Slashdot Mirror
210 Degrees of Heads-Up Display: Hands-On With the InfinitEye
First time accepted submitter muterobert writes "InfinitEye is a prototype head mounted display that uses dual 1280×800 displays to create a massive 210 degree field of view. I traveled to Toulouse, France to be the first journalist in the world to go hands-on with the unit. These are my thoughts on the trip, the team, and the HMD itself. 'Natural and Panoramic Virtual Reality' is the best phrase I can come up with that summarises the InfinitEye's capabilities. If using the Oculus Rift is like opening the sunroof on a virtual world, the InfinitEye takes the roof clean off — at least if you base your opinion solely on horizontal FOV. But the new HMD also offers 1280×800 per eye in comparison the current Oculus Rift Dev Kit's 640×800 (and only slightly fewer pixels per eye than the Oculus Rift HD prototype)."
210 or 120? :>
This might be a new record or maybe not. The headline currently states "120 Degrees..." when it should say "210 Degrees..." Summary and article both state 210 degrees.
See my journal for slashdot ID's by year. Mine created in 2005. http://slashdot.org/journal/289875/slashdot-ids-by-year
From the video the latency seems acceptably low.
The border edges don't need to be very high res, since our eyes are only high res at a small area, and you don't normally use only your eyes to look all the way to the left/right, you'd normally also move your head a bit towards the area you want to look at.
First you allow a submission with a summary that completely contradicts the article; then you put the wrong specs in the headline when it's clear from both the summary and the article itself what the correct specs should be? Get yourself a cup of coffee, come back and try again.
Everyone wants these for VR, but I want something I can code with. Anytime, anyplace. Maybe even laying in bed with a feeding tube snaked down my nose. :-)
How close are these to being a replacement for a reasonable monitor? (Absolute minimum would be 24x80 text that's usable without headaches/etc for several hours at a stretch.)
I'm much more interested in having a single display that is ultra wide, with the possibility of it being curved. For work, it's not bad to have a bunch of monitors as they will most likely contain separate windows anyway. But for gaming, having a single, continuous monitor with no borders in the middle works a lot better. Currently, you either have to have 3 monitors, so the middle of the screen isn't obstructed, or have 1 monitor. Having a single monitor that's as wide as 2 monitors would probably be wide enough for a lot of tasks. Also, even for work, It would be nice to have an ultra wide monitor, because there would be more usable space. There's kind of a dead zone in the middle because you don't want windows sitting between 2 monitors. If you have 2 monitors, it's hard to display 3 things side by side (Firefox , IE, and Chrome for instance).
Anthropic principle: We see the universe the way it is because if it were different we would not be here to see it.
Yes it is so big, you would type 120 when you mean 210.
sed -e 's/Chuck Norris/Rajnikant/g' joke > fact
These people are just jumping on the wave being created by Oculus Rift.
The current Oculus Rift is concentrating on latency, not resolution. Extra resolution is a given once all the other problems are finally sorted out. Delaying the high-res screens is a good move because it gives screen technology a bit of time to advance and keeps the dev kits dirt-cheap at the same time.
So...don't put too much emphasis on big headlines about screen resolution. The final Oculus Rift may well be lag-free and 1080p per eye when everybody else is still figuring out the latency issues on their high-res headsets.
No sig today...
CastAR has a higher resolution and it can switch between AR and VR as well as Projected AR. Get on the Kickstarter now!
http://www.kickstarter.com/projects/technicalillusions/castar-the-most-versatile-ar-and-vr-system
boom goes the dynamite....
I think we're approaching a point where it's within the right price range for the most practical consumer use: gaming.
VR was hyped heavily because people thought a 3D world would naturally be more intuitive than a 2D world. Evidently this is no longer a big deal, and VR for the office environment (aside from a few niche areas) probably doesn't really add much at this point.
But the novelty factor and immersion capabilities combined with the decreasing price point (both the head gear and the PC to actually drive it) would seem to make it a big potential for gaming.
Somewhat on-topic: there is an awesome BBC miniseries called "The Machine That Changed The World" from 1992. The first 3 parts are timeless and provide a pretty damn good overview of the history of computers. The last 2 parts aged less well, however they provide a very interesting view into what some very smart people thought the future would be like with regards to AI and user interfaces.
There was a lot of hype about VR stuff in the 90s, and the whole thing did not get much traction.
Because they were rubbish. Back in the 90s you'd have to pay $100,000 for something that was worse than the $300 Oculus Rift devkit.
(plus another million for a computer powerful enough to drive it)
Are things significantly better now?
Yes.
No sig today...
In fact, as they were showing, this display has a wider horizontal FOV than you can actually see, leading to wasted space. They also seem to be using last-gen 7" tablet displays (1280x720 or 1280x800), which are good, but something like the display from an iPhone 5S on each side would make it lighter, higher resolution, and somewhat more immersive. Though it also seems to me that this system could be driven by most higher-end video cards natively (albeit with an added software shader to create the fisheye-like effect needed for the fresnel lenses). So make a standard head-tracking mount, with modular and interchangable displays that run off a standard connector (MicroHDMI, for instance, or Micro DisplayPort). You could even have the same lenses, so it's just the display being changed, and then the displays themselves could also be used as tertiary information displays on systems. It would give the product longevity, and upgradability, and would require no software changes (Other than to pick the new, higher resolution for the displays in-game), and no hardware changes to the HMD if you use a standard size and mounting (5" smartphone screen, say).
The Occulus has a ~100% overlap factor, meaning that the same arc of FOV is presented to both eyes. Put another way, the left and right sides of both eye views are the same.
This device has less than 100% overlap. I'm guessing it's around 60% from looking at the monitor images. When the overlap decreases too much, it gives you the impression of having a very large nose that blocks each eye from seeing part of the other eye view. This can be annoying.
The overlap factor for real people varies, of course, due to facial structure. But you don't really want a device that has a much smaller overlap than your actual body has.
It is extremely difficult to maintain a large overlap factor as FOV is increased. The right side of the left display will encroach into the space needed for the left side of the right display. Avoiding this requires making the displays smaller and closer to the eye, which increases the demands on the optical system to refocus the image. In addition, there is less space for eye glasses, and other useability parameters may also be reduced (although weight can be decreased). At some point, you can no longer look at lens-based optics, and have to take a different approach altogether.
Note also that increasing the FOV tends to make the rendering a more difficult job as well. Fortunately, this isn't as big an issue these days.
The problem is that LCD panels have horrible color and brightness shift issues as you view them at an angle, and the eye here is so close that different parts of the screen are at dramatically different angles to the eye.
It _may_ be possible to solve some of that in software. Or it could require the use of OLED displays.
There was a lot of hype about VR stuff in the 90s, and the whole thing did not get much traction.
They're developing consumer versions that are far superior (and cheaper) than the $1000 minimum 256 color, low FoV junk from the 90's (looking at you VFX-1!). Better, professional units quickly went up to the 10s of thousands of dollars.
Are things significantly better now?
The reason why it's better now is due to cheap high resolution displays (thanks to phones and tablets) and precise accelerometers and gyros.
On the Occulus Rift side, they sidestepped the old design requiring two separate screens by using one screen split between your two eyes and using optics to make the narrow (per eye) screen appear wide. Also the optics concentrate more pixels in the center of your field of view (where you need them most). The distortions created by this are counteracted in software. So this new approach + cheaper displays + cheaper sensors = time for cheap and awesome consumer VR headsets!
Georgia Tech, the leader in Chia(tm) technology.
Because they were rubbish. Back in the 90s you'd have to pay $100,000 for something that was worse than the $300 Oculus Rift devkit.
(plus another million for a computer powerful enough to drive it)
Your second point has more validity than the first. Computing power was the real issue. Reasonable (if heavy) HMDs weren't even $10k, much less $100k. Its been a long time, but I think most of the ones I used in the mid 90's were in the $2k-$3k range.
But even more than that, the market for that kind of device is vastly different now. In the 90's, you didn't have anything close to the kind of market for real forward looking devices like that. It was a time when AOL ruled the space for what little of the general population was doing things on computers, and anything VR related wasn't just nerd but a real fringe of nerd, likely in academia.
Given the availability of the high density display technology in use in cell phones and such, I think it's absolutely shameful that the VR companies keep plugging crap like this instead of actual high quality displays. There is just flat out no excuse for it.
I do not fail; I succeed at finding out what does not work.
Memory of stuff like that is hazy at best, and it's been 20 years....
I remember trying a few systems back in the '90s and being distinctly underwhelmed(and that was back when just being near an SGI workstation was exciting)
I bought an Oculus VR dev-kit last month and I'm quite impressed. Yes, the screens are low-res, but I knew they would be (and I know it will be addressed). The overall impression is very good though.
No sig today...