MS Research: IllumiRoom *spin*

Back in August Samsung bought a projector technology company called bTendo. So Samsung has remained very intrested in developing projectors.
I thought it was ST-micro? Still, Samsung is interested in pico-projectors. They want to integrate them in all their portable devices, so using lasers and SLM (SOM?) makes a lot of sense, it's the only reasonable way to make pico-projectors.
 
To which I reply, so what? ;) What has Samsung adding pico projectors to its devices the same as Sony does got to do with MS securing a low cost, bright HD projector for IllumiRoom?
 
It'd just be a blurry image, or flickery. Or worse, a rainbow mess. A 180Hz DLP is projecting the same 60fps footage 3x to eliminate rainbow effects.

If there's any tech that's giving MS an advantage, it'd exist in the projector market itself also.

If there's any tech that's giving MS an advantage, it'd exist in the projector market itself also. Note there are cheaper laser projectors out there - that revolution happened, with projectors in mobiles and camcorders. They just aren't high res or bright yet. If they were, they'd have replaced lamp-based projectors.


Possibly, although the existence of the video doesn't really show that. MS have a patent. This is PR. Maybe they can drum up interest from 3rd parties for niche activities, rather than create a real home product? Even if they have a tech they want to introduce (which MS have always had with their RnD concept home - many, many techs that haven't yet made it to the mainstream), it would only be as an expensive extra like Sony's HVD. Maybe they can release it for £400 as a peripheral, but it's not going to be a standard component of Durango. It's way too costly and clearly niche.

Here's a concept vid from Sony showing Move's use in projection mapping tracked to a camera. It's somewhat different to MS's 3D scanning, but the same sort of idea with no special effects applied, released publicly. It's not a commercial project - just a proof of concept for PR. Like patents, the existence of a video doesn't mean the intention to create or release a product idea, and certainly it says nothing about the affordability of such a product.
There is a difference. That is clearly a conceptual vid not even trying to promote a game experience related product but the PSN itself ;)
The MS video is "promoting" it as a gaming related product that may see its way to the console
 
I'm sure people would like to say that MS created magical hardware technology, but they never did. Their strength in RnD have always been software and algorithms. Even the Kinect is practically an off the shelf product they bought from PrimeSense, but it's most interesting part is the software, the way they managed to take a messy point cloud and turn it into usable vectors and limbs, it's a big achievement. That's why I think if Sony adds a depth channel to the eyetoy, they're still a long way from the Kinect, they'll need to develop the software which is the hard part of this technology.

OTOH, the hardware part of this video looks like an ordinary projector, and the software part doesn't look difficult, it's just mapping the projection in the gaming space, and mapping the TV. It's trivial, maybe I'm missing something.
 
I thought it was ST-micro? Still, Samsung is interested in pico-projectors. They want to integrate them in all their portable devices, so using lasers and SLM (SOM?) makes a lot of sense, it's the only reasonable way to make pico-projectors.

Yes ST not Samsung.

I don't think they are using DLP for this stuff but images wouldn't flicker. PWM is used to generate many sub-fields. TI has said they are working on glasses free projection...how they will achieve that I have no clue...special screen maybe?
 
TI has said they are working on glasses free projection...how they will achieve that I have no clue...special screen maybe?
Because 3d requires each eye to get a different image, beyond 12 inches from the screen it becomes exponentially difficult to do auto-stereoscopy in any way, there's a way to track the person and beam to each eye, but it's not marketable yet, and only works with one person in the room, and definitely needs a special expensive screen.
 
So how does this work when you play with a PJ already, and the complete wall is filled with an image :)

Ok, i guess i could use it in my living room with my TV :)
 
Ultimately I could see this going beyond that one wall, but I'm not holding my breath on this tech at all, it's going to be a while.
 
Thats pretty good, the major problem I see with it (aside from the snow etc stuff) is the Programmers will have to work a lot more to support it, Sure worth it if everyone gets it but if its an add-on it'll prolly not be worth the bother
 
Kinect was a software solution using ordinary optical parts. There's nothing that can be done in the projector space using software to make hardware perform more impressively.
Laser scanners and direct green laser diodes are becoming ordinary now. For a darkened room with TV brightness turned down a bit they could probably do it in a 100-150$ device right now ... Microsoft seems unlikely to make such a niche device though, especially for core gamers (the only people who would darken the room to play games).
 
I have my doubts. The area they are projecting over is massive compared to a typical laser projection. Intensity is going to be very low - nothing like that demo vid. It's certainly doable in some form at some price, and I wouldn't be surprised if this is released as a product, but I expect MS will recognise the niche and charge a suitable premium to make it worthwhile to them. The projector would be sold as a projector too I expect - seems a considerable waste to buy a projector and have it only able to fill a room and not project a conventional large image.
 
Hmm, actually how about really lowering the resolution ... LED light density is getting pretty decent, you might be able to get away with simply scanning collimated "beams" from LEDs.

This would be substantially cheaper.
 
Hmm, actually how about really lowering the resolution ... LED light density is getting pretty decent, you might be able to get away with simply scanning collimated "beams" from LEDs.

This would be substantially cheaper.

Doesn't resolve the brightness issue though.

If you wanted to go monochrome, there are some very bright laser diodes, that could be coupled with a scanner of some sort. But i don't know what the lifetimes are like for the ones on the higher end of the power spectrum.

One possible solution might be to go back to a vector display, this resolves the brightness issue because you only scan the lit pixels.
 
Lumen per $ for LEDs is cheap enough to light up a room.
I've found an LED projector for $200 rated at 2000 lumens. I'm pretty sure that won't be bright enough to match the concept vid. LEDs will also need optics, which aren't cheap for ultra-short-throw projection. We're a ways off getting this at a main-stream price, I still think. Of course, when Durango launches it'll be a cheaper option, so maybe there will be a +$200 SKU with added projector?
 
I've found an LED projector for $200 rated at 2000 lumens. I'm pretty sure that won't be bright enough to match the concept vid. LEDs will also need optics, which aren't cheap for ultra-short-throw projection. We're a ways off getting this at a main-stream price, I still think. Of course, when Durango launches it'll be a cheaper option, so maybe there will be a +$200 SKU with added projector?

I hate to guess at figures like this, but given you are projecting on an imperfect "screen", with what looks like well over 200in diagonal in those clips, and have to do it with potentially poor light control, I'm going to guess you need at least 300 lumens and probably closer to 4000+ lumens, most home projectors don't get anywhere near that.
 
I had a 480p laser projector that was about the size of an ipod touch (including the battery) that could pretty much replicate what I see in that video. Note that the room is pretty much unlit except for the projector and TV. It's not meant to be used in a well-lit area.

That advantages of the laser projector I has was that it needed no focusing, it was always in focus, no matter how far away you held it from the wall. With the recent LED and laser diode advances, I could easily imagine an addon device in the $200-$300 range. the trick is to get an image of our room, it needs a camera as well as a projector, and you can't use the kinect, since it would have to be behind you.
 
I'm not talking about imaging projection, imaging doesn't work any way because you can't rely on a planar projection surface and even if you did you don't want the player to have to fool around with adjusting the optics for focus.

What I'm talking about is using really low resolution (lets say 128x128 or something like that). At that point the pixel size becomes large enough you might be able to use a collimated LED beam(s) and scan them across the room ... just like with scanned lasers you don't have to worry about focus in this case. You don't have to worry about concentrating all the light in a tiny homogeneous lit square (this limits how much lumen you can get into a LED projector) and you don't have to pay for high quality imaging optics (which is not to say the optics are straightforward, but it's all going to be acrylic so cheap in mass production).
 
I'm not talking about imaging projection, imaging doesn't work any way because you can't rely on a planar projection surface and even if you did you don't want the player to have to fool around with adjusting the optics for focus.
Sorry, course you weren't. :oops:

What I'm talking about is using really low resolution (lets say 128x128 or something like that).
128x128 is WAY below the concept vid. I don't think SD would cut it either, although it might.
 
I had a 480p laser projector that was about the size of an ipod touch (including the battery) that could pretty much replicate what I see in that video. Note that the room is pretty much unlit except for the projector and TV.
The projection is lower brightness than the TV, but still pretty strong. I'll experiment with my projector at some point and see what I find personally.

That advantages of the laser projector I has was that it needed no focusing, it was always in focus, no matter how far away you held it from the wall.
Yeah, lasers a given. Won't really work without. Well, it might as it's peripheral.

With the recent LED and laser diode advances, I could easily imagine an addon device in the $200-$300 range. the trick is to get an image of our room, it needs a camera as well as a projector, and you can't use the kinect, since it would have to be behind you.
Ha ha! Good point I had overlooked. You'd need a projector + imple Kinect (just camera) in a package. I expect Kinect level scanning is pretty cheap now though. Could you not also move the Kinect by the projector for scanning and, as long as the projector isn't moved, that'd work? Not very user friendly but an acceptable cost-saving measure methinks.
 
Back
Top