Touchscreen on Cnet, how are they doing this ?

[Modelworks]

I just downloaded a cnet segment on my tivo where they reviewed a new touch screen design by microsoft called surface computing.

It was basically a flat screen made into a table like surface that was the size of a 30" screen.

They were able to draw on it with there fingers,using multiple hands, two people at one time, use a regular paint brush to paint on it like it was the real thing.

By placing small chips on objects they could also make the table recognize and interact with real world objects.

One example they gave was ordering food off a menu built into the table that allowed you to scroll the menu with your hand on the table, order, etc. It also could recognize your credit card by placing it on the table and you or the person across from you could split the bill, pay , etc .

I know how regular touch screens work, but this seems like a significant leap above that.
The fact that it could allow painting with a standard paintbrush ( not chipped, etc) and show the bristle detail in the painting was really impressive.

Found a link to the video :
http://reviews.cnet.com/Micros...0-10621_7-6737942.html

 

[NikolaeVarius]

This is news is a bit old., but its very interesting.

But, I think it might be 10 years before we see this in consumer technology.

 

[SuperFungus]

I saw one way they do this multi touch stuff awhile ago, so if i screw something up in my reconstruction please be gentle .

basically you have a plane of plexglass or something, i don't know exactly what they where using. with a cross section like this:

____________________________________

____________________________________

and you beam IR through it in such a way that you achieve what i believe they called "frustrated total internal refraction". like this

____________________________________
/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/

What you have now is a plate of glass which has IR light in it, but because of the index of refraction none of it (an even amount of it perhaps?) escapes. But, wherever you, or anything else touches the surface of the screen the index of refraction changes and your previously even field of IR gets disrupted. A camera positioned behind the screen tracks these disruptions and turns them into x-y coordinates for your computer.

I wish i had the link to the website i was reading about it on, but i can't seem to find it. Again this is all just stuff i remembered so anyone who knows better feel free to set me strait. hope that helps.

 

[Modelworks]

Thanks for the info.
I knew it had to be something more than the capacitance type touch screens.

 

[makken]

Originally posted by: SuperFungus
I saw one way they do this multi touch stuff awhile ago, so if i screw something up in my reconstruction please be gentle .
[...]

strange, I don't see how that can track very precise positional information though - ie. how do you know the IR light escaped from a specific location on the screen unless you embed it with a huge number of sensors?

An interesting note is that that technology is also in used for "rain sensing wipers" on cars - ie. they send the light with perfect reflection along the windshield, droplets of water on the glass disturb this, and let some of the light out. A sensor at the very end measures the amount of light loss, and turns on the wipers if it is significant. My mom's car has the first generation of this tech, and while its interesting, it's far from perfect.

 

[Nathelion]

I figure these screens would be extremely vulnerable to scratching. Bring out the silk gloves, people! Let's all be british 19th century aristocrats wearing white silk gloves and eating plum pudding.

I don't really get the camera behind the glass either. You could localize leaking light, sure, but would there be light leaking backwards in the first place? Frustrated total internal reflection would work as a method of creating disturbances since the evanescent wave would transfer energy into the touching material, but that doesn't mean that any light would exit through the back of the pane. To do that it the reflected light would have to change frequency, which wouldn't happen - there would only be an intensity loss in the "ray" passing by the contact area. That would create an interference pattern that could possibly be detected with sensors placed along the edges of the pane, but I still don't see how it would leak backwards.

Hmm. Very interesting.

 

[Zolty]

Behold the power of surface.

 

[Aluvus]

They use several infrared cameras.

 

[Zumbador]

The problem I see with Microsoft Surface, Apple iPhone, etc, is fingerprinting/smudging. Is SC Johnson going to make a killing in the next few years thru increased Windex sales?

Then again, most people are slobs and probably don't care. I, on the otherhand, and paranoid about this sort of thing (even though I own a machine shop with oil everywhere).