Oculus "Rift" : An open-source HMD for Kickstarter

[Libertine]

The reason i was suggesting an interchangeable set of lenses is that i thought any content that wasn't pre-distorted wouldn't work with the Rift. Is that not true? Someone mentioned media players. Perhaps they distort the content for the Rift? Another reason was to double the resolution for 2D content, especially since movies and perhaps even Imax movies would not use all the available pixels.

@Msat: If you incorporated a simple rectangular section in front of the lens with tracks on either side to guide it into place, you could ship the rift with the default lenses assembly and make the other one available for what? $10.00? Imax and movie theater right in your pocket.

Hey Libertine, I haven't used a Fresnel before, but I'm wondering, would it help with eye strain from staring at a monitor all day? I am considering getting one for work.

Im honestly not sure. I don't get eye strain though. It does focus your eyes further out. I'd google that one.

[vtms]

After seeing the Rift I bet Sony improves their fov.

Or maybe Oculus could make a separate 2D IMAX HMD that would be light years better than HMZ-T2 and much cheaper. There's an opportunity there.

[PasticheDonkey]

That would not work, since both eyes needs to see the same.
Both halves would look blurry/ghosted on top of eachother

you could however sample that half image with a half pixel offset for each eye. this would give you better percieved res equal to the normal displays full res when not split in two.

so you'd need some kind of lens prism per eye to widen and shrink the image. anyone know something that would do that?

[MSat]

Imax and movie theater right in your pocket.

Hehe. Hardly fit in your pocket.

I think if you want to watch movies, then Namielus' project is the most ideal way to go about it. Otherwise, a huge FOV and no headtracking would lead to an uncomfortable viewing experience.

Remember, the current device is an early dev-kit, so resolution will go up once the consumer product is ready. But even with 640x800 per eye, it's not far off from DVD's native 720x480, so by turning your head just a little to the left and right gets you full resolution easily.

[Libertine]

Imax and movie theater right in your pocket.

Hehe. Hardly fit in your pocket.

I think if you want to watch movies, then Namielus' project is the most ideal way to go about it. Otherwise, a huge FOV and no headtracking would lead to an uncomfortable viewing experience.

Remember, the current device is an early dev-kit, so resolution will go up once the consumer product is ready. But even with 640x800 per eye, it's not far off from DVD's native 720x480, so by turning your head just a little to the left and right gets you full resolution easily.

Yeah I should have said backpack. But an Imax screen "on your person" is still a pretty significant achievement. I also think headtracking is just about necessary for high fields of view as well, but i don't see any reason it wouldn't work ok with a 2D lens.

I can't wrap my head around this, using the Rift's current lenses, would a movie end up with blurry or softer looking sections around the center if its was pre-distorted for the Rift?

[PasticheDonkey]

the rift is higher res at the centre of your view and blurrier to the edges.

[vtms]

It would really be nice to have 7680x2160p 7-inch panel. That'd cover both 2D IMAX and 3D viewing at 4K per eye. It might be possible with IGZO tech.

[Bretspot]

I just found this video that shows the leap motion's FOV and interactive area.

Holy Smokes! if that was strapped onto the front of an Oculus Rift, you'd have your full hand/arm tracking, maybe even your legs/body/feet if you look down that way. And hell I bet it could even do a great job of doing full 3DOF motion tracking of your head.

[youtube-hd]http://www.youtube.com/watch?v=pn-Phl1BN4o[/youtube-hd]

[German]

Holy Smokes! if that was strapped onto the front of an Oculus Rift, you'd have your full hand/arm tracking, maybe even your legs/body/feet if you look down that way. And hell I bet it could even do a great job of doing full 3DOF motion tracking of your head.

They've said on their forums that it doesn't return point cloud data so it's not like a Kinect where you could interpolate all that. The sweet spot for good data is probably a rectangular solid only a couple inches thick and all the way up/out above the sensor.

[Mystify]

All the stats I saw implied a 2ft cube(8 cubic feet) area above the device.

[Namielus]

https://www.youtube.com/watch?list=PL94 ... page#t=42s

watch from about 0:40
Look at when he brings a pencil into the picture, maybe its just a simulated point cloud but it seems legit to me.


I think its more like the SDK at this point do not return point cloud data but will, and remember people are still under NDA at this point.

I would not be surprised if its announced that some upgrades to the SDK will let you get access to more useable data.

[Naru]

The impressive thing about the leap is not the device itself, but the software that comes with it. It can detect a pointcloud, but what use is it if you can't tell what corresponds to your hand, arms, legs, and fingers. Getting usable data from a pointcloud is a lot of hard work. This is why they focused on only the hands and some other small items. For the kinect, Microsoft(or that other company they bought it from) spent their time creating the software to correctly detect a human body and it's pose from all the point data. I'm thinking for the leap, they don't want to add body tracking as that is not the use they intend for it. I'm waiting on Microsoft to release a more competent version of the Kinect, which I know they've had for years now.

[2EyeGuy]

Would it help if internally the 2D screen is rendered at double the resolution, and then the left half of the screen shows only the odd lines and columns, while the right half shows all even lines and columns? Would the eye "combine the detail" of both images?

I can't imagine that working well.

That should work. People keep saying you can recognise half a pixel of separation, which is what it would look like.

Of course you'd still have the aspect ratio problem, since the movie would still be squashed into half the width. And it isn't really compatible with the warping.

[blasphemer]

The Leap interaction volume is for tracking fingers, isnt it? To do whole body tracking we need less precision, so it should work over much greater distances, maybe even whole room..

[Mystify]

Would it help if internally the 2D screen is rendered at double the resolution, and then the left half of the screen shows only the odd lines and columns, while the right half shows all even lines and columns? Would the eye "combine the detail" of both images?

I can't imagine that working well.

That should work. People keep saying you can recognise half a pixel of separation, which is what it would look like.

Of course you'd still have the aspect ratio problem, since the movie would still be squashed into half the width. And it isn't really compatible with the warping.

But its not really a half pixel of separation. Its putting the picture through a paper shredder, tracking every other row, and gluing it back into 2 pictures, showing them to both eyes, and hope that somehow your brain figures out the extra detail, even though this type of vision task is completely unlike anything else your vision has to do. At best, I think it would make the image look blurry.

[MSat]

Just as a quick experiment, I placed a divider between by eyes, and looked at objects with both no overlap and partial overlap in the center. While it certainly worked as I was able to read text just fine, it was a bit uncomfortable. That leaves me wondering if it might be at all practical to utilize more of the horizontal resolution for 2D use. What exactly is the cause of the eye strain I experienced, and is it possible to reduce it? Is it a matter of slicing the image in half, and shifting each one outwards, and perhaps adding overlap in the central region?

[LikeMike]

Hello there, newbie here. I am really excited about the Rift - but I still have a couple of questions, perhaps some of the people here can answer:

1. As I understand it and see it in the videos, the screen is basically divided in two halfs showing basically the same thing - doesn`t that mean, that we have an about half as wide field of view than we would have on a monitor? And basically a larger vertical field of view than a horizontal? To me that seems kinda odd...

2. I am not a programmer by any means - but I can code a bit and enjoy level editors, so I would be intrigues to try some things. How easy would it be to get UDK or Unity creations into the rift? Is it basically just checking a box ("add support for Rift") - or do I have to really code it in there somehow?

3. For people that have already tried it - how immersive is it? Do you really get the feeling that for example this knight is physically standing right in front of you - or is it like great 3D really close to the screen?

3. Positional tracking seems to be the biggest thing missing from the developer kit - is that something that could be added later with a software update, or is the necessary hardware missing from the devkit?

4. How easy will it be to get and load user created content? I can only imagine what you guys will produce - and I can`t wait to wander on famous movie sets, see historical events, walk the moon or dive the great barrier reef - but like I said, I am not a programmer, so would it be easy to get and use those?

5. Any idea when the consumer version comes out? I know there is no release date yet, but is there a timetable of some sort? Like this year, early next year, christmas next year... something like that. I am really tempted to get the devkit. But it wouldn`t be here till May probably or even later - and if the consumer version comes out only a few months later I guess I´d rather wait.

[PasticheDonkey]

But its not really a half pixel of separation. Its putting the picture through a paper shredder, tracking every other row, and gluing it back into 2 pictures, showing them to both eyes, and hope that somehow your brain figures out the extra detail, even though this type of vision task is completely unlike anything else your vision has to do. At best, I think it would make the image look blurry.

passive 3d tvs do this and it works, in that case tho the images your brain puts together basically have scan lines that make the picture not as good. rift wouldn't have this problem. test it by creating a half side by side image using one source (one an offset sampling of the other), and use parallel or cross-eyed 3d techniques, it should work.

oh i see MSat did test. discomfort might be from them not being aligned with your ipd.

[Mystify]

Hello there, newbie here. I am really excited about the Rift - but I still have a couple of questions, perhaps some of the people here can answer:

1. As I understand it and see it in the videos, the screen is basically divided in two halfs showing basically the same thing - doesn`t that mean, that we have an about half as wide field of view than we would have on a monitor? And basically a larger vertical field of view than a horizontal? To me that seems kinda odd...

2. I am not a programmer by any means - but I can code a bit and enjoy level editors, so I would be intrigues to try some things. How easy would it be to get UDK or Unity creations into the rift? Is it basically just checking a box ("add support for Rift") - or do I have to really code it in there somehow?

3. For people that have already tried it - how immersive is it? Do you really get the feeling that for example this knight is physically standing right in front of you - or is it like great 3D really close to the screen?

3. Positional tracking seems to be the biggest thing missing from the developer kit - is that something that could be added later with a software update, or is the necessary hardware missing from the devkit?

4. How easy will it be to get and load user created content? I can only imagine what you guys will produce - and I can`t wait to wander on famous movie sets, see historical events, walk the moon or dive the great barrier reef - but like I said, I am not a programmer, so would it be easy to get and use those?

5. Any idea when the consumer version comes out? I know there is no release date yet, but is there a timetable of some sort? Like this year, early next year, christmas next year... something like that. I am really tempted to get the devkit. But it wouldn`t be here till May probably or even later - and if the consumer version comes out only a few months later I guess I´d rather wait.

1. Its more complex than that.The screen isn't just sitting there in front of you, its warped around your vision with lenses, and there is only partial overlap, so even the effective resolution isn't halved. It also encompasses a much wider field of view than a monitor because of how close it is.

2. UDK and Unity support is going to be included with the dev-kit. I don't know precisely what it will entail to get it working, but I think the hard work has been done.

3a. Haven't tried it

3b. Positional tracking would require new hardware.

4. It should be as easy as getting any other game and playing it.

5. There is no telling right now, as that would depend on the feedback they get from developers. I'd recommend waiting for the consumer version if you aren't a developer, there won't be a lot to do with it at first, and the dev kit it will be inferior to the consumer version.

But its not really a half pixel of separation. Its putting the picture through a paper shredder, tracking every other row, and gluing it back into 2 pictures, showing them to both eyes, and hope that somehow your brain figures out the extra detail, even though this type of vision task is completely unlike anything else your vision has to do. At best, I think it would make the image look blurry.

passive 3d tvs do this and it works, in that case tho the images your brain puts together basically have scan lines that make the picture not as good. rift wouldn't have this problem. test it by creating a half side by side image using one source (one an offset sampling of the other), and use parallel or cross-eyed 3d techniques, it should work.

oh i see MSat did test. discomfort might be from them not being aligned with your ipd.

in a passive 3d tv, they are using it to show you two different images, which your brain combines into 3d. you are trying to show the same image, and get more resolution out of it. I don't think that makes any sense.

[PasticheDonkey]

Hello there, newbie here. I am really excited about the Rift - but I still have a couple of questions, perhaps some of the people here can answer:

1. As I understand it and see it in the videos, the screen is basically divided in two halfs showing basically the same thing - doesn`t that mean, that we have an about half as wide field of view than we would have on a monitor? And basically a larger vertical field of view than a horizontal? To me that seems kinda odd...

your brain puts the 2 images together into something wider.

2. I am not a programmer by any means - but I can code a bit and enjoy level editors, so I would be intrigues to try some things. How easy would it be to get UDK or Unity creations into the rift? Is it basically just checking a box ("add support for Rift") - or do I have to really code it in there somehow?

the point of the SDK is to makes adding rift support to things running on those engines possible. don't know how easy it'll be tho

3. For people that have already tried it - how immersive is it? Do you really get the feeling that for example this knight is physically standing right in front of you - or is it like great 3D really close to the screen?

haven't tried it but many hands on impressions say this is the case.

3. Positional tracking seems to be the biggest thing missing from the developer kit - is that something that could be added later with a software update, or is the necessary hardware missing from the devkit?

it's been mentioned that there's a possibility that a positional tracker could be ordered and used to update the kit when ones ready.

4. How easy will it be to get and load user created content? I can only imagine what you guys will produce - and I can`t wait to wander on famous movie sets, see historical events, walk the moon or dive the great barrier reef - but like I said, I am not a programmer, so would it be easy to get and use those?

only as difficult as any other user created content.

5. Any idea when the consumer version comes out? I know there is no release date yet, but is there a timetable of some sort? Like this year, early next year, christmas next year... something like that. I am really tempted to get the devkit. But it wouldn`t be here till May probably or even later - and if the consumer version comes out only a few months later I guess I´d rather wait.

there is no information about this.

[PasticheDonkey]

in a passive 3d tv, they are using it to show you two different images, which your brain combines into 3d. you are trying to show the same image, and get more resolution out of it. I don't think that makes any sense.

if the images are sub pixel offset versions of the same image then they are not the same image and your brain treats them like it would something completely flat in 3D. like for example the screen you are looking at now.

[Mystify]

in a passive 3d tv, they are using it to show you two different images, which your brain combines into 3d. you are trying to show the same image, and get more resolution out of it. I don't think that makes any sense.

if the images are sub pixel offset versions of the same image then they are not the same image and your brain treats them like it would something completely flat in 3D. like for example the screen you are looking at now.

But I don't think that will increase your perceived resolution.

[donkaradiablo]

You know what would be a killer app for the Rift? A Superman game. Aim by looking, shoot laser from your eyes, and fly. With "Man of Steel" coming to the silver screen next year, the timing couldn't be better.

[PasticheDonkey]

But I don't think that will increase your perceived resolution.

not over the original 2d version but it's better than both eyes looking at the same half horizontal res image.

[Mystify]

I can't do anything with those images. Could someone else try and tell me how it works?

[PasticheDonkey]

I can't do anything with those images. Could someone else try and tell me how it works?

you can't do parallel viewing? that was how i looked at 3D stuff before 3D tvs. anyway my opinion is that the combined view is better than either of the images individually. but someone else's opinion would be more impartial. but i did want you to see it for yourself.

[Mystify]

I can't do anything with those images. Could someone else try and tell me how it works?

you can't do parallel viewing? that was how i looked at 3D stuff before 3D tvs. anyway my opinion is that the combined view is better than either of the images individually. but someone else's opinion would be more impartial. but i did want you to see it for yourself.

I can barely look at 3D stuff when there is equipment to do it.

[MSat]

But I don't think that will increase your perceived resolution.

not over the original 2d version but it's better than both eyes looking at the same half horizontal res image.

I can't tell much of a difference when they're merged. Perhaps try it with something like text instead?

[LikeMike]

Thanks everybody for the replies... that was quick. A quick follow up here:

With the standard resolutions standards - is it safe to say that the vertical fov is always larger than the horizontal (since the horizontal needs to be "cut in half" for each eye) in practically all hmd? Isn`t the extra vertical fov "wasted" - meaning that you can`t see it anyways since the natural fov seems to be for a lack of a better word widescreen? Couldn`t you save processing power by just deactivating some pixels at the very top or the very bottom, that you would never see anyways?

[EdZ]

is it safe to say that the vertical fov is always larger than the horizontal (since the horizontal needs to be "cut in half" for each eye) in practically all hmd?

Not really. It's true in the Rift due to the way it divides it's single panel, but most HMDs either use independent displays for each eye, or optics to view a single display from the same viewpoint for both eyes.

Isn`t the extra vertical fov "wasted" - meaning that you can`t see it anyways since the natural fov seems to be for a lack of a better word widescreen?

No, the current vertical FoV does not quite cover (though it's very close) the entire vertical FoV of the human eye. Disabling any of it would start to introduce black borders at the edge of your vision. This may change if the FoV is expanded while maintaining the same aspect ratio.

[MSat]

Thanks everybody for the replies... that was quick. A quick follow up here:

With the standard resolutions standards - is it safe to say that the vertical fov is always larger than the horizontal (since the horizontal needs to be "cut in half" for each eye) in practically all hmd? Isn`t the extra vertical fov "wasted" - meaning that you can`t see it anyways since the natural fov seems to be for a lack of a better word widescreen? Couldn`t you save processing power by just deactivating some pixels at the very top or the very bottom, that you would never see anyways?

The pixels along the top are not wasted. The vertical FOV is larger than the horizontal. With your eyes looking straight forward, your vision has >180 degrees horizontal FOV, whereas the Rift has ~90. The only wasted pixel space is in the far corners of each half.

[Mystify]

Thanks everybody for the replies... that was quick. A quick follow up here:

With the standard resolutions standards - is it safe to say that the vertical fov is always larger than the horizontal (since the horizontal needs to be "cut in half" for each eye) in practically all hmd? Isn`t the extra vertical fov "wasted" - meaning that you can`t see it anyways since the natural fov seems to be for a lack of a better word widescreen? Couldn`t you save processing power by just deactivating some pixels at the very top or the very bottom, that you would never see anyways?

Only if the vertical FOV was greater than the vertical FOV for a human, which I'm pretty sure its not.

[PasticheDonkey]

I can't tell much of a difference when they're merged. Perhaps try it with something like text instead?

well try this. the scaling isn't so good in Paint it seems, so you get weird colours interfering.

[LikeMike]

That confuses me a little... people that tried it usually say something like: you can`t see any borders - or, only if you try really hard can you see a little black border. So it seems like almost all the eye sees is the screen. I know, that a persons fov is usually a lot bigger than 90, I though the fisheye lenses and stuff like that just kinda "expanded" that to make it seem, like there are no borders (I don`t know much about the technical side, so I may be dead wrong here). But if thats the case, it should "expand" the horizontal view just as much as the vertical view, shouldn`t it? So if the expanded 90 for horizontal is "almost" enough to give you a borderless view, the expanded 110 for the vertical view should be "too much" - since your vertical view isn`t as wide as the horziontal view anyways.

Did I understand anything wrong, or are there more borders to be seen than I thought?

[cybereality]

The vertical FOV on the Rift basically covers all you can see. I mean, if you look for the edge, surely you will see an edge, but if you are just looking forward playing a game you probably won't notice it. On the horizontal edges you can see the border, but its further out, not at all like the toilet paper tubes you got with older HMDs.

So the FOV certainly has some room to improve, but it is absolutely massive compared to previous efforts. I don't think anyone can honestly complain, especially not for $300 bucks.

[PasticheDonkey]

That confuses me a little... people that tried it usually say something like: you can`t see any borders - or, only if you try really hard can you see a little black border. So it seems like almost all the eye sees is the screen. I know, that a persons fov is usually a lot bigger than 90, I though the fisheye lenses and stuff like that just kinda "expanded" that to make it seem, like there are no borders (I don`t know much about the technical side, so I may be dead wrong here). But if thats the case, it should "expand" the horizontal view just as much as the vertical view, shouldn`t it? So if the expanded 90 for horizontal is "almost" enough to give you a borderless view, the expanded 110 for the vertical view should be "too much" - since your vertical view isn`t as wide as the horziontal view anyways.

Did I understand anything wrong, or are there more borders to be seen than I thought?

put you arms at 90 degrees to each other than look in between them.

[MSat]

That confuses me a little... people that tried it usually say something like: you can`t see any borders - or, only if you try really hard can you see a little black border. So it seems like almost all the eye sees is the screen. I know, that a persons fov is usually a lot bigger than 90, I though the fisheye lenses and stuff like that just kinda "expanded" that to make it seem, like there are no borders (I don`t know much about the technical side, so I may be dead wrong here). But if thats the case, it should "expand" the horizontal view just as much as the vertical view, shouldn`t it? So if the expanded 90 for horizontal is "almost" enough to give you a borderless view, the expanded 110 for the vertical view should be "too much" - since your vertical view isn`t as wide as the horziontal view anyways.

Did I understand anything wrong, or are there more borders to be seen than I thought?

A good example is to place your hands on each side of your head. Move them forwards and backwards. By moving them forwards, you decrease the horizontal FOV, and moving them back increases it. Yes, you can always see them, and they slightly obstruct your FOV, but it's not all that different from wearing ski goggles. Unless you're wearing an HMD with a real crappy FOV, it's not always apparent how much FOV you're missing compared to pure, unobstructed vision.

[Alkapwn]

You guys have to remember that you're also viewing it through the lenses which are in front of your eyes. so you technically can't even see your full field of view on the screen even if it were possible.

[Kazioo]

I made a quick mockup comparing Rift to Sony HMZ-T1 and static eyes



white = binocular vision (stereo 3D, eyes overlap),
hatchet = monocular vision.

5.6" Oculus Rift had 90-degree horizontal FOV and 105-degree vertical FOV.

[PasticheDonkey]

You guys have to remember that you're also viewing it through the lenses which are in front of your eyes. so you technically can't even see your full field of view on the screen even if it were possible.

i believe the side of the lens facing you is concave meaning it can surround your eye.

edit: naw checked this isn't true.