like mad, and for VR tracking needs to be PERFECT.

could you mitigate the low framerate of the kinect by using it along with intertial sensors? so rather than using the camera as a pure positional movement sensor, you use it more as a continuous calibration. Use the intertial sensors for basic movements, and the kinect to understand when the person isn't moving, so adjusting continuously for drift?

 

[plainr_]

why would you want to overlay stuff on the dirty real world, when you can escape to the imagisphere of VR?

Right. Your AR experience will vary from room to room. With VR, you can be taking a shit in your bathroom while scaling Mt. Olympus.

 

[Giant Panda]

For gaming at least, AR is nearly useless compared to VR. That's why I was confused that the AR glasses MS were developing were specifically tied to Xbox.

 

[gamechanger87]

Right. Your AR experience will vary from room to room. With VR, you can be taking a shit while scaling Mt. Olympus.

True, just need a really really long cable..

 

[Alx]

You have no idea that you are talking, both Rift DK2 and Morpheus are using external cameras, and they will continue to use them with consumer models.

Yes, and like I said the cameras are used to correct drifting. Sony said the positional tracking was done at 1000Hz, do you really think the PS camera can go that high with a decent resolution ?

 

[Cyriades]

agreed. AR doesn't seem good for gaming imo, but for enhancing our ability and access to information as we go about our daily lives.

VR seems to me to be about placing yourself in a new world to titillate different experiences.


AR enhances your daily life, VR puts you in a different world.

It does seem MS bought this for non-game applications

 

[enzo_gt]

why would you want to overlay stuff on the dirty real world, when you can escape to the imagisphere of VR?

Because I can play multiplayer with my buddies sitting next to me and actually see them while doing so, not compromising the best type of multiplayer in favour of more parasocial experiences that the industry has been leaning towards since online multiplayer debuted on consoles. Furthermore, this opens up far more than just gaming possibilities. I see far more uses for projecting manipulable environments that people can interact with than plugging away into the Matrix. Not every game is or should be about making it as much of an escape from reality as possible, which is VRs biggest sell, but again, AR promises concessions in case you want that too, just maybe not as immersive.

 

[1.21Gigawatts]

You're misunderstanding, I'm saying 'eventually'. I know the difference. I do. But all you need is a webcam to make any VR into AR (simply speaking).

I'm not saying they are the same, I'm saying that the devices used to create either experience are very similar that eventually they will most likely coincide.

You'd need screens that can turn from normal screens to fully transparent screens.

 

[flyinpiranha]

To me, VR and AR are not really two technologies aiming at the same end (and that is readily apparent in the name). I think you have to stretch to compare the two, and even then it seems a bit weak.

I also could be wrong on this, but VR seems way further along. I don't think we have anything even approaching the sci-fi magic of something like Minority Report, but Occulus Rift and Morpheus seem to indicate we're not too far from very commercialized VR that creates very believable presence.

Well, not through our 'eye's' like holograms and movement that are projected "in a 3d world". But through a screen, or something similar. For example this app:

https://itunes.apple.com/us/app/magicplan/id427424432?mt=8

And this stuff is still in its infancy. But you strap that onto a camera with Google's Project Tango and you could have a designer step into your room, change paints, move furniture, etc ... same goes for hide and seek games or any other crazy shit the more creative out there could think of.

The main difference between AR and VR for me is that VR is mostly stationary (sans walking on a treadmill) while AR is best for movement and roaming around with an overlay.

I think the two are inherently different but the devices used to display them could be very similar. Think years from now your sunglasses would darken for VR and lighten for AR or something.

I do agree that they are going for something different though, but the tech to achieve both is so similar that we're going to see a lot of overlap in development and such.

For example take a game, one renders your house down to a tee and you can walk around it with VR goggles and a treadmill with monsters jumping out of stuff and Lego pieces waiting to be stepped on. Take that same idea BUT instead you have the same goggles, it overlays monsters and Lego pieces and you actually walk around your house with your badass Finger Glun (glove+gun - patent pending). The two are very similar with the device you are using, but completely different experiences, yet ...kind of the same.

There is obviously a difference, but I see the roads crossing over more and more.

You'd need screens that can turn from normal screens to fully transparent screens.

Of course, I'm not talking about today. And why would I need screens that turn fully transparent? Do the 3DS and Vita not do AR? They just use the camera to project on your screen. Implement a camera on a VR headset and it's basically the same.

 

[gamechanger87]

Because I can play multiplayer with my buddies sitting next to me and actually see them while doing so, not compromising the best type of multiplayer in favour of more parasocial experiences that the industry has been leaning towards since online multiplayer debuted on consoles. Furthermore, this opens up far more than just gaming possibilities. I see far more uses for projecting manipulable environments that people can interact with than plugging away into the Matrix. Not every game is or should be about making it as much of an escape from reality as possible, which is VRs biggest sell, but again, AR promises concessions in case you want that too, just maybe not as immersive.

100% agree. Both AR and VR are aiming for different experiences.

 

[Phades]

At any rate, your original statement was clearly wrong. And since no one has actually brought a product to market yet, it's premature to be awarding points for innovation anyway. Lots of companies are working on this stuff, and have been for a while. Nobody's shipped anything yet, which makes me think it's probably hard to get right.

With what AR is being developed as currently, isn't what the military been using for aircraft huds and helmets for targeting and tracking basically a "first step" into AR and been around for a long time now? Sure nothing consumer grade has been readily available, but this doesn't seem like a new idea at all. Hell even some car manufacturers were toying around with HUD style projections onto the windshield at one point. I still believe one still is in regard to night vision cameras for safety, I just can't think of which manufacturer it is.

 

[mrklaw]

Because I can play multiplayer with my buddies sitting next to me and actually see them while doing so, not compromising the best type of multiplayer in favour of more parasocial experiences that the industry has been leaning towards since online multiplayer debuted on consoles. Furthermore, this opens up far more than just gaming possibilities. I see far more uses for projecting manipulable environments that people can interact with than plugging away into the Matrix. Not every game is or should be about making it as much of an escape from reality as possible, which is VRs biggest sell, but again, AR promises concessions in case you want that too, just maybe not as immersive.

I agree with you on the couch co-op, but I fear it is pretty much dead. Ironically VR might be something that brings that back - albeit you'd be physically on different couches, but you could be virtually on the same one.

I have no idea why everything is online and not shared screen anymore. so sad.

 

[tensuke]

That's a strange way to look at it. MS didn't seem the least bit interested in AR until after Google unveiled Google Glass. They won't be interested in VR until after Oculus Rift or Project Morpheus become consumer products.

MS was definitely planning stuff like Fortaleza before Glass was announced (not that Google wasn't working on it for a while too).

 

[KoopaTheCasual]

If they've bought VR patents they don't need to do anything in the space, they can just apply a patent tax to anyone who does.

Too bad they haven't bought any VR patents.