Inside MultiTouch: Team, Demo, Lab Tour

I've always found Surface to be too big because it needs to be rear projected and has cameras. This is far more usable and I hope we see a real product emerge from this soon.

Again this "Unable to load media" message, like posted in several other threads.

Arrrghhhh..

I wonder if the customers are ready for this technique.

And please someone tell mr butler to cut his nails 

Yeah, hard to enjoy this video with the popup, and then the inability to actually play the video. Oh well, back to downloading...

So how does the iPhone do all its multi-touch stuff then?

So, they mentioned resolution, but what other reasons are there not to use this for Surface computing? Looks like it'd cut down the form factor, which would immediately make it a lot more suitable to more environments. That last one they showed was already a small table, after all... If they stuck the Surface software on one of these devices, what wouldn't they be able to do that a normal Surface computer is able to do?

The tablet pc/IR sensing hybrid they mentioned sounds great, by the way. Manipulation using your fingers, and then writing using a stylus? That sounds ideal.

Great video, Charles. MSR Cambridge is fascinating.

Loved that... going back soon, once they've advanced a bit, I presume?

 

Thanks for another great vid;)

Hey, the error message is gone! I don't know what you did, but thanks!

Bas wrote:

Hey, the error message is gone! I don't know what you did, but thanks!

Still there for me. They go away on occasion (on the same videos) but are there more often than not.

Nuts, you're right, it's back now.

Great job MSR Cambridge

If anyone wants to DIY this check out NUI Group

hard working hassall wrote:

So how does the iPhone do all its multi-touch stuff then?

The iPhone uses a capacitive touch panel built in the display.
Apple shows it some at the following two links:
http://www.apple.com/iphone/features/index.html#technology
http://www.apple.com/iphone/features/index.html#touch

Oggelito wrote:

I wonder if the customers are ready for this technique.

And please someone tell mr butler to cut his nails 

Are you suggesting he should switch to nailless classic guitar playing technique?

Long nails might provide some additional benefits like tweezers substitute.

Apple's tech is cool but may be considered 'old school' when compared to this approach.

They're attempting to show the benefits of using IR over the traditional capacitive touch panels.
 
As noted in the video, using a smartphone, TV remote or any IR pointer to control the UI is very cool.

A likely product from this is a "low-cost" Multi-touch TV Remote.

PocketXP wrote:

 Apple's tech is cool but may be considered 'old school' when compared to this approach.

They're attempting to show the benefits of using IR over the traditional capacitive touch panels.
 
As noted in the video, using a smartphone, TV remote or any IR pointer to control the UI is very cool.

A likely product from this is a "low-cost" Multi-touch TV Remote.

Bear in mind that an IR remote does not have the pinpoint focus of a laser pointer; it's diffuse. So, regardless of the resolution of the IR receiver array, IR reception from a distance will be as accurate as a BIG fat finger. Point-click-move UI control will only be possible in either close proximity to an average-sized screen or maybe from across the room from a 80+ inch screen. UNLESS it's capable of accurately detecting an IR point source from a distance.

If not, as with standard IR remote control from a distance, any IR remote in this scenario will still have to function via a binary protocol and will translate "next", "prev", "first", "last" buttons and arrow keys to manipulate the UI. One way to resolve this is to have the remote control device fitted with an electronically-controlled focusing lens and at the beginning of each remote control session have the device and computer communicate and perform a quick auto-focus function that will ensure that the IR beam is in perfect pinpoint focus relative to its distance from the screen. Then you can have accurate point-click-move UI manipulation. Another way would be to have the remote control itself have an embedded multi-touch system, as you mentioned, that transmits touch information to the screen via a standard IR binary protocol.

I'm no Apple fanboy, but Microsoft's technology may appear "old school" when compared to Apple's "next" approach, their full screen image sensor array technology. The apple screen will be able to see you... from across the room. If the image/object analysis is good enough it will be able to detect when you are pointing or drawing with your forefinger. Draw out an imaginary screen with your forefinger to allow the Apple screen to register the X/Y extent of your motions (or automatically and constantly adjust according to perceived shoulder width to allow and account for changes in proximity to the screen) then use multiple fingers, elbows, eyes, eyebrows, mouth shape to point-click-move, draw and communicate to your hearts content. This could also extend to multi-person interfaces down the road where the faces and gestures of a variable number of people control the UI and applications.

A hybrid of the two technologies may solve a lot of issues and create new opportunities that neither alone can address. I could see a cross-licensing agreement between Microsoft and Apple. If Apple's image sensor can vary or expand the light spectrum that they detect into IR then all that would need to be added to each pixel would be the IR transmitter element.

A cross-licensing agreement and possible cooperative research and development could save time and money leading to a more robust product and an industry standard rather than competing technologies that fragment the marketplace. Individually, if costs of the technologies are close to being equal then I imagine that Apple's future tech will find greater adoption as, on the surface and from pure speculation, it seems to facilitate all of the potential applications of MS MultiTouch and more.

Oggelito wrote:

I wonder if the customers are ready for this technique.

And please someone tell mr butler to cut his nails 

Considering it was only on one hand I'm assuming he plays guitar with his nails.

So MS’ Surface Computing uses a rear projection/sensing system which has high sensing resolution but is bulky and expensive. MS’ MultiTouch uses tiny, rear-mounted infrared transceivers, offering a more compact (and cheaper?) setup than Surface but with lower resolution.

I wonder what the pros and cons of these technologies are compared to the iPhone’s use of capacitive sensing technology?

I can see that Microsoft’s methods would allow the use of a stylus or gloved finger, whereas the iPhone will not. And Microsoft’s methods (presumably) work with a hard glass surface, so you don’t have the fragility or limited lifespan of a resistive touch screen’s soft plastic (polyster) surface.

But I don’t think that support for non-conductive input devices (gloved finger/stylus) alone necessarily justify the cost and complexity of these over the iPhone’s technology, so I’m curious if there are other advantages to these.

iPhone accomplishes it mostly via great PR.   

They use a capacitive touchscreen, which provides non-paired point data (i.e. x1, x2, y1, y2).   So you'll find the only true multi-touch gesture they support is pinch.  They can tell you have at least two fingers down, and can tell if the two outermost fingers are getting further apart or closer together, and where the aprpoximate midpoint is, but that's about it - they can't truly do multi-touch.  For example, fire up the ons-screen keyboard, press the shift key with one finger and try typing any another key.  You can't.  It's a modal-shift keyboard in large part because they can't tell the difference between (X1,Y1)(X2,Y2) and (X1,Y2)(X2,Y1).

One big advantage of capacitive over the more common resistive is that optical clairty is far better, as evidenced by the super sharp display on the iPhone.

Check this multitouch technology I developed with a friend.
It still a prototype, but the concept is valid

www.youtube.com/watch?v=jNCOs1TGwTY

It does not have the ghost points problem like capacitive, and can detect N points without problem. The dimensions of the prototype are big because we made it by hand and needed space to work, the actually size could be 3mm thick. This technology can be integrated to a LCD screen.
Also, this technology is very cheap, the prototype cost us US$20, and we estimate that the cost of do it in a 20'' screen is less than US$100

I promess to send one with audio next time