They are selling a software dual touch which works on existing touchscreen hw (you must change only your single touch IC with their dual touch IC), but why fake ?
If it works correctly and from your sw side is seen as a real dual touch where is the problem ?

Well, the difference is whether or not it's "emulation" or true "multitouch". It can still be true multitouch and be a software thing - the ability of a capacitive screen to figure out how many touches are doing what is a software thing too, I think - the hardware detecting differences in current is approximately the same in all cases; the software that calculates the differences is what changes, I thought.

Anyway, the way I understood that article, they were saying they used 4 layers of screen, instead of 2, to get the multitouch solution they were using. (Which is basically like having two resistive screens fused together.) Obviously, more technical details than that, but that was the general idea as I understood it. I may have read it wrong though....

Yes, basically a resistive screen is made of two planes of parallel cables (one set perpendicular to the other) with elastic spacers in between. When you touch a point you actually short one horizontal cable say y=2 with one vertical x=1. So you have your coordinates. If you touch two points you have two x's and two y's so four possible points. That is the inherent problem of basic resistive screens. Stantum screens on the other side have 3 layers of conductors (one being diagonal I think) so they can do the math and produce two seperate touch points.
HOWEVER those 4 possible points are perfectly enough for both pinch and rotate, the most used multirouch gestures.


Ok i replied to an old post by accident, but i leave the reply there for historical reasons. Also I stated 3 layers probably incorrectly. More than 2 anyway.
________
How To Roll A Joint

You mean the Fujitsu stuff? The issue that I find is that the mention both a 4-wire resistive screen (which is pretty much the most common one), and a "software based method". But nothing about layers..

Capacitive multi touchscreen usually uses 200-240 sensible points (capacitors) and theorically can relieve one contact about at every 5mm (on 4" screen).
But they can relieve multi contacts in parallel so then software interpolation or other software methods increase the effective resolution but I doubt they can be pixel precise (above all on screen with 240 or more dpi) as resistive touchscreen are claimed

Four layer is refered to Stantum screen ?
Because Fujitsu solution is four wires (so a normal resistive touchscreen) and not four layers

PS
I doubt that a single accelerometer can precisely emulate a gyro.
I have partially read some complicated articles that claims to make gyro emulation using two accelerometers and even so is it is a poor/imprecise emulation

No, I was referring to the Fujitsu one - I misinterpreted the four wire thing to mean four layers.

Re: qwazix: Actually, Stantum uses the same normal amount of layers too. As I said in the longer previous post I made a page or so back, the difference is they have spacer dots. A normal resistive screen can detect every single one of their sensor 'points' (intersections of the wires in the screen) at once - it's just that it has no way of understanding that it's multiple touches, vs one odd-shaped one, because two touches push in the screen area between them, and thus trigger all those points in between.

What the Stantum screen's spacer dots do is prevent this 'spill over', because the spacer dots keep the pressure on any given section of the screen from pushing down any part of the screen beyond the sector contained within a set of spacer dots.

The only problem is that it means that they can't detect true pressure differences per se (unless each section is itself precise enough to have multiple sensor points); because a normal resistive screen measures pressure based on how far away from the 'center' of the touch the sensor points are activated by the press(es). IE, I press lightly, screen bends in only so much, and only so many points around the touch get pressed. I press harder, and a larger area of the screen it pressed to the point of contact.

However, because of the spacer dots, the screen will never get pressed down in more sectors that the touching object actually touches. (So - and they admit this on the site - the way Stantum screens detect pressure is by how much area is affected - which makes sense for objects that bend/flex/have-any-give-and-area, like fingers, paint brushes, etc. A finger that presses down harder flattens itself more, and thus pushes on more sectors. However, I could drive a needle through the screen so hard it stabbed though, and the stantum screen wouldn't be able to detect the pressure difference from that, and a feather light press of the same needle. I'm sure this would be resolved as resistive screen sensor resolution becomes better, as sectors could have far more sensor points each, and math done on how much pressure each sector has would give a good pressure measurement on top of sectors-pressed-based measurements.)

Alternatively, someone could invent pressure-detecting spacer dots soon enough, which would handle that part of it.

Precisely, no, not exactly. I AM pretty sure that it can emulate most of the functions with complicated enough mathematics. However, much like emulating multitouch on a device that can't distinguish more than one touch at once, there's going to be cases where the emulation confuses some other set of events for normal gyroscope behavior. EG, a certain type of device movement can make the accelerometer values be exactly the same as they would be during the gyroscope-detected rotation, etc.

I haven't read anything on the subject though, nor have I thought TOO detailed'ly about this, so it's probably true that there's other exceptions I haven't thought of that are that much harder to properly account for.

Either way, gyroscopes remain far more resource efficient, probably far more so to justify using gyroscope emulation on an embedded device regularly if at all.

With one accell you can detect all orientation changes except those along the axis aligned with the up/down vector

With two accells you would think things would improve, you do get better precision along two axes and you can detect yaw changes, but now you're still blind to rotations when the line between the two accells is aligned with the up/down vector.

Now with three 3 axial accells there is no blindness to absolute yaw and no gimballock, but i think there is still some hidden issue that would make 4 non-coplanar 3 axial accells necessary ( i once read a little essay on the web on this and i have the impression it explained things all the way to 4 acells)


And that is all with perfect accelerometers, the cheap ones used in mobiles and game controllers have little jitters in the measurement big enough to easilly drift the rotation calculations off the real value fast enough to not be usefull for anything but the briefest usages (even the most expensive ones would still suffer from too much imprecision due to quantum fluctuations and stuff)


Gyrometers (most electronic gyroscopes actually measure rotation rate and not absolute rotation) still suffer from imprecisions and drift, but because the raw values from them are a few steps closer to the result in the math the imprecisions don't get amplified as much and it takes much longer to drift away from the real value significantly.

Hm, would it be possible to "hack" the touch sensor in the N900, switching wires from positive and negative to the opposite, and measuring the output on both sides, allowing one to tell the difference between single and multtouches?