Resistive Touch Screen As A Scale?


Status
Not open for further replies.

devhace

Member
Joined
Oct 1, 2008
Messages
147
Okay so apparently the touch screen has like 16 levels of sensitivity right? Wouldn't that mean that there must be certain amount of force needed for every level? If the amount could be determined then could we maybe use it as a scale? This would be incredibly useful for weighing erm, small things.
 
devhace said:
If the amount could be determined then could we maybe use it as a scale?
No, not with any degree of accuracy. Resistive pressure sensitivity is very half-assed, and the degree of accuracy varies greatly depending on the shape and composition of what is making contact. See here for a more thorough explanation.
 
Last edited by a moderator:
Mjlink said:
I've wondered the same thing about my iPhone wishfully.

Bummer
The iphone uses a capacitive touch screen. That is why it is capable of multi-touch, but only has one level of sensitivity per point of contact.
 
Last edited by a moderator:
notaz said:
fahrstuhl said:
BTW, speaking of the touchscreen... Are these actually capable of recognizing variable pressure? Or do they just have 'click' and 'no-click'?
I've been messing around with the driver, and yes it can do some pressure sensing, returns around 16 different values for pressure.

For those interested how it will be exposed, it will be (already is, actually) accessible through standard Linux event interface (raw values), or tslib (calibrated).

Doesn't answer the scale question, but interesting info none the less.
 
Last edited by a moderator:
Gruso said:
notaz said:
fahrstuhl said:
BTW, speaking of the touchscreen... Are these actually capable of recognizing variable pressure? Or do they just have 'click' and 'no-click'?
I've been messing around with the driver, and yes it can do some pressure sensing, returns around 16 different values for pressure.

For those interested how it will be exposed, it will be (already is, actually) accessible through standard Linux event interface (raw values), or tslib (calibrated).

Doesn't answer the scale question, but interesting info none the less.


chip pretty much answered it... maybe it will work, but you would need a fixed-size object between the weight and the display. Also, the display must be recalibrated before pretty much every usage.

Nobody really knows the scale and the steps it can detect, so wait for the final device.
Writing a program for this function shouldn't be too hard.
 
Last edited by a moderator:
conso said:
maybe it will work, but you would need a fixed-size object between the weight and the display.
Even then it's going to be hideously imprecise. It's nowhere near sensitive enough to distinguish increments of single grams, let alone fractional grams. It's a stretch to say it could differentiate between single whole ounces. At that resolution, you'd be better off just hefting the object in your hand and making an educated guess.

It was not a bad idea, but it is completely impractical and unworkable in any meaningful way.
 
Last edited by a moderator:
It might just about be able to say "First object was heavier than second object", but even then if they were too light, it wouldn't know the difference.

Secondly, why would you want to put various objects onto your Pandoras screen? Personally, I'd want to not touch it at all.
 
Squidge said:
It might just about be able to say "First object was heavier than second object", but even then if they were too light, it wouldn't know the difference.

Secondly, why would you want to put various objects onto your Pandoras screen? Personally, I'd want to not touch it at all.
I can imagine what those 'erm, small things' are B)
 
Last edited by a moderator:
mali said:
I can imagine what those 'erm, small things' are B)
I think we can all read between the lines there. Since it is clear that the Pandora will be completely useless for this particular purpose, I'm going to close the thread now before anybody starts incriminating themselves.
 
Last edited by a moderator:
Status
Not open for further replies.
Back
Top