Ok so people have been confused over the tech thats behind the Iphone/G1 capacitive dual touch screen, and all WinMo phones resisitive screens. So i thought i would do the great job of researching the tech behind both methods. 10 minutes later i came across this info on how both methods work, as well as a more expensive and even better method using acoustic mumbo jumbo. So for all you out there that are too lazy to do the diggin, heres the info that once and for all puts the rumors to bed about "...i heard its dual touch capable from so and so website..."
PLEASE READ>>>
The resistive system consists of a normal glass panel that is covered with a conductive and a resistive metallic layer. These two layers are held apart by spacers, and a scratch-resistant layer is placed on top of the whole setup. An electrical current runs through the two layers while the monitor is operational. When a user touches the screen, the two layers make contact in that exact spot. The change in the electrical field is noted and the coordinates of the point of contact are calculated by the computer. Once the coordinates are known, a special driver translates the touch into something that the operating system can understand, much as a computer mouse driver translates a mouse's movements into a click or a drag.
In the capacitive system, a layer that stores electrical charge is placed on the glass panel of the monitor. When a user touches the monitor with his or her finger, some of the charge is transferred to the user, so the charge on the capacitive layer decreases. This decrease is measured in circuits located at each corner of the monitor. The computer calculates, from the relative differences in charge at each corner, exactly where the touch event took place and then relays that information to the touch-screen driver software. One advantage that the capacitive system has over the resistive system is that it transmits almost 90 percent of the light from the monitor, whereas the resistive system only transmits about 75 percent. This gives the capacitive system a much clearer picture than the resistive system.
On the monitor of a surface acoustic wave system, two transducers (one receiving and one sending) are placed along the x and y axes of the monitor's glass plate. Also placed on the glass are reflectors -- they reflect an electrical signal sent from one transducer to the other. The receiving transducer is able to tell if the wave has been disturbed by a touch event at any instant, and can locate it accordingly. The wave setup has no metallic layers on the screen, allowing for 100-percent light throughput and perfect image clarity. This makes the surface acoustic wave system best for displaying detailed graphics (both other systems have significant degradation in clarity).
Another area in which the systems differ is in which stimuli will register as a touch event. A resistive system registers a touch as long as the two layers make contact, which means that it doesn't matter if you touch it with your finger or a rubber ball. A capacitive system, on the other hand, must have a conductive input, usually your finger, in order to register a touch. The surface acoustic wave system works much like the resistive system, allowing a touch with almost any object -- except hard and small objects like a pen tip.
As far as price, the resistive system is the cheapest; its clarity is the lowest of the three, and its layers can be damaged by sharp objects. The surface acoustic wave setup is usually the most expensive.
How Stuff Works people, its a great website. heres a picture i made to show you how the tech works, or at least how it appears to me.
{
"lightbox_close": "Close",
"lightbox_next": "Next",
"lightbox_previous": "Previous",
"lightbox_error": "The requested content cannot be loaded. Please try again later.",
"lightbox_start_slideshow": "Start slideshow",
"lightbox_stop_slideshow": "Stop slideshow",
"lightbox_full_screen": "Full screen",
"lightbox_thumbnails": "Thumbnails",
"lightbox_download": "Download",
"lightbox_share": "Share",
"lightbox_zoom": "Zoom",
"lightbox_new_window": "New window",
"lightbox_toggle_sidebar": "Toggle sidebar"
}
Very informative, thanks.
cool info, thanks for your effort.
but does this mean we may be able to improve drivers for a multi touch on our h/d or on other resisitive screens?
This whole lack-of-multi-touch-on-the-HD thing got me thinking. One problem is obviously the gaming side of things, where you can't push two virtual buttons at once to go at a diagonal say.
If you open up an SMS and use the Transcriber option, then push at two points on the screen, there is a line drawn that stops mid-way between the two pushed points.
Would it not be possible therefore, for a cleverer person than I, to program a virtual D-Pad that could work out when two buttons were pushed simply by triangulating the position of the virtual cursor?
Does that make sense?
That's an interesting point, Major.
Try this also, in the notes app - turn it to drawing mode. Then hold one finger on the screen in one position, and then tap the screen or drag around the screen with your other finger. You see how the touchscreen makes lines that fan out between the two points of contact.
i believe that it may be possible to somehow utilize the fact that something does happen when pushing in on two points of a resistive screen.
the real problem with it is that when you push on two seperate points, the screen makes an indentation that spans across the area between the two. the layer you are pushing on is simply not elastic enough to register two smaller seperate points or indents. its just read as one big indent on the layer between the two points of pressure. this is why the the screen only reads a point between the two. the sensor is just picking up one giant blob that centers at the mid point between your two fingers. this wouldnt really help much because the farther apart your fingers are the more random the point location becomes---this is because the thin layer will be contacting the layer below at multiple random spots between your two fingers. idk. it seems like it wasnt meant for multi touch. and a driver would need to have a more precise input than that.
Ahh, but consider a simple d-pad arrangement, with 4 direction buttons and maybe 2 action buttons (à la SNES or similar?). Diagonals would be easy enough with the point of contact appearing between the 2 buttons pushed. A move-and-fire combo could be detected with clever placement of the virtual action buttons giving the device a good clear area of detection.
If I get time today, I'll try to draw a simple graphic. A picture speaks a thousand words and all that.
Edit: OK, excuse my laughable drawing, but you get the idea. With better placement, the buttons could be made quite in-sensitive to where the fingers/thumbs are pushing on them.
My thoughts on the matter.
I think it would do emulator or even virtual d pad developers well to add this option into their d pad. When you press on two points the screen reads in between right? So if you're playing mario and you're trying to run and jump over a hole, just have a setting that allows right button to continue if suddenly the pointer reads in the middle of the screen. In otherwords, it would make no sense for a nintendo player to suddenly press the middle of the screen, so the d pad would read that as a whichever butten first pressed, plus button now being pressed.
If it reads exactly between both points, then the software should be able to guess which button you're pressing by calculating which buttons are exactly in line with eachother. There will be a beginning reference set by the first point pressed, so it is only a matter of auto executing o---^---0. I hope this idea makes some sense to someone other than me.
Just think of Orion's belt and it's symetry to get the right idea of what I'm talking about.
Caid.
Just a bit more clarity, if the d pad developers set up a reference of center screen points, the could technically program their pads to be multitouch souley off the basics that each two buttons pressed together would have a center point at an exact or near exact point in between the left and right side of the d pad. This wouldn't work for diagnal, but for any direction plus ABXYLR there would be a center point between each of them, a place in between them that the phone would register the touch.
Is this exploitable?
Sorry for constantly editing, but ideas keep crystalizing.
Beyond the first idea, to register an A+b, have slide vs tap settings on the d pad. That means, if I slide from a to b, the d pad will hold a and activate b as opposed to just switching between the two. Once again, this should technically be possible with the first idea as the middle reference point should also read a shift. In fact, if you're pressing on both sides of the screen at once (d pad and abxy) then all the actually pressing action would be happening in the center of the screen or rather between both pads.
tazbo28 said:
the real problem with it is that when you push on two seperate points, the screen makes an indentation that spans across the area between the two. the layer you are pushing on is simply not elastic enough to register two smaller seperate points or indents.
Click to expand...
Click to collapse
I seriously doubt that it's technically any more difficult than calculating the two touch positions on a capacitive digitizer (keeping in mind that in reality "multi" means "dual").
it's actually hysterical raisins (historical reasons) that resistive is single touch only - when PDAs were first developed, the designers had to cater for both stylus (very precise) and finger use. Finger use is spread over the fingertip so the designers hard-coded algorithms that would average out the touch coordinates so the middle of the touch events.
The mentality basically went along the line of "why would you want to touch two points on the screen at the same time? Nothing uses that! Can you think of a use for that? No? Ah well, we don't need that then"
This has held until Apple decided to use the more battery draining capacitive method (resistive only uses battery when the two layers touch, capacitive has a constant charge and therefore constant drain) and decided to use multi-touch as a gimmick (and, lets be completely blunt here it IS a gimmick - pinching is NOT easier than running a circle around the point of interest and it's certainly more clumsy than a slider for zooming!) under the pretense of a new and awesome feature for mobiles (it does have practical use in laptop trackpads though)
Resistive technology does support multi-touch, but it's a bigger issue to code the function because the touch events are processed by the touch controller then sent to the operating system as x/y coordinates. The operating system can't do anything about this. To get resistive technology to support multi-touch you'd have to re-code the controllers...
Major_Sarcasm and Caid444, I agree 100%. If I just had the time, given that I can develop applications using .Net Framework, I should be able to replicate this in a simple application. Will try and get back to you... I'll start with a Win XP app, just the concept, and then I will deal with Compact Framework (if it's still called 'Compact'). Making drivers to work in any application is another story, though.
tazbo28 said:
Ok so people have been confused over the tech thats behind the Iphone/G1 capacitive dual touch screen, and all WinMo phones resisitive screens. So i thought i would do the great job of researching the tech behind both methods. 10 minutes later i came across this info on how both methods work, as well as a more expensive and even better method using acoustic mumbo jumbo. So for all you out there that are too lazy to do the diggin, heres the info that once and for all puts the rumors to bed about "...i heard its dual touch capable from so and so website..."
PLEASE READ>>>
The resistive system consists of a normal glass panel that is covered with a conductive and a resistive metallic layer. These two layers are held apart by spacers, and a scratch-resistant layer is placed on top of the whole setup. An electrical current runs through the two layers while the monitor is operational. When a user touches the screen, the two layers make contact in that exact spot. The change in the electrical field is noted and the coordinates of the point of contact are calculated by the computer. Once the coordinates are known, a special driver translates the touch into something that the operating system can understand, much as a computer mouse driver translates a mouse's movements into a click or a drag.
In the capacitive system, a layer that stores electrical charge is placed on the glass panel of the monitor. When a user touches the monitor with his or her finger, some of the charge is transferred to the user, so the charge on the capacitive layer decreases. This decrease is measured in circuits located at each corner of the monitor. The computer calculates, from the relative differences in charge at each corner, exactly where the touch event took place and then relays that information to the touch-screen driver software. One advantage that the capacitive system has over the resistive system is that it transmits almost 90 percent of the light from the monitor, whereas the resistive system only transmits about 75 percent. This gives the capacitive system a much clearer picture than the resistive system.
On the monitor of a surface acoustic wave system, two transducers (one receiving and one sending) are placed along the x and y axes of the monitor's glass plate. Also placed on the glass are reflectors -- they reflect an electrical signal sent from one transducer to the other. The receiving transducer is able to tell if the wave has been disturbed by a touch event at any instant, and can locate it accordingly. The wave setup has no metallic layers on the screen, allowing for 100-percent light throughput and perfect image clarity. This makes the surface acoustic wave system best for displaying detailed graphics (both other systems have significant degradation in clarity).
Another area in which the systems differ is in which stimuli will register as a touch event. A resistive system registers a touch as long as the two layers make contact, which means that it doesn't matter if you touch it with your finger or a rubber ball. A capacitive system, on the other hand, must have a conductive input, usually your finger, in order to register a touch. The surface acoustic wave system works much like the resistive system, allowing a touch with almost any object -- except hard and small objects like a pen tip.
As far as price, the resistive system is the cheapest; its clarity is the lowest of the three, and its layers can be damaged by sharp objects. The surface acoustic wave setup is usually the most expensive.
How Stuff Works people, its a great website. heres a picture i made to show you how the tech works, or at least how it appears to me.
Click to expand...
Click to collapse
Thanks tazbo28
Very usefull info here!
Here what the guys working on android have been able to produce with a capacitive screen:
http://gizmodo.com/5097166/android-multitouch-proof-of-concept-aint-pretty-but-it-works
Multi-touch that works "not-so-well-but-at-least-it-does"
tazbo28 said:
the real problem with it is that when you push on two seperate points, the screen makes an indentation that spans across the area between the two. the layer you are pushing on is simply not elastic enough to register two smaller seperate points or indents.
Click to expand...
Click to collapse
Your source?
I don't think so...
Just a little comment about the two points creating a crease down the screen. The distance between the metalic layer and the plastic screen is minimal at best, not to mention the material on the top layer is very flexible. I don't think it's likely at all that the screen would have a crease or ridge between the two points, there's just not enough indent to create this scenario.
One more thing, I'm not 100 percent sure, but I get the impression that the space between the two layers isn't actually filled with air but in fact some sort of non conductive fluid or gel. It would help the two laters to retain their space and elasticity from eachother.
But what do I know without poking holes in my screen (and I'm not about to do that.)
Caid.
just got this from wikipedia
Resistive touchscreens are composed of two flexible sheets coated with a resistive material and separated by an air gap or microdots. When contact is made to the surface of the touchscreen, the two sheets are pressed together, registering the precise location of the touch.
For example, during operation of a four-wire touchscreen, a uniform, unidirectional voltage gradient is applied to the first sheet. When the two sheets are pressed together, the second sheet measures the voltage as distance along the first sheet, providing the X coordinate. When this contact coordinate has been acquired, the uniform voltage gradient is applied to the second sheet to ascertain the Y coordinate. This operation occurs instantaneously, registering the exact touch location as contact is made.
Resistive touchscreens typically have high resolution (4096 x 4096 DPI or higher), providing accurate touch control. Because the touchscreen responds to pressure on its surface, contact can be made with a finger or any other pointing device.
the part about "microdots" is true i can see the grid of super tiny dots on my 8125 touch screen
Related
Am I the only one that seems to be having a major issue with this? It comes up more with the added on screen keyboard in 1.5, it seems.
The corners of the G1 screen do not respond to touch at all, and I find myself always having trouble hitting the keys on the corners of the virtual keyboard. Applications that require you to touch any corner of the screen with precision (ie tic-tac-toe) are annoying as well because it's difficult to get it to work.
Is there any fix in sight or is it a hardware limitation?
definitely a hardware limitation.
hardware limitation means: normal hardware limitation or limited due failure????
I only experience this problem with the right edge of my phone. The top, left, and bottom don't have any problems.
Touch also seems to be less sensitive when the phone is rested flat.
My phone has the sensitivity problem on the edges too (mostly on the right and top sides).
For example, it's almost impossible to drag an icon to the left screen, because it doesn't seem to reach the threshold area.
The protective screen (boxwave anti-glare) made the problem even worse, but since it's much more comfortable than the glass, I won't be taking it off.
The easiest way to test all this is to use the whiteboard application.
So, I ask the developers: Please don't place the buttons on the edges of the screen!
The worst apps in this regard I've seen so far were the Video player (don't remember its exact name, its seek slider was unusable), and the sudoku with the number chooser in the bottom. OpenHome's app slider has the same problem.
is the screen heat or pressure sensitive, and how is the multitouch?
Neither. And so far I can't complain about the screen one bit. I do find myself simply tapping the screen to auto-size the browser, been easier than using two hands.
Its heat although the proper name is Capacitive meaning is only works with fingertips or a specialized " iphone " stylus, which i have and dont work very well tbh although mine was a cheapy off ebay (like 60cent for 2)
i just tried one at the store and noticed that it was not as sensitive as the iphone's. if my finger was on the hero light enough there was no response, but no matter how light i try to go on the iphone it always responds.
It is not heat.
Wiki:
"A capacitive touchscreen panel consists of an insulator such as glasshttp://en.wikipedia.org/wiki/Glass, coated with a transparent conductor such as indium tin oxidehttp://en.wikipedia.org/wiki/Indium_tin_oxide (ITO). As the human body is also a conductor, touching the surface of the screen results in a distortion of the local electrostatic field, measurable as a change in capacitance."
And who cares if it's not as sensitive as the iphone? When would you ever try to press a button as lightly as possible in real life? I find it better that you actually need to touch the screen for a button press, leads to fewer mistakes with random light button brushing. I have yet to find a time when the screen doesn't respond when I want it to.
Feels just as sensitive to me. I very lightly ran my finger over the screen and it responded just as well.
could anyone share any experiences for a good working stylus for the hd2?
this phone is stupendous but alas i have to use it for work and the software we use requires very tiny touches and signature signing.
have tried the ones on e-bay (rubber topped pencil turned upside down)
which is crap...have now tried the htc and dagi ones with the tiny red dot and they are just not cutting it.
has anyone done any homemade experiments or come across a moe "pointy" stylus
toodle pip amigos
I got some, and they were crap...
davidcampbell said:
I got some, and they were crap...
Click to expand...
Click to collapse
Unfortunately you are all too correct!
With the glass screen!
Put a protector on it and they work!
I have been using the dagi long slim one and it works ok, I do have a screen protector on though so maybe this helps... makes selecting text and writing Chinese characters much easier (I use my hd2 to study/translate Chinese so being able to use a stylus to do this is pretty great)... its a bit fiddly to start with as you need to learn how to use it properly, but once you get used to it everything is ok...
Most of the styli are crap, I need something with more of a point, but that's probably not possible. Still waiting for some sort of battery powered pen that offers pin point precision, a guy can dream right...
i'll try to put a screen protector on as suggested and try...
my problem is it's no good mastering the dagi myself as i have to get customers signatures and most people simply can't use it...
it deffo needs something more pointy.
i'm sure i read somewhere on the forum about a guy who was filing something to a point till it stopped working???
i'm so frustrated as this is the only phone in 8 years that uses my works software flawlessly...except stylii
I have ordered the official htc stylus 400.
arrives in 2 weeks and i'll post comments.
read a review after ordering that it actually scratches the screen!
I guess i should have waited for the 400/2 version!
look on the bright side, will probably misplace it by the second day anyway
I've tried using my DS stylus but that doesnt work, seems like it was made for bigger objects to come in contact with the screen (ie fingers).
ILLEATYAFACE said:
I've tried using my DS stylus but that doesnt work, seems like it was made for bigger objects to come in contact with the screen (ie fingers).
Click to expand...
Click to collapse
It`s not about the size of objects. HD2 has capacitive screen.
"A capacitive touchscreen panel consists of an insulator such as glass, coated with a transparent conductor such as indium tin oxide (ITO).[2][3] As the human body is also a conductor, touching the surface of the screen results in a distortion of the body's electrostatic field, measurable as a change in capacitance. Different technologies may be used to determine the location of the touch. The location can be passed to a computer running a software application which will calculate how the user's touch relates to the computer software."
Wikipedia
thanx for input guys..i least i know about capacitive screens now thanx to kregowski...although i sorta knew.
so if someone could come up with a small pointy thing that gives off some sort of electro staticy stuff...we in..
the search continues
Got the dagi (10 euro from Hong Kong). I am not amused... Touching the screen with the complete tip flat on the screen is annoying to get used to. Often only the side of the tip touches the screen.
When the tip is flat on the screen (which it should) it just does not feel OK. Not "delicate".
Which the screen responsiveness lowered and the tap-reaction time increased (which I did just to be able to get some decent fingertyping results) the dagi does not function adequate. I have to aim more accurate then with my fingers and have to press the dagi on the screen.
Putting the screensettings to default (which I did not try yet) reinstates all the drawbacks in the other functionalities.
sometimes my phones touchscreen becomes unresponsive when its flat on its back on a table for example. I unlock the screen with the trackpad or power button so that the screen comes on. Then when i try to slide the lockscreen it doesnt do anything. When i begin sliding the bar from the total edge of the screen it sometimes works.
Usually i just have to pick up the phone to get the touchscreen working again. actually, when i even lift up the phone the slightest bit, it works again.
Weird right?
anyone else experiences this issue ?
I assumed it was designed to work like this. Makes me move my phone rather than just slide to snooze the alarm then go back to sleep.
Same here. I think it has to do with the capacative toucscreen.
Sent from my HTC Vision using XDA App
wileykat said:
I assumed it was designed to work like this. Makes me move my phone rather than just slide to snooze the alarm then go back to sleep.
Click to expand...
Click to collapse
you're saying you move your phone to snooze the alarm? didnt know it could do that, gonna try that out tomorrow morning
forgot to mention the issue is not ROM specific. I have witnessed it on virtuous, CM, MIUI, stock
Ive had this problem as well. Slightly annoying!
I don't think its a design issue..
you don't have to lift your phone to get the screen to respond.
notice that it is enough just to touch the body of the phone to make the screen responsive again..
for me its very annoying, especially not being able to snooze the alarm in the morning
Let me point out that this happens with my G1 also so its not just the G2.
Yeah, there are a few discussions about this on the forums already. It's most likely a capacitance issue. When your phone is just sitting on a table and you touch the display, you may not create enough capacitance with the touchscreen for it to register as a touch (as opposed to resistive touchscreens which needed only physical contact but were less accurate as a result). Touching the body of the phone basically creates a common ground or voltage reference point for the phone, allowing your finger to create the needed amount of capacitance (in technical terms, no longer a floating ground). Whenever I touch the screen while the phone is lying on a table, I just always make sure to rest either my thumb or ring finger on one of the edges.
A lot of factors will also come into play---whether or not your phone is in a case, type of material the phone is resting on, the level of humidity in the air, even potentially the type of shoes you are wearing. This is why sometimes it works and sometimes it doesn't.
ianmcquinn said:
(...) Touching the body of the phone basically creates a common ground or voltage reference point for the phone, allowing your finger to create the needed amount of capacitance (in technical terms, no longer a floating ground). (...)
Click to expand...
Click to collapse
yes! I just figured this out myself. merely touching the body of the phone makes the screen respond to my fingers.
well in that case its not really a big problem. from now on i just touch teh body with my middle finger while sliding with my index finger =D
liory said:
notice that it is enough just to touch the body of the phone to make the screen responsive again..
for me its very annoying, especially not being able to snooze the alarm in the morning
Click to expand...
Click to collapse
Will give this a go tomorrow, I hope it doesn't work!!!.... I'm gonna be forever snoozing with 2 fingers now!!!
This seems to be temperature related (and it doesn't happen on a mytouch 3g)
I live in Minnesota and I deliver pizza. When I would need to call a customer to get into an apartment building, I would set the phone on top of the pizza bag and dial with one hand.
When I first got my G2 (early november) I had no problems doing this, then one day it wouldn't respond at all to touch. I still had my old mytouch 3g with me because the old number wasn't ported yet, and I was able to use that to make the call. It wasn't until later that I realized that the problem was caused by the cold and that I needed to touch the metal part to make it work, so I thought that the touchscreen on my 2 week old G2 had just died.
At room temperature the touch screen always works (I can silence the alarm while it's sitting on the table), but temperatures below 50F cause problems
I started noticing this as well. Sometimes I'll hit the power button to turn my phone on from idle-state and then I'll drag the unlock button with to no avail! But then again, I do use my thumb to do it so maybe the capacitance by using the thumb isn't good enough to unlock it.
I use my phone at work a lot so in order to text while being discreet, I usually have the phone in my right hand and text / unlock the phone using my thumb. I doubt it is a defect but it had me worried too. Guess I'm not the only one experiencing this. (Although I must say, dragging my thumb to swype and using the side of my thumb probably is the reason why it sometimes doesn't work)
Back when I had my g1 it did the same thing and experience this on my g2....if its laying down its like it thinks its just on in your pocket so the touch seems limited
Sent from my T-Mobile G2 using XDA App
Elegy said:
sometimes my phones touchscreen becomes unresponsive when its flat on its back on a table for example. I unlock the screen with the trackpad or power button so that the screen comes on. Then when i try to slide the lockscreen it doesnt do anything. When i begin sliding the bar from the total edge of the screen it sometimes works.
Usually i just have to pick up the phone to get the touchscreen working again. actually, when i even lift up the phone the slightest bit, it works again.
Weird right?
anyone else experiences this issue ?
Click to expand...
Click to collapse
Not weird. TOTALLY normal and expected.
It is because it has a CAPACITIVE TOUCHSCREEN.
This means that when you touch the screen, your finger is actually COMPLETING AN ELECTRIC CIRCUIT.
When the phone is lying on a table and you touch it, you are NOT completing the circuit since you are NOT touching the phone's body.
When the sensor is on, there is an electrical difference between the surface of the screen and the body of the phone. When you touch BOTH, you alter the electric field in that localized area of the screen. The sensor detects the change in voltage and determines the location of the contact.
dhkr123 said:
Not weird. TOTALLY normal and expected.
It is because it has a CAPACITIVE TOUCHSCREEN.
This means that when you touch the screen, your finger is actually COMPLETING AN ELECTRIC CIRCUIT.
When the phone is lying on a table and you touch it, you are NOT completing the circuit since you are NOT touching the phone's body.
When the sensor is on, there is an electrical difference between the surface of the screen and the body of the phone. When you touch BOTH, you alter the electric field in that localized area of the screen. The sensor detects the change in voltage and determines the location of the contact.
Click to expand...
Click to collapse
So how do I adjust the sensitivity? As I understand it I could increase the voltage to the screen making it easier to complete the circuit...such as through thin gloves.....
Sent from my HTC Vision using XDA App
dhkr123 said:
Not weird. TOTALLY normal and expected.
It is because it has a CAPACITIVE TOUCHSCREEN.
This means that when you touch the screen, your finger is actually COMPLETING AN ELECTRIC CIRCUIT.
When the phone is lying on a table and you touch it, you are NOT completing the circuit since you are NOT touching the phone's body.
When the sensor is on, there is an electrical difference between the surface of the screen and the body of the phone. When you touch BOTH, you alter the electric field in that localized area of the screen. The sensor detects the change in voltage and determines the location of the contact.
Click to expand...
Click to collapse
This is mostly true.
Put the phone on the table and dont touch it and try to swipe to unlock.
No touch a finger on the phone body and do it again, itll work.
Doesn't seem to always fit that behavior, but more often than not.
TheNewGuy said:
So how do I adjust the sensitivity? As I understand it I could increase the voltage to the screen making it easier to complete the circuit...such as through thin gloves.....
Click to expand...
Click to collapse
You don't. You should NOT even if you could. It would not work as you expect since it is NOT a matter of sensitivity, but of CONDUCTIVITY.
klmsu19 said:
This is mostly true.
Put the phone on the table and dont touch it and try to swipe to unlock.
No touch a finger on the phone body and do it again, itll work.
Doesn't seem to always fit that behavior, but more often than not.
Click to expand...
Click to collapse
It is not "mostly" true, it is PRECISELY true.
dhkr123 said:
It is not "mostly" true, it is PRECISELY true.
Click to expand...
Click to collapse
No reason for the condescending tone in every post (or if it's not your intention to be condescending, you may want to restrain your constant use of caps). You had stated:
dhkr123 said:
This means that when you touch the screen, your finger is actually COMPLETING AN ELECTRIC CIRCUIT.
When the phone is lying on a table and you touch it, you are NOT completing the circuit since you are NOT touching the phone's body.
Click to expand...
Click to collapse
Someone was just pointing out that if this was "precisely" true as you put it, then the screen should never be responsive if you aren't touching the phone's body (since the circuit is, based on your reasoning, still broken). This isn't the case though since it does work sometimes. How could your statement be precisely true if it's known to be false sometimes (hence the "mostly true" comment)?
And it's actually not true. You are not completing an electric circuit. No current flows from your finger into the screen and no components within the phone are electrically coupled because of your touch. As you stated, you are altering the electric field when you touch the screen. This results in a change in capacitance in that area. Note, this change occurs even when you are not touching the phone's body, just to a different degree. Touching the phone will typically increase the amount of capacitance as opposed to when you are not though (since you are basically "grounding" yourself to the phone). The more capacitance there is, the easier it is to measure. If there is too little capacitance, the touchscreen's microcontroller will not register the change at all. Even if it does register the change, there will be additional filtering that occurs to that raw data either in the microcontroller itself or in software (or both).
This is why the screen works sometimes when you touch it without touching the rest of the phone. You are still causing a change in capacitance. The amount of change depends on numerous factors (this is what I stated in an earlier post). These factors effect how grounded both you and the phone are relative to earth ground (in the electrical sense, not geographical). Typically speaking though, if you are relatively close to earth ground, then you will be able to use the screen without touching the body of the phone. For example, someone stated earlier that the screen doesn't work when it's cold. Cold weather results in lower humidity/drier air and you being less grounded (basically why people see an increase in static electricity as well). Chances are that the touchscreen would work again without the touching the phone's body if you had physically been touching a large piece of metal with your other hand (to help ground you).
So in response to the other question posed, yes, technically by raising the voltage level to the touchscreen, it may result in a more sensitive screen since there will be a larger increase in capacitance when you touch it. It may also reduce the life of the touchscreen though, if not damage it completely, so not a good idea. It would be better to figure out the filtering mechanism if possible and adjust that in software. Not sure if any devs have looked into it (though my guess is that if it was simple, it would have been done already).
I am so happy I found this thread..
Lately I thought my touchscreen was broken because I was unable to snooze my alarm in the morning (just as some of you seem to have been trying to do).
I think it's temperature related as suggested by the person who mentioned pizza delivery.. and I think this because it only happens on mornings where I've left the bedroom window open and it's very cool. My phone is often on a cold wooden surface. however even picking up the phone doesn't seem to help (I'll have to pay more attention next time) but after a few moments of struggling it eventually works (probably warms up in my hand and gets a better circuit).
I can also vouch that the screen works great with just a single finger on a flat surface, provided it's room temperature. I don't need to touch the body at all, and I do this a lot without issues (aside from cold mornings waking up to my alarm).
im not sure if this is a widespread issue most most of us. as we are all using SHIELD for different forms of entertainment. but i have noticed that with running just about any game designed for digital input, the D-pad is almost useless.
let me explain:
when playing say, Super Mario World. 90% of the time you're always holding "right" on the D-pad because thats the direction of game play. but for some reason you have to REALLY try to only push on the right input of the D-pad. reason being the sensitivity of the D-pad is so strong the if you ever so slightly start to push "down" even in this slightest, Mario will duck and stop running... its sooo f&%$*ng annoying!
this problem is not limited to this game alone, obviously. but to test to see if i was over reacting i popped in the SNES cart in my SNES console (along with many other titles) and had no issues whatsoever with controls on my SNES controller.
then even to run more "tests" to see if i was going crazy i had my brother (also a "old school" gamer) try out my his favorite game Contra. for test purity i never mentioned my thoughts on the D-pad to him, but after 3min of playing he looked at me a said "is there any way you can adjust the sensitivity of the D-pad? on this thing?"
i even had my Mom try out her fav, Tetris and her thoughts were similar... input errors....
so i ether have a unit that has some sort of defect and im completely alone here on this forum? or maybe there are a few more of you out there with this same problem.
well i might just maybe have a solution...
here is the PCB for the shield:
here is the PCB of a SNES controller: (my favorite controller )
here is the mod im planing:
so for clarification, the RED is where i plan to paint over or shave down the conductive contact materal on the PCB. this should in theory make the contact area act more like the D-pad us old school gamers are familiar with. so instead of the contacts kind of "weaving" its more like the 1/2 circles contact patch seen on the button side of the PCB.
i feel this may be a pretty easy mod for most of us and there is already a handy teardown via ifixit:
http://www.ifixit.com/Teardown/Nvidia+Shield+Teardown/16212/1
thought?
suggestions?
Sounds interesting.
I'd suggest starting with something non-permanent like cutting a piece of tape to size and using that first to make sure it does what you expect.
Also take a look at the d-pad contacts and make sure that those are large enough to still complete the circuit.
I don't actually own a shield but had the same issue with the xbox 360 controller and fighting games. Let us know if it actually works.
masterfud said:
Sounds interesting.
I'd suggest starting with something non-permanent like cutting a piece of tape to size and using that first to make sure it does what you expect.
Also take a look at the d-pad contacts and make sure that those are large enough to still complete the circuit.
I don't actually own a shield but had the same issue with the xbox 360 controller and fighting games. Let us know if it actually works.
Click to expand...
Click to collapse
good points. the D-pad does move (or skate?) from side to side about 1/8-1/4" inside the housing. so id need to make sure the D-pad contact pad will still hit both sides of the PCB contacts at the furthest positions to complete the circuit.
i was thinking the paint might be better for the short term test, as tape might "smudge" out of the way during a heated game like streetfighter?
if the idea works id prob scrape away the unwanted conductive material (the lines marked in red) with a hobby knife for a more durable and permanent solution.
only thing im worried about is the pressure needed for circuit closing will change from having the contacts painted (raised up surface) vs taking down the unwanted material (lowered surface) when referenced to the actual contact material we are actually trying to hit. not sure if that is easy to explain without pictures.
Don't do it. Electronics hobbyist here.
Those buttons are a digital 0 or 1. No sensitivity there. The rubber mat underneath the keys simply contacts against both sets of PCB contacts to connect them. No resistance or capacitance readings occur, its either connected or not connected. Shaving the contacts probably wont change that very much. You want to inhibit the contact between PCB and rubber mat from occuring without increased force which your proposed hack wont do to a great extent.
The problem would be purely mechanical in nature, ie the exact way the buttons being mounted perhaps allowing for unintended wobble and as a result perhaps if you shift your thumb forwards and back a bit the rubber mats for the up and down arrows may well come into contact.
A better solution would be to adjust the height between the PCB and the rubber mat, not a simple task. Perhaps the underside of the buttons have some sort of plastic nub which if filed down a tiny amount might work and require a harder press to force the mat down, not sure. Not reversible though.
Or a very thing piece of sponge with 4 holes cut aligning with the PCB contact area (but ever so slightly smaller) would cause a greater force to be required on the buttons while still allowing the buttons to contact. I would sketch what I mean but my skills with drawing on a PC are non existent and I have neither a scanner or camera right now.
I would actually lean towards experimenting with the sponge method, actually inspired by me working my way through a tin of biscuits while playing Xbox one morning and getting crumbs in the controller, started to require alot more force on the A button to get it to work. Open the controller and biscuit crumbs were everywhere preventing the buttons moving properly. Sponge should have the same effect but reversibly and cheaply. Namely, take a kitchen sponge and cut it up a bit, 2mm thick sections perhaps. Kitchen sponges are cheap anyway, pretty sure I saw a 10 pack in a £1 store once, experiment to your hearts desire.
---------- Post added at 07:43 PM ---------- Previous post was at 07:39 PM ----------
Come to think of it. If you put paint on the area in between the contacts then that may work.
Your not going to be able to reliably fix this without altering the feel of the buttons sadly. I think they should have gone for tactile switches on the DPad, they are generally perfect for this.
It is the DPad skate which I believe is causing this.
less permanent if you want to try it is white glue spread with a toothpick. you can wipe that off carefully with a bit of alcohol or water.
SixSixSevenSeven said:
Don't do it. Electronics hobbyist here.
Those buttons are a digital 0 or 1. No sensitivity there. The rubber mat underneath the keys simply contacts against both sets of PCB contacts to connect them. No resistance or capacitance readings occur, its either connected or not connected. Shaving the contacts probably wont change that very much. You want to inhibit the contact between PCB and rubber mat from occuring without increased force which your proposed hack wont do to a great extent.
The problem would be purely mechanical in nature, ie the exact way the buttons being mounted perhaps allowing for unintended wobble and as a result perhaps if you shift your thumb forwards and back a bit the rubber mats for the up and down arrows may well come into contact.
A better solution would be to adjust the height between the PCB and the rubber mat, not a simple task. Perhaps the underside of the buttons have some sort of plastic nub which if filed down a tiny amount might work and require a harder press to force the mat down, not sure. Not reversible though.
Or a very thing piece of sponge with 4 holes cut aligning with the PCB contact area (but ever so slightly smaller) would cause a greater force to be required on the buttons while still allowing the buttons to contact. I would sketch what I mean but my skills with drawing on a PC are non existent and I have neither a scanner or camera right now.
I would actually lean towards experimenting with the sponge method, actually inspired by me working my way through a tin of biscuits while playing Xbox one morning and getting crumbs in the controller, started to require alot more force on the A button to get it to work. Open the controller and biscuit crumbs were everywhere preventing the buttons moving properly. Sponge should have the same effect but reversibly and cheaply. Namely, take a kitchen sponge and cut it up a bit, 2mm thick sections perhaps. Kitchen sponges are cheap anyway, pretty sure I saw a 10 pack in a £1 store once, experiment to your hearts desire.
---------- Post added at 07:43 PM ---------- Previous post was at 07:39 PM ----------
Come to think of it. If you put paint on the area in between the contacts then that may work.
Your not going to be able to reliably fix this without altering the feel of the buttons sadly. I think they should have gone for tactile switches on the DPad, they are generally perfect for this.
Click to expand...
Click to collapse
thank you as well.
i am aware this is a "on / off" situation. but the weaving of the PCB contact material makes for a very light press to be needed for circut closing. (my sensitivity problem)
the "sponge method" is also something i thought about. i wanted to source some other D-pads from other controllers to try this but im not sure how a modded one would hold up to button mashing in the long run. so it would basically have to be almost identical dimensions to work.
ive been trying to source SHILED parts to experiment with. but the guy i was speaking with has family obligations tying him up right now. so...
i do really like the feel of the controller. they did a awesome job. but i just wish the PCB contact layout was identical to the NES/SNES controllers. as in with the a, b, x, y buttons being the more sensitive (weaved) contacts and the D-pad being the less sensitive 1/2 circle style once.
i wish i was just being picky here but i just flat out cant play any retro games in the current state. 15 years of muscle memory just doesn't go away.
SixSixSevenSeven said:
It is the DPad skate which I believe is causing this.
Click to expand...
Click to collapse
i agree. its a lot of movement...
i can center the D-pad and game play is much better for a few seconds till it moves again. in fairness, every D-pad (SNES, SEGA, Etc) does skate a little (except for PSX). the SHIELD is by far the one with the most movement.
for ppl not understanding what we mean by the D-pad "skate" here is a very crappy rez vid:
http://www.youtube.com/watch?v=LYj5-L02fiI&feature=youtu.be&noredirect=1
s0me guy said:
i agree. its a lot of movement...
i can center the D-pad and game play is much better for a few seconds till it moves again. in fairness, every D-pad (SNES, SEGA, Etc) does skate a little (except for PSX). the SHIELD is by far the one with the most movement.
for ppl not understanding what we mean by the D-pad "skate" here is a very crappy rez vid:
http://www.youtube.com/watch?v=LYj5-L02fiI&feature=youtu.be&noredirect=1
Click to expand...
Click to collapse
My PSP has quite a bit of DPad skate. But the centre point of the Dpad does seem to stay centred, it just seems to rotate on the spot if you know what I mean. It skates but it remains perfectly usable.
to stop the skate we could find (make?) rubber grommets to fit around the mounting holes for the screwdowns (yellow circles are the tips). they would act as sleeves or bumpers keeping the plastic D-pad assembly centered.
i have no idea where i would even look for a part like this... McMaster? lol
http://www.mcmaster.com/#rubber-bumper-spacers/=o6yfz8
That centre nub between the 4 screws. They should have made that larger or the hole in the rubber piece smaller so that they match size better. I reckon that the cause of the skate is that nub.
just realized i should probably stop brainstorming using the images from ifixit as a reference.
they could have quite possibly used a different pre production unit and it might be nothing like the final public retail units. i know there were a few different revisions along the way (D-pad included).
guess ill have to try to find time to tear this down over the weekend. i've been hesitating up until this point mostly because our 1mth old baby requires a lot of attention ( who would have thought? ).
id just hate to have this thing half taken apart get sidetracked and forget were i was... lol
i can easily see this mod taking up an entire saturday to get 100% working without a newborn.
oh yeah, this D-pad is really really ****ty...
delete..
MUCH BETTER! :victory:
now to fix the skate issue!
wow... it feels much better? so you DID it. nice! :good::laugh:
I want to try in once, too
david419kr said:
wow... it feels much better? so you DID it. nice! :good::laugh:
I want to try in once, too
Click to expand...
Click to collapse
yea now its much more like the other D-pads im used to. (NES, SNES, N64, PSx, PS2, Gameboy, NDSlite)
i swapped the button diaphragms also. as the A B X Y button side Diaphragm has a small hold in the middle of it and it tends to minimise the "skate" issue.
BEFORE
AFTER
BEFORE
AFTER
all in all id say these mods made the D-Pad 100x better, but its still not perfect. i still make input errors all the time. but its nowhere near as bad as it was. the main issue thats still present is that the D-pad is acting like a rocker more than a switch. so when you press dead center on the "right" arrow on the D-pad the diaphragm on the "right" contact is pressed down 100% and the "up" & "down" contacts are pressed 75% down (almost touching but not making contact). but... it you ever so slightly press off of dead center on any arrow the contacts next to the one you're pressing end up touching. then your input becomes a 45deg. input instead of a 90deg one.
i honestly don't know how to fix this without designing a different bottom half of the D-pad assembly. my current idea is to take a D-pad from a Donor SNES controller (its dimensions are pretty close on the bottom side), grind off the "arrow pad" and attach the remaining bottom portion to the SHIELD's "arrow pad". since the SNES bottom part has more of a Dome shape than a flat one like the SHIELD it will have more bias to the direction your inputting and less Bias to the directions on either side of that direction.
descriptive pic coming soon
my main issues is right now everything i do is a MAJOR commitment because i have no access to Extra SHILED parts. if i mess up... well thats it... so...
hopefully the guy i was talking to about getting extra parts will get back to me soon. i also was talking with a guy who has a 3D printer that may be able to print the modded D-pad for us after its complete. :fingers-crossed:
also i might add that i used 2000grit sand paper to rough up the surfaces on the D-pad & buttons. the glossy-ness of them got very slippery after a few minutes of gaming as the finger oils build up.
they have WAY MORE grip now and the matte finish is much more appealing over the Gloss finish. :victory:
this has been already kind of talked about in a few other threads with users complaining of "screws falling out" but i would also would like to stress to anyone disassembling their SHIELD to be very very VERY careful with the threads. it seems about half my screws were slightly over tightened from the factory and i have some slight stripping going on.
the build up of plastic dust on the one screw in this pic is not something you want to see. as the less plastic there is in the threads, means the less "meat" the screw can grab on to. in-turn that means the less torque can be applied to fasten the 2 parts which can lead to movement and then internal ware... i may have to find slightly bigger screws then drill and tap all the holes...
Makes you wonder how this got past QA testing. I also have the same problem with a newly purchased Shield and it makes emulators difficult to use (unless you map stuff to the analog stick)
georaldc said:
Makes you wonder how this got past QA testing. I also have the same problem with a newly purchased Shield and it makes emulators difficult to use (unless you map stuff to the analog stick)
Click to expand...
Click to collapse
emulators were probably not something they were heavily focused on and all the testers probably were more worried about "feel" than function. IMO if you are able to think about the controller during game play, its not made right.
i hate to sound like a whiny brat or something here but if i wanted a sub par experience i would have stuck a $40 Moga controller on my Galaxy S3 or somthing. i mean isn't the point of SHIELD to be a premium product for a Niche market? and isnt the controllers function the most important part of the "function" of SHIELD? sure nice screens and speakers are great and a part of the experience... but what does it matter if you cant play the the thing. you may as well use a GameKlip and android Phone.
its like driving a highend sports car on all season tires. all kinds of awesome engineering and power thats just frustrating to drive cuz you cant keep the thing on the road.