CastAR launched on kickstarter

[Psi]

Have you ever used the Oculus Rift 3D Glasses? I did.. talk about heavy, hot, and not very good on the eyes ... i was not impressed, i had a hard time keeping my balance too afterwards, and my eyes were watering from the heat.

You did not try the oculus rift.  You tried the developer prototype.

[EEVblog]

http://www.kicktraq.com/projects/technicalillusions/castar-the-most-versatile-ar-and-vr-system/
trending toward $5,452,864

It seems to be going exactly as you'd expect. At an average $242 per pledge, it's to be expected the campaign would meet it's target just based on Jeri's existing loyal audience. It's how many additional gamer nerds it gets that will determine how high it goes.
I'd be surprised if it doesn't reach 10,000 backers, so at the same $242, that's at least $2.4M
That trending wouldn't be accurate yet because it's presumably extrapolating the initial loyal audience numbers, but yeah, probably not far off the mark.

[EEVblog]

I have a lot more faith in any project Jeri is working on than most other kickstarter projects.
We'll end up with a lot more than 2 slices of bread when this is done

And at least she's setting a realistic timeline. She will very likely need that full year to deliver, there is an awful lot to go yet, and a lot to go wrong too. Let's hope the ASIC works first spin!
My ridiculously simple PCB rulers are going to be a month late 

[Rasz]

$1M is reachable.
At this point the most important thing is exposure. Dragon/Comic/whatever-con goers should be priority for Jerri. Concentrating on Board games was a great decision - after all those are people with imagination

I was looking into retroreflective materials, alibaba has some listed starting at $2-3 per square meter, seems too cheap?

[mikeselectricstuff]

Re. focus. in this vid. she says they have a small aperture on the projector, giving depth of focus between a couple of feet and several yards

[mikeselectricstuff]

A few interesting technical snippets and other Jeri stories in this podcast
http://embedded.fm/episodes/2013/10/16/go-for-everything-that-i-want

[Markybhoy]

Twit done a video about this with Jeri recently -

http://twit.tv/tri124

[Balaur]

I'm still unclear on how they will manage to design and manufacture the ASIC in quantity. The $400,000 from the initial goal may not even cover that part (according to my experience as an ASIC designer). Not to mention production schedules (for rather recent 28/40/45nm processes, we are talking about 3-6 months between the tapeout and delivery), the sheer difficulty of a small company dealing with some foundries, or getting correctly-priced EDA tool licenses that allow commercial applications.

I will be really happy to see that I was wrong.

[krenzo]

It didn't sound like she was going with a full scale ASIC but one of those services that turns an FPGA into an ASIC with reduced NRE cost.

[thm_w]

I'm still unclear on how they will manage to design and manufacture the ASIC in quantity. The $400,000 from the initial goal may not even cover that part (according to my experience as an ASIC designer). Not to mention production schedules (for rather recent 28/40/45nm processes, we are talking about 3-6 months between the tapeout and delivery), the sheer difficulty of a small company dealing with some foundries, or getting correctly-priced EDA tool licenses that allow commercial applications.

I will be really happy to see that I was wrong.

She discussed this on podcast 147: http://www.theamphour.com/the-amp-hour-147-absorptive-augmented-actuality/
Its lower cost/faster turnaround: https://en.wikipedia.org/wiki/Application-specific_integrated_circuit#Gate-array_design

NRE was ~10-50k

[Balaur]

Krenzo, thm_w thank you for your comments. Much appreciated.

[Rasz]

I was just sitting down taking number two when an idea hit me:

 Big problem in VR is frame/tracking latency. Move your head too fast for displayed framerate and brain notices discrepancy.
What if you mounted projector lenses on micro galvo arms and compensated head rotation directly with the glasses? Shift current image in sync with tracker until video pipeline has new frame ready, then snap lens back into starting position during vsync.
Of course it wouldnt work for lateral movement, but maybe (no idea, my intuition, head mass, and shaking my head furiously tells me so) rotation is a more sudden and less predictable movement - easier to rotate big mass on its center than to move it.

[jancumps]

Well done technical illusions!
 

[EEVblog]

They cracked the $1M mark thanks to a last minute surge.
Great result of course, but a fair bit less than I originally guessed such a ground breaking gamer system would have generated. The gamer geeks didn't really seem to swarm to it like I thought they would. Just under 2500 who ordered the hardware is a good result though.
I wonder who the two people are that shelled out $10K each for a hand build proto?

[whatchitfoool]

I wonder who the two people are that shelled out $10K each for a hand build proto?

One of the 2 is a founder of the Oculus Rift!  How about that? using other peoples money from kickstarter to buy into another kickstarter to get a leg up on your competition for ones said kickstarter.
In his defence, he does have the worlds largest collection of head-mounted displays...

[Rasz]

excavating
http://venturebeat.com/2015/08/19/castar-raises-15m-from-android-creator-andy-rubin-to-build-augmented-reality-gaming-glasses/

boom! 15mil from former Googler. Jeri is moving up in the world

[sswift]

I don't get the hype about these.  It's a dead-end technology. 

It only works when you're looking at the retro-reflective screen, and it surely has poor contrast.  Have you ever used one of those pico-projectors?  You have to be in a dark room to be able to see the image unless you're inches away from the surface you're projecting onto.

And I say it's a dead end technology, because if you think about what ultimately you would want AR glasses to be, this isn't it.  You would want glasses that are lightweight, run on batteries for hours, can 'project' a solid high contrast image anywhere in a room, and don't shine light into other people's eyes while doing so.

The only way to get there that I can see is to have a pair of thin transparent OLED screen which can make portions of the display partially or completely opaque.  And that technology is so far removed from this, research on this won't lead to that.  Transparent OLED screens (flexible no less) are also right around the corner, and should be in commercial products in five years or so.  And Microsoft's offering, though flawed, is still far closer to the ideal than this.

[sswift]

A lot of long time forum members are probably thinking you sound like you know what you are talking about. NOT

By all accounts the retro reflector throws back most of the light and gives a bright image. Significantly it allows the viewer to see things at a natural optical distance and reduces headache and disorientation some users experience with other devices.

Why not listen to the old Amphour episode Jeri did with Chris and Dave to hear more about it. Then come back with your comments still not having worn a pair and seen for yourself.

It's funny you should talk about seeing for myself when you clearly have not seen them with your own two eyes and are just parroting back what others tell you.

I may not have used these glasses personally, but I have used the next best thing: a pico projector. And I can tell you that with a pico projector projecting onto a white surface in a well lit room, is barely visible with the projector a few feet away from the surface.

Now, the retro-reflective surface will improve this some, but there's one factor you cannot escape: Light cannot make a surface darker, and the retro-reflector will always appear gray.  That right there kills your contrast, and is why whenever anyone uses any projector, even ones which can put out far more light, they do so in dark rooms.   

Have you not noticed that every demo of this thing is in a dark room?

Here is a screenshot from their demo, with the image projected from the camera's point of view so you can see it:


Do you see any black in that projected image?  No.  The contrast is terrible.  And that's in a dimly lit room. 

Let's see what happens if we crank the light up in the room a bit:


This is simulated of course, but should be close to what the image would look like if the room were more well lit.  Now imagine if instead of it being a room of average brightness in the evening, it were a sunny day.  My living room is twice as bright as this with all the shades pulled down in the summer and I live in the northeast where we don't exactly get the most sun.

This really shouldn't come as a surprise either.  Just how much light do you think you could get out of three tiny LEDs in a headset without it getting up to 150 degrees in a few minutes?  They're probably driving the LEDs with no more than 1W so the headset gets warm but doesn't burn you, and they're spreading what little light they have over a large area.

Anyway, all speculation aside, the photo above proves that I said about the contrast.  Even in the best of conditions, it's not great.

[sswift]

http://www.roadtovr.com/castar-hands-latest-glasses-thru-glasses-gameplay-video/

At this point it’s worth pointing out that the combination of bright projectors and the unique properties of the retro-reflective material mean it’s possible to play with castAR in ambient light. Our time with castAR was at around midday and there were lamps and outside sources of ambient light, yet the projected world was clearly visible. As this is projected light however, playing this way means that contrast ratio is lacking, although your brain does a good job of adjusting

http://www.roadtovr.com/castar-interview-jeri-ellsworth-rick-johnson-speak-road-vr/

Contrast ratio is however obviously only as good as the darkest shade on you projection surface, which means CastAR will never excel at deep blacks.

http://hackaday.com/2015/01/12/castar-hands-on-and-off-record-look-at-next-version/

Got to try an earlier version at Maker Fair last year. A couple things:

* The field of view is limited. I understand they went through some trouble to get the widest beam angle possible out of the projectors, but it still doesn’t come close to covering your peripheral vision. This is not really noticeable, though, when you are looking at a table.
* The image above gives you a pretty good idea of what the contrast and colors are like. Its better in a dark room, but obviously projectors aren’t going to be as vibrant as an OLED screen strapped to your face.

[mamalala]

Imagine if people had the same mindset about telephones...

They are a dead technology! I mean, who wants those bulky things? They have a lousy sound quality, you need to be in a rather silent room to hear the person on the other end, and you have to almost yell into them so that the other person can hear you. And don't forget the cables. The cables make those fancy telephone machines quite stationary. Plus, who on earth would ever want to weave a web of cables to connect all those telephones?

Right?

Greetings,

Chris

[TMM]

The only way to get there that I can see is to have a pair of thin transparent OLED screen which can make portions of the display partially or completely opaque.

How is that going to work when you are focusing on something meters away and the 'AR' LCD is centimetres away from your eyeball? You'll only be able to focus on the LCD or the real environment, not both at the same time. CastAR is the only solution that completely avoids that problem without using ridiculously expensive optics. Obviously there are practical limitations, but alternative AR solutions like Hololens have other limitations that CastAR doesn't.

[sswift]

How is that going to work when you are focusing on something meters away and the 'AR' LCD is centimetres away from your eyeball?

I don't know, but Microsoft's Hololens couldn't work if they hadn't solved that problem.

You'll only be able to focus on the LCD or the real environment, not both at the same time. CastAR is the only solution that completely avoids that problem without using ridiculously expensive optics. Obviously there are practical limitations, but alternative AR solutions like Hololens have other limitations that CastAR doesn't.

Yes, but an augmented reality device which only works in dark rooms is hardly the end-all be-all of augmented reality.  The end-all be-all of AR would work outdoors, in the sun, and the batteries would last for a couple hours.  And a good step in that direction would be to work indoors, in a well lit room, tethered to a PC, when looking at any surface; not just a retro-reflective screen.   There is no way for Cast-Ar to ever evolve into the ideal.  The technology it uses is fundamentally flawed. 

Of course I'm not suggesting that one should never develop technologies that will never lead to the ultimate solution, but the need to use a special surface, and work in dark rooms... those are serious limitations.  And there is some serious competition coming very soon.  My opinion of Hololens isn't much more favorable, I think Microsoft has been misleading everyone by making it appear the images look solid and fill your entire field of view, but the potential for that technology to evolve into something better is tangible.  If they could just add an LCD in front which can turn selectively opaque they could probably make objects appear solid.  And the FOV thing can probably be improved a great deal. 

As for requiring expensive optics, pico projectors aren't cheap either my friend.  And CastAR requires TWO of them.

[sswift]

You remain entitled to your opinions and I'll continue to think you have rushed to judgement based on insufficient personal experience.

And does Gabe Newell also have insufficient personal experience?  Considering he and his team have practically reinvented VR while working with the Oculus team, I would say he has plenty.  Yet he decided that Cast AR wasn't worth continuing to invest in, and allowed the team to take the project they developed at Valve with them. 

I think he made the right decision.

But I guess we'll see what happens.  Personally, I think CastAR is going to go the way of the OUYA, and that had way more developers supporting it.

[sswift]

Imagine if people had the same mindset about telephones...

They are a dead technology! I mean, who wants those bulky things? They have a lousy sound quality, you need to be in a rather silent room to hear the person on the other end, and you have to almost yell into them so that the other person can hear you. And don't forget the cables. The cables make those fancy telephone machines quite stationary. Plus, who on earth would ever want to weave a web of cables to connect all those telephones?

Imagine if people were satisfied with yelling into pipes you mean.

I'm not against AR.  I'm against this implementation of AR.

And even the first telephones were game changing.  They didn't gain traction because someone had vision of what they could become with years of research.

[sswift]

Why are you even attempting to make a case that one is better than another? What of it?

Why does anyone here discuss the merits of different Kickstarter projects?  You're asking me why I have an opinion?

I have no ulterior motive.  I just think it's a waste of money, like solar roadways, or the Ouya. 

Is it technologically neat?  Sure, I'd be proud of myself if I built it.  That doesn't mean it's a good product.