Lockheed Martin 180 DEGREE FOV HMD - SIZE MATTERS?

[cleverusername]

viewtopic.php?f=26&t=4072&p=120308#p120308

I don't know guys, 150 FOV, 180 FOV, 100 gram AR, sheesh! You might think we got GANDALF himself on dis project, direct descendent of Ivan Sutherland and Alan Kay, F35 CIO ARCH WIZARD about to start slaying some dragons left and right!

HOLY SHIT! My jaw is hanging agape! Aint no damn FALSE PROPHET HERE!




We got DAVID SMITH in DA HAUS!







Take us out Elvis, David Smith has changed the planet!

All of facebook's 2 billion in plans wiped out by one Gandalf and a hobbit!?




Is my Uncle Butch and Congressman Joe Wilson on the armed services committee gonna support a 110 FOV device for our military and veterans, or this LOCKHEED MARTIN 150 FOV device? Hmmm.... Will my buddy Jim Kennedy, former director of NASA KSC, want to do underwater astronaut training and NAVY SEAL stuff with 110 FOV or 150 and higher LOCKHEED MARTIN tech? Sheesh, decisions decisions....

[willste]

The extra FOV even if it ends up being just horizontal is going to be awesome. It will be interesting to see when they demo a new creation with the new Samsung screen tech.

[KBK]

A lens for a single curve screen is ridiculously difficult to make, and introduces nasty problems in application and implementation, especially for set up for multi-varied IPD's and potential eye issues.

If at all possible it would be best to go to DUAL curve and a meniscus lens set, which would eliminate just about all potential problems and be a FINAL design consideration that probably would not change until direct neural interface replaces it.

[willste]

"KBK" wrote:
A lens for a single curve screen is ridiculously difficult to make, and introduces nasty problems in application and implementation, especially for set up for multi-varied IPD's and potential eye issues.

If at all possible it would be best to go to DUAL curve and a meniscus lens set, which would eliminate just about all potential problems and be a FINAL design consideration that probably would not change until direct neural interface replaces it.

I am sure they won't go with a design that requires industrial grade manufacturing precision to work correctly. It will be interesting to see what screen tech they take from Samsung.

And with regards to this random thread, Oculus has prototypes that are supposed to far exceed DK2. Most of which will probably never see the light of day. Oculus is clearly paying attention to most of the correct problems when it comes to making consumer VR, this is why I think they will come to market with the best consumer VR experience. Not because of how DK1 and DK2 measure up to every interesting prototype that may pop up along the way.

I am certain defense research can come up with some compelling VR tech too along the same design lines as Oculus. That doesn't mean they give two craps about gaming or concerns that matter there. All the same the more VR research the better.

[KBK]

Once Samsung manufactures dual curve sections for/as OLED's...... it's all over.

Just the one innovation is required and it's done: "All necessities provided, all anxieties tranquilized, all boredom amused."

[MrMonkeybat]

"KBK" wrote:
A lens for a single curve screen is ridiculously difficult to make, and introduces nasty problems in application and implementation, especially for set up for multi-varied IPD's and potential eye issues.

If at all possible it would be best to go to DUAL curve and a meniscus lens set, which would eliminate just about all potential problems and be a FINAL design consideration that probably would not change until direct neural interface replaces it.

By dual curve do you mean curved vertically as well as horizontally so its like the section of a sphere? Current curved screens are made the same way as flat screens and flexed afterwards, such a concave screen would require much bigger changes to the manufacturing. Meniscus lenses flatten the focal plane allowing focus on flat screens easier any way, I suspect Lockheed might be using some.

And there could be something better in the future that is not a neural interface. Electro optics and holographic lenses could make a direct retinal projector that can cover your entire retina in any eye position in which your pupil is still visible giving you much greater FOV than real life and even the great Geekmasters :ugeek: design. Through foveated projection it could provide as much detail as a 16k screen with only 1 megapixel of beam scanning. It could also adapt instantly to any eye prescription and give you a printout to take to your optometrist.

[willste]

"mrmonkeybat" wrote:

"KBK" wrote:
A lens for a single curve screen is ridiculously difficult to make, and introduces nasty problems in application and implementation, especially for set up for multi-varied IPD's and potential eye issues.

If at all possible it would be best to go to DUAL curve and a meniscus lens set, which would eliminate just about all potential problems and be a FINAL design consideration that probably would not change until direct neural interface replaces it.

By dual curve do you mean curved vertically as well as horizontally so its like the section of a sphere? Current curved screens are made the same way as flat screens and flexed afterwards, such a concave screen would require much bigger changes to the manufacturing. Meniscus lenses flatten the focal plane allowing focus on flat screens easier any way, I suspect Lockheed might be using some.

And there could be something better in the future that is not a neural interface. Electro optics and holographic lenses could make a direct retinal projector that can cover your entire retina in any eye position in which your pupil is still visible giving you much greater FOV than real life and even the great Geekmasters :ugeek: design. Through foveated projection it could provide as much detail as a 16k screen with only 1 megapixel of beam scanning. It could also adapt instantly to any eye prescription and give you a printout to take to your optometrist.

I am not an optics engineer but I would assume that the current lens solution would have issues if you suddenly placed a curved screen in front of it, certainly it would change the distance to the pixels in the horizontal direction likely distorting the image differently. But its entirely possible that the image could be warped in a way that would negate the horizontal distortion this would introduce.

It definitely complicates the problem but if it can all be handled with software screen warping without any physical changes needed and it can give you some extra FOV will keeping size and weight in check then i don't see a problem.

[ImRoot]

kevinw729,I heard you quote that you were "toying with writing a feature on the system for RTVR".I don't know about anyone else,but I would love to see an article about this with more info and better pictures.

[MrMonkeybat]

I have been measuring the FOV of my vision with paper, pencils, doorway and table tops so I thought I might as well share my results here:

I measure a 120 degree vertical FOV and 220 degrees horizontal FOV. With a binocular overlap of 100 degrees with the bridge of my nose protruding 10 degrees in reducing it to 90, and the tip of my nose protruding in another 10 degrees. My cheeks don't seem to obstruct my vision at all and my eyebrows only where they meet my nose. So If you include my nose each eye has a FOV of 160 by 120 degrees a 4:3 aspect ratio. To make use of the entire FOV an HMD would have to have some exotic concave hemisphere lenses that wrap around the eye brushed by your eye lashes. If the DK2 has a vertical fov of 90 degrees to get the same angular resolution at my ultimate FOV would require a res of 1440 by 1920 per eye for a total of 1440 x 3840 or 5.53 megapixels.

If you use eye tracking the resolution outside of 15 degrees could be halved. That Microsoft foveated rendering experiment had a 1080 by 1920 monitor covering 60 degrees so scaling up there 1/5 peripheral resolution to 120 x 160 degrees would be 768 x 1024 but I expect the extreme periphery could be reduced even more. They had a 15 degree resolution of 240 pixel wide circle and within that a 5 degree foveal resolution of a 160 pixel wide circle. Which in the ultimate FOV HMD would require a screen resolution of 5120 x 3840 per eye.

[cleverusername]

http://techland.time.com/2013/04/02/an-interview-with-computing-pioneer-alan-kay/

http://www.reddit.com/r/oculus/comments/2887st/alan_kay_on_hmds_and_the_future_and_past_of/

This was just posted at redditt. (Where I am censored again even though palmer FREEMAN luckey was kind enough to gift me reddit gold and try to stop my censorship) It is a good article by Alan Kay. David smiths advisor and mentor. I enjoyed Alan Kay's perspective on how things have not progressed and some of the pitfalls that should have been avoided. His insight into the state of the industry was Also interesting. Especially his insights into funding, profiteering, corporate america etc. What do the rest of u think about Alan Kay's thoughts?

http://m.youtube.com/watch?v=iTgXOHAKfEM

The freemen from dune. Like Palmers middle name.

[cleverusername]

Dr. Jerald on David smith and Lockheed optics (Jason is one of the kewlest kats in kyrandia - where the hero always wore flip glops/sandals - like someone we all know palmer - who knew Westwood made a game w palmer as the hero)

http://www.mtbs3d.com/index.php?option=com_content&view=article&id=13883:dr-jason-jerald&catid=39&It...

http://m.youtube.com/watch?v=_c5QrnYYk3Y