JF1.41 Bug? changes the way the touchscreen works? - G1 Android Development

On my unrooted phone the "Steamy Window" application worked fine
If I pressed a little it would make a little line in the "steam", and if I were to push my thumb down on a large area, it would detect that and make a "thumb" sized print.
Now using JF1.41 no matter what I do it only makes a TINY TINY line, about a pixel or two in size.
Did the modification of the screen change an api that the Steamy Window app is using to work different?

Well, maybe you should go back to an unrooted phone so that you can play with your steamy window....

LPChris, I would expect a little more maturity from a "senior member"
I am trying to contribute to make JF the best it can be. Maybe your just having a bad night or something, but theres no need to be so childish.
:ed:
Looking through your posts it seems like your just having a bad night
Back on topic:
While this may be a stupid little problem with one application. Should finger-area sensitive applications continue to be made this could isolate JF1.41 from the market. Naturally I dont care much for steamy window, but I would like to prevent other issues in more... useful... applications

I've seen some weirdness where my home screen has a jitter. I've also had accuracy issues with clicking menu items and buttons. I'm going to backup my phone and load JF 1.31 back on to see if it is the screen or the img.
Interesting...I just reflashed with 1.41 RC30 and the problem went away. I was using the fade to blue theme .99.

sjbayer3 said:
On my unrooted phone the "Steamy Window" application worked fine
If I pressed a little it would make a little line in the "steam", and if I were to push my thumb down on a large area, it would detect that and make a "thumb" sized print.
Now using JF1.41 no matter what I do it only makes a TINY TINY line, about a pixel or two in size.
Did the modification of the screen change an api that the Steamy Window app is using to work different?
Click to expand...
Click to collapse
Ditto about LPChris lol, a lot of his posts sound like that, but yeah I noticed the same thing about the steamy window app before and after using JF 1.41.

Steamy window, LOL! sorry guys. anyway keep up the great work.

Yes, this is a trivial example, but it does identify a real issue.
I was reading through the various google android groups last night and I ran into some info that seems to addresses this issue. From what I was able to gather, the JF1.x implementation of multi-touch uses a community hack that reassigns the value that normally keeps track of the size of the touch input. This value is now used to keep track of the distance between two touches. (see: http://groups.google.co.uk/group/android-platform/browse_thread/thread/209b512a9fdf7367?hl=en)
your Steamy Window app seems to make use of the size value as it was originally designed. Now with the change, with only one finger input, this value is set to some minimum, producing a thin line.
To test this theory, see if when you touch the screen with two fingers the size of the mark gets bigger.

Why can't a capability bit or other mechanism be defined that could be used to indicate whether or not an application were multi-touch aware, so that the touch handling behavior at the lower levels would revert to "standard" behavior for normal (non-multi-touch) apps? Trying to make the multi-touch hack fully backwards compatible may not be practical.

I tested your idea about multituch possibly fixing steamy window
My only results are really strange.
If I touch with two fingers it will make a line about the width of my 2 fingers, but its located in the top left, no matter where my fingers actually are. Furthermore the line isnt straight its curved.
Now, after my first touch, it seems if I touch it again it just crashes.

ex-MN said:
From what I was able to gather, the JF1.x implementation of multi-touch uses a community hack that reassigns the value that normally keeps track of the size of the touch input. This value is now used to keep track of the distance between two touches. (see: http://groups.google.co.uk/group/android-platform/browse_thread/thread/209b512a9fdf7367?hl=en)
Click to expand...
Click to collapse
Edit: Hmm... the x-y data of the second finger is stored in separate fields. I think the usage of the size field is just as a boolean to tell multitouch capable programs that second finger data is being sent.

themes and JF RC versions. be careful
TheDudeOfLife said:
I've seen some weirdness where my home screen has a jitter. I've also had accuracy issues with clicking menu items and buttons. I'm going to backup my phone and load JF 1.31 back on to see if it is the screen or the img.
Interesting...I just reflashed with 1.41 RC30 and the problem went away. I was using the fade to blue theme .99.
Click to expand...
Click to collapse
it seems as tho themes have to be matched to the rom, a theme for 1,30 may flake out on 1.41, most of the themes I have seen and used so far were marked as to which rom the were intended to theme...
BhanG

bhang said:
it seems as tho themes have to be matched to the rom, a theme for 1,30 may flake out on 1.41, most of the themes I have seen and used so far were marked as to which rom the were intended to theme...
Click to expand...
Click to collapse
After some more testing I have this same issue on 2 1.41 compatible themes (rusty metal and fade to blue) I am going to go back to themeless 1.41 but I think it is more related to the multitouch 'hack' than the theme themselves. I find it hard to believe that a bunch of icons could mess this up. I watches my screen jitter between the left and right screen for 3 minutes today before I just performed a soft reset on my phone.
Note: I also applied each theme from a fresh flash of 1.41...not on top of each other.

this happens with screen crack also, but it doesnt affect me that much until i start not being able to select stuff. i know it may be important its just not to me... yet *knocks on wood*

Yeah Ive noticed it too..
I thought that it was just the app at first.
I might be switching back to .31 then at least til this gets fixed.
Steamy window def is NOT reading my fingers location or size properly.

I've also seen it affect touch paint found in the API Demos. Doesn't even register touch anymore.

It appears multi-touch is premature at this point and to implement multi-touch throughout the OS and not just the browser like everyone seems to think isnt good. It is definitely is throughout the OS and probably will affect any application that uses the second touch value to register touch size.
My other concern is that dev's won't take this hacked multi-touch into account so they will continue to write programs that won't be compatible since its not the "norm"
Taken from and android forum...this is my biggest concern
"Please think about soft keyboards when considering multi-touch APIs,
where the problem isn't detecting the distance between two finger, but
rather is allowing a 2nd finger (or thumb) to come down on a 2nd key
before the first finger (or thumb) has been released. The proposed
API to provide only a centroid and the distance between fingers won't
work at all for this sort of application.
I think you'd ultimately want to package this as additional
GestureDetector subclasses, depending on whether the type of gesture
detection wanted was pinch/zoom or multiple/overlapping taps.
Keyboards (and maybe games) would use one class, browsers another.
Providing the right information from a touch panel to drive either
detector might be challenging, especially if the panel provides
'unpaired' x and y coordinates rather than individual touch locations,
but I believe it could be done. "
JesusFreke,
Any thoughts on this?

Yeah lets hear what JF's got to say.
I originally thought that the multitouch was embedded in the browser, It's too soon to have it throughout the entire OS...
How about 1.41=with multitouch and 1.42=without

I've installed JF1.41 and I have an unusual bug....sometimes the screen behaves like it isn't calibrated..i press the screen on a button and it is pressed another one....have anyone encoutered this problem?

I invite everyone to do the following to find out how ur touch registers on ur screen. Open anycut. Get it if u don't have it.
Click activity and find raw image viewer. Click ok and then click on the shortcut.
Now start moving your finger around and see the results.
Let's all post our results here so we can figure this out.
Post in the same format as me:
ROM: JF1.41
1 finger - it surpisingly followed my finger perfectly. It only registered the very center point of my finger.
2 fingers - it appears that it made the same point as the 1 finger, however now the point was exactly in between my fingers and stayed there as I moved them around.
Ok so now we need ppl with stock rc30 and 1.31 so we can test this out!

You wanna have some real fun? Try it with 3 or 4 fingers.. It goes nuts..

Related

Very Strange Fuze Keyboard Troubles

My keyboard has been lagging and doing other strange things that has got me worried. I did a hard reset on it yesterday and before I even installed any software I was forced to use the on-screen board for a spell as the slide out was un-responsive. I chalked it up to some background process being a hog and moved on. Now I'm finding some consitencies.
On a fresh reboot if I open the notes app in landscape mode and start typing I can see the cursor moving but nothing else till I get about 75% across the screen and then there are some black artifacts that now look like a ton of characters all bunched up in the space of about 3 positions wide. If I keep typing till it does a return then the entire line of text shows up and the process repeats itself on the next line.
If I have a full page of text open in say Total commander's text editor in portrait mode and slide the keyboard out, it will make a pattern of those black artifacts diagonally down the screen toward the left side and it will not correct itself. If I go back to portrait mode all is well again. There is seriously something wrong here.
I can't find any information from searching on the problem but hoping one of the experts here could shed some light on it and let me know if it's time to think about a return. Will try to uncover more if I can.
Many Thanks,
Eidolen
I took some screens of the problem. Both pics on the bottom are the same document but left is with TC's editor and the right is Word. Upper right is word as well. Hopefullt the attachment makes it through, if not I can host and link to it.
If you are running a stock ROM without any customizations (you said you hard reset your device), then chances are you have a defective Fuze. You should exchange it through AT&T's 30 day return policy.
If you're not running a stock ROM, then maybe you have an unstable one. Flash Da_G's clean ROM and see if the issue still occurs. Since that ROM has very little installed by default, the chances of a software conflict are slim.
Please post back here with your status.
I was thinking the same thing but the one thing I can't reproduce is when the keyboard doesn't respond at all. Maybe the graphics and the responsiveness are 2 different things?? I'm wondering if clear-type might have something to do with the graphics but that shouldn't lag the keyboard.
Gonna disable Clear-type now and will report back.
Eidolen
Well if I would have done a search with cleartype I would have found my answer, sorry I didn't figure it out sooner. At least pople can see what cleartype looks like on the Fuze in landscape mode if they were wondering. Apparently there is no fix known other than to turn it off. Hopefully the laggy response clears up as well but not getting my hope up. Thanks for your response.
Eidolen
What an odd thing for ClearType to do in landscape mode... as for laggy response, flashing a clean ROM would give you a good gauge as to how preinstalled apps can affect such lag.
Same thing happens to me but in txt'in sometimes my keyboard dont work at all and i have to restart the device, but this has happen to me only 3 times and the keyboard still have life cuz when i hit the fn key or shift the light comes on indicating that i put them on. as for screen lag i bearly get screen lag, because i did some hack i saw in this thread and it seems to of help alittle
by the way am using the ROMeOS2, i was wondering if i was the only person with this problem but apperently not.
Yep I got the same problem. I installed the NullKb cab on my phone to keep the kb from popping up in Opera. It didn't work. I uninstalled NullKb, It still shows up as selectable under other input options. The lag got progressively worse and even resets will not get the keyboard to work.
I've got the same problem. It's really strange.
I used ROMeOS but now am back to Australian Three official rom.
It seems like, when ClearType is active, I can't see what I am writing with the keyboard in LandScape mode. Even stranger is that when you write an SMS, the keyboard writes fine as there is T9 active.
I really need a fix for this as this has started to frustrate me.
I found having "clear type" activated was my problem as well. I turn it off and my kb is back to usual normal. I have a euro stock unit. Very happy with the rest of it's performance. I can live without cleartype.
eidolen said:
My keyboard has been lagging and doing other strange things that has got me worried. I did a hard reset on it yesterday and before I even installed any software I was forced to use the on-screen board for a spell as the slide out was un-responsive. I chalked it up to some background process being a hog and moved on. Now I'm finding some consitencies.
On a fresh reboot if I open the notes app in landscape mode and start typing I can see the cursor moving but nothing else till I get about 75% across the screen and then there are some black artifacts that now look like a ton of characters all bunched up in the space of about 3 positions wide. If I keep typing till it does a return then the entire line of text shows up and the process repeats itself on the next line.
If I have a full page of text open in say Total commander's text editor in portrait mode and slide the keyboard out, it will make a pattern of those black artifacts diagonally down the screen toward the left side and it will not correct itself. If I go back to portrait mode all is well again. There is seriously something wrong here.
I can't find any information from searching on the problem but hoping one of the experts here could shed some light on it and let me know if it's time to think about a return. Will try to uncover more if I can.
Many Thanks,
Eidolen
Click to expand...
Click to collapse
Hi:
I had the same problems. I fixed it by disabling cleartype in landscape (using Diamond Tweaks), and then installing the Fuze Keyboard fix cab, which I am attaching here for you, in case you want to try it. I'm sorry I don't remember the original post I got it from. After doing these 2 things, everything seems to be working great for me.

Auto rotate to the left AND right

as of now, auto rotate for the JF updates and the droidsans app only allow us to rotate to the left.
is it at all possible to make it so the phone can be rotated to the right as well for us left handers? the blackberry storm has this feature and i thought it was pretty cool but im not sure if the g1 can support it.
I'd like to see something like that, too. Mostly because the phone won't stand on the left side on its own, but it will on the right.
yup, i like this idea
I'd quite like a way to lock (a toggle) the rotation too if at all possible, possibly from the menu, or maybe by pressing the camera button.
Use? When using at an angle such as laying in bed and you really want landscape mode without having the keyboard open. Also when you lay the g1 flat on a desk it can switch to portrait by accident.
darkflare said:
yup, i like this idea
I'd quite like a way to lock (a toggle) the rotation too if at all possible, possibly from the menu, or maybe by pressing the camera button.
Use? When using at an angle such as laying in bed and you really want landscape mode without having the keyboard open. Also when you lay the g1 flat on a desk it can switch to portrait by accident.
Click to expand...
Click to collapse
that is an excellent suggestion!
erasedgod said:
I'd like to see something like that, too. Mostly because the phone won't stand on the left side on its own, but it will on the right.
Click to expand...
Click to collapse
lol thanks for pointing this out. what a pos
yeah this would be a nice adition, but what i would like more is a keyboard over ride that forces the screen to landscape mode when he keyboard is open.
i would LOVE to see this happining on the G1 it would awesome !
i would like JF to see this and say what he thinks, and if its possible to do it, please do it =D
IIRC this isn't possible. When they originally unlocked the auto rotate feature it was stated that the os only had coordinates for 0 and 90 degrees. They would have to make extensive edits to the source to add 180 and 270 degrees for orientation. I know, that's a bummer dude.
Well I just found an app that rotates both ways... It's a game called "Wordzor Lite"
I'm pretty sure this shows that it CAN work but i have no idea how hard it would be to incorporate it into the home screen and such...
left-handed people are evil!
golferguy17 said:
Well I just found an app that rotates both ways... It's a game called "Wordzor Lite"
I'm pretty sure this shows that it CAN work but i have no idea how hard it would be to incorporate it into the home screen and such...
Click to expand...
Click to collapse
No one is saying that it's not technically possible, we're just saying that the current state of autorotate is easy because Android is already set up to rotate when you open the keyboard, whereas there is nothing already in place for rotating the other direction.
erasedgod said:
I'd like to see something like that, too. Mostly because the phone won't stand on the left side on its own, but it will on the right.
Click to expand...
Click to collapse
hey, you have made my day! i never realized that..
Hey guys. Not that I would ever want to take anything away from the amazing work of people like JesusFreke, and as someone who has never posted before...
Just to say that I am working on a slightly more stable auto-rotate patch for the kernel with exactly that feature of auto-rotate left. My intention is to make this selectable in three ways.
Off
Right only
Left only
Both
milestone.it said:
Hey guys. Not that I would ever want to take anything away from the amazing work of people like JesusFreke, and as someone who has never posted before...
Just to say that I am working on a slightly more stable auto-rotate patch for the kernel with exactly that feature of auto-rotate left. My intention is to make this selectable in three ways.
Off
Right only
Left only
Both
Click to expand...
Click to collapse
keep us updated
i know of several apps on the marketplace that already have this feature but would it be much different to get the whole OS to rotate?
...i'm guessing it would lol
Actually, its not so different. The bigest problem is stability. Trying to patch in a deliberate action for autorotate feature. Like people have said elsewhere it can be a little bit too sensitive. Should have first version available next week (unless I manage to brick it!)
Anything people want to chip in about autorotate whilst I'm patching it? I am not the worlds greatest programmer but I can usually get there.
I was just reading this thread and what came to mind was the Bubble app on the market. It does this but of course this is on a much smaller scale and not across the entire OS of the phone.
But the nice thing is that you can turn the phone end over end and the bubble rotates with it. Not sure if the Dev for that app is on here, or if you wanted to email him to see if he could provide a bit of insight to possibly assist.
right. i am still working hard on this. any suggestions as to what deliberate action you would like to force rotate? I have thought about an up/down flick then turn the direction you want.
the only problem with patching four-directional autorotate is that it creates interesting problems of when it is perfectly still, it still manages to rotate itself. ie, i left it still on the desk and it was happily rotating itself.
any suggestions from coders. also, is there a quicker way to compile than to recompile the whole stack? ie selective compiling?
cheers.
milestone.it said:
right. i am still working hard on this. any suggestions as to what deliberate action you would like to force rotate? I have thought about an up/down flick then turn the direction you want.
the only problem with patching four-directional autorotate is that it creates interesting problems of when it is perfectly still, it still manages to rotate itself. ie, i left it still on the desk and it was happily rotating itself.
any suggestions from coders. also, is there a quicker way to compile than to recompile the whole stack? ie selective compiling?
cheers.
Click to expand...
Click to collapse
instead of having a force rotate, can you implement a Lock Rotate instead?
That way you can rotate to the desired position, and lock it.
darkflare said:
instead of having a force rotate, can you implement a Lock Rotate instead?
That way you can rotate to the desired position, and lock it.
Click to expand...
Click to collapse
or i could just do that! that would actually be a lot easier to code
I thought it might ~_^
How is the patch applied? if you need a tester give me a pm as i'd be interested in testing and providing feedback
~DF

[REQ] Have all software support screen rotation

This is mainly directed towards programs developed right here, by fellow members (I don't see this problem in professional programs).
Of the countless programs from here that I've ran on my phone, just about all of them would screw up/not have any particular handling for when my screen rotates. Generally, the bottom half of the program disappears to some imaginary area off screen, where my keyboard would be. There won't even be any way to scroll down to see what got cut off. Also, as the program is left-justified, the right side of the screen is blank and unused. It looks wholly unprofessional and sometimes a perfectly suitable solution is just to make the program not rotate! I can't imagine such a thing being very hard to implement, yet I very rarely see any program make any effort to deal with this issue.
Honestly I don't understand why I even have to request something like this. I consider it common sense, but I guess not.. >_>
I feel like I'm going to get flamed for this but I hope for the better.

Touch Capacitive Panel (bottom pad)

Hello to all, am going to ask if anybody of you have a problem in TPs Touch Capacitive Pad? I used this app
Nav Debug Tool by HTC
http://forum.xda-developers.com/showthread.php?p=2511016
To check out this multi touch stuff in my TP (though I know before that it has a multi touch feature, I just want to check if it's working properly) but it seems that the pad has a problem. The problem was sometimes, it can register my touch and most of the time, not.. sometimes it can register the multi touch, and most of the time, not .. even the nav wheel has a problem which doesn't register the rotation most of the time. Is this a common problem in TPs or even Diamond??
I even use the signed os WM6.1 but still has a problem.. so I wonder if any of you guys having the same..
thanks!
It's been happening to me for about a year already as well. Seems like it's just defective. It resets thought after sleeping the phone and waking it. I'm not sure if it was because I used to use Sensor scroll or it's a little scratched up.
I hear the d-pad was a good conceptual idea but that is about it... I personally don't use the d-pad at all because of this, much less any of the buttons (to include the power button). On the other hand, from what I can tell about the d-pad "sensor" from using the NDT(Nav debug tool), it does seem pick up points but it seems to not like repetitive touches. It almost seems like the sensor overloads the processor to the point that it fails in terms of keeping up with the actual location that is being touched; if that makes any sense?
R^7Z said:
I hear the d-pad was a good conceptual idea but that is about it... I personally don't use the d-pad at all because of this, much less any of the buttons (to include the power button). On the other hand, from what I can tell about the d-pad "sensor" from using the NDT(Nav debug tool), it does seem pick up points but it seems to not like repetitive touches. It almost seems like the sensor overloads the processor to the point that it fails in terms of keeping up with the actual location that is being touched; if that makes any sense?
Click to expand...
Click to collapse
that made sense but why would HTC permit a buggy front pad???
You know, it seems many phone brands/cooperations come under fire due to lack of something or another; maybe proof of concept? Perhaps the debug tool is to blame? Or even the sensor degradation over time? One can only speculate so far before it becomes a volatile thought... Again, I point out the idea being good, conceptually. So, I don't know where that exactly puts us as being end users? Then again, I've never really had a decent experience with phones, much less with newer phones; my thoughts and expectations may be slightly on the negative/lesser side of things because of this.
I have the same issue. I used to use GScroll to flip pages in e-books, but finally it got so buggy I had to dump it (other GScroll features were almost impossible to use). I've used NavPad DBug. What I do find is that running the compact heaps command of SK Tools will temporarily fix things, at least most of the time. When I have NavPad DBug running, if it's logging compact heaps will clear out the traces. Somehow or other, it cleans things out when the nav pad gets confused or overloaded with inputs. But even compact heaps doesn't work all the time.
It would be sweet if the capacitative area really worked, but unfortunately it appears to get worn out about the time that you figure out all the things you can do with it. Lol.
Edit: I wonder if the device has problems distinguishing a single touch of your finger vs. several touches. That could really screw up the function of the device.

[Nook HD+] Touchscreen sensitivity experimentation (updated GUI app on 07/13)

Thanks to some testing by @alal690 based on my instructions in the post on how to play with touchscreen sensitivity we have some suggested reduced sensitivity values.
I spent entire day figuring how to do more things in android apps, and here's my next terrible UI attempt: NookTouchscreenSensitivityAdjuster-4.apk (backup link).
It's recommended that you run it with 07/11 cm10.1 release, but it will work with older releases too, just all settings (all but touch sensitivity on 07/06) would reset on screen off.
This minimalistic tool would allow you to change various touchscreen parameters in GUI, so not need to play with adb.
Right now this tool just lets you to play with values, the new values persist until the screen turns off (and the touchscreen resets restoring to default values).
The "save" button only saves the sensitivity and focus settings at this time, and the values persists across reboots (important to grant it root access the first time you run it!), everything else is not saved anywhere (since I am not sure there's any value in that and only included the other settings so you can experiment with them and perhaps get back to me with your awesome finds).
save button only works with 07/06 release, in previous releases save button does not persist anything.
the "reset" button resets values to the "old" values in the touchscreen, meaning pre-sensitivity change (so to 0x11, instead of the new 0x16).
Old post:
Anyway, here's a test zip that you can flash to try the new value. I encourage everybody to try it both if you do and do not feel like you have a touchscreen sensitivity issue.
If there are no downsides (so far I see none on my end), we might make this one to be a standard new default.
touchscreen-test.zip - this is for Nook HD+ (noop on the Nook HD). (Note, this is now included into 6/30 build for ovation)
Do not flash into stock, it'll break it.
I know the raw TS settings is a bit intimidating, so I plan to make an app for that, though that's a bit more effort than what I have time for ATM, so we'll make do with just this topical treatment.
will this work with carbonrom also? i like to switch between them.
What about those of us who felt it wasn't sensitive enough? Is there some sort of weird effect where its so over-sensitive that it feels less sensitive?
Apk07 said:
What about those of us who felt it wasn't sensitive enough? Is there some sort of weird effect where its so over-sensitive that it feels less sensitive?
Click to expand...
Click to collapse
Try it and tell me what you find out.
Feel free to play with other register values as per that other post referenced if you can. No two devices are totally the same. So we'll need a significant amount of testing by all interested parties to come to some middle groumd
Either that, or an app where people would be able to setup volume, touchscreen and other settings for themselves, but we'd still need to have sensible defaults.
bckrupps said:
will this work with carbonrom also? i like to switch between them.
Click to expand...
Click to collapse
It probably will work there too, but I did not test it on anything but my setup.
verygreen said:
It probably will work there too, but I did not test it on anything but my setup.
Click to expand...
Click to collapse
i tried it on both. What a difference already. especially on stock browser. sometimes i wanted to smash the thing lol. you are one hell of a dev. scrolling takes a bit more work but clicking links is so much better now. sometimes i would click a link four or 5 times lol
verygreen said:
Thanks to some testing by @alal690 based on my instructions in the post on how to play with touchscreen sensitivity we have some suggested reduced sensitivity values.
Anyway, here's a test zip that you can flash to try the new value. I encourage everybody to try it both if you do and do not feel like you have a touchscreen sensitivity issue.
Click to expand...
Click to collapse
Can I just flash this zip from CWM... or do I need to do any of the other steps mentioned in the post on playing with touchscreen?
dbrickg said:
Can I just flash this zip from CWM... or do I need to do any of the other steps mentioned in the post on playing with touchscreen?
Click to expand...
Click to collapse
You just flash this one from cwm and you are good to go for just this particular setting change.
verygreen said:
You just flash this one from cwm and you are good to go for just this particular setting change.
Click to expand...
Click to collapse
Thanks. That's what I thought. So I went ahead and did it while I was waiting for your response. So far it seems just fine. I guess I don't notice any great deal of difference. But I'm not sure what the symptoms were that you were trying to cure. I can't figure out how to read any of the context around the link to the post you mention in the OP
dbrickg said:
Thanks. That's what I thought. So I went ahead and did it while I was waiting for your response. So far it seems just fine. I guess I don't notice any great deal of difference. But I'm not sure what the symptoms were that you were trying to cure. I can't figure out how to read any of the context around the link to the post you mention in the OP
Click to expand...
Click to collapse
Basically the problem is: there are multiple complaints about people having too sensitive/too insensitive touchscreens. I don't think I am seeing anything like that, but there is more than one report so perhaps different people have different tolerances and different hardware has it's quirks.
The goal here is to arrive at a different set of defaults that would improve the situation for people that feel there is a problem and at the same time don't make situation worse for everybody else who feels current settings are ok.
verygreen said:
Basically the problem is: there are multiple complaints about people having too sensitive/too insensitive touchscreens. I don't think I am seeing anything like that, but there is more than one report so perhaps different people have different tolerances and different hardware has it's quirks.
The goal here is to arrive at a different set of defaults that would improve the situation for people that feel there is a problem and at the same time don't make situation worse for everybody else who feels current settings are ok.
Click to expand...
Click to collapse
mine personally was very insensitive. when i'm on slickdeals and i know you are also or bleacher report clicking on links such as hot deals and my subscriptions would take multiple tries. i didn't notice much system wide but mostly on the internet clicking smaller links. whatever he did with this zip made mine personally much much better.
verygreen said:
Basically the problem is: there are multiple complaints about people having too sensitive/too insensitive touchscreens. I don't think I am seeing anything like that, but there is more than one report so perhaps different people have different tolerances and different hardware has it's quirks.
Click to expand...
Click to collapse
Aha... then I guess I *did* know about the problem. I noticed it most when reading books. After a few minutes, the page turning touches on the lower right margin would seem to stop. I finally just gave up even trying to use them, and substituted a swiping movement. Anyway, with this zip installed, the sensitivity doesn't seem to be any worse, and might well be better. It hasn't been very long, but I'll keep an eye open for differences.
PS: Just read some of the other people's posts, and see that they were complaining of difficulty in clicking links in web pages. I had certainly noticed that, but hadn't even thought of the possibility of fixing it, so had just ignored it. With this new zip, it does seem like web link sensitivity is much improved.
dbrickg said:
Aha... then I guess I *did* know about the problem. I noticed it most when reading books. After a few minutes, the page turning touches on the lower right margin would seem to stop. I finally just gave up even trying to use them, and substituted a swiping movement. Anyway, with this zip installed, the sensitivity doesn't seem to be any worse, and might well be better. It hasn't been very long, but I'll keep an eye open for differences.
PS: Just read some of the other people's posts, and see that they were complaining of difficulty in clicking links in web pages. I had certainly noticed that, but hadn't even thought of the possibility of fixing it, so had just ignored it. With this new zip, it does seem like web link sensitivity is much improved.
Click to expand...
Click to collapse
mine is much improved. whatever adjustments were made I feel like mine could go even a little more in that direction. I'm at like 85% now lol
Thanks, alal690! I think he deserves to get some "Thanks", too, but he doesn't have any posts! And thanks to verygreen for providing the tools for alal690 to tweak with, of course!
I have noticed more mistypes on the keyboard verygreen if you are still needing input but the positives outweigh that in my book. it really is much better on websites. please thank the other dev that worked on this also alal690.
Big improvement for me, thanks I suspect the issue with pressing ui elements was caused by jitter and phantom touches that may have been caused by oversensitivity even though it seems somewhat counterintuitive... Selecting text was the big issue for me and it's now vastly improved. As has my accuracy on they keyboard - normally I'd have had to retype half of a post this length
BTW, it woul be great if people with some sort of a protective film over the touchscreen will give this a try to make sure it's also ok for them (I know those screen protectors are really popular amongst some people).
I've spent the last week trying to fix this, but it looks like it could be a problem with a driver or the actual hardware itself. If you go to Dev tools, then Pointer Location, you can see the reported values for touch pressure and size. When holding down my finger anywhere on the top 2/3 of my screen, the xy and size values jitter like crazy. If I touch the bottom 1/3 of my screen, there is no jittering and the size value is constant.
I think people that are complaining about low sensitivity are mistaken. I think the touch screen response is actually too sensitive, thus causing the jitter and making it hard to click on hyperlinks.
Reducing the touch sensitivity scale and increasing the touch size in the IDC file has helped a lot, but the jitter is still a problem that I hope someone can figure out how to fix.
Sent from my BN NookHD+ using Tapatalk 4 Beta
phammichael said:
I've spent the last week trying to fix this, but it looks like it could be a problem with a driver or the actual hardware itself. If you go to Dev tools, then Pointer Location, you can see the reported values for touch pressure and size. When holding down my finger anywhere on the top 2/3 of my screen, the xy and size values jitter like crazy. If I touch the bottom 1/3 of my screen, there is no jittering and the size value is constant.
I think people that are complaining about low sensitivity are mistaken. I think the touch screen response is actually too sensitive, thus causing the jitter and making it hard to click on hyperlinks.
Reducing the touch sensitivity scale and increasing the touch size in the IDC file has helped a lot, but the jitter is still a problem that I hope someone can figure out how to fix.
Sent from my BN NookHD+ using Tapatalk 4 Beta
Click to expand...
Click to collapse
I agree with phammichael that the real issue here would be jittering, but my symptom is different.
This is what I found so far.
- Pinch out to zoom in something (in gallery or web browser) and hold the 2 fingers on the screen.
- If my fingers are on the right side of the screen surface (portrait mode), the screen jitters. You can see the touch values changing fast and continuously by tiny values.
- If my fingers are on the left side of the screen surface, the screen won't jitter and stay calm.
Am I the only one experiencing this issue?
BTW, this is before I applied verygreen's touch screen patch. I will report back after applying it.
pins2u said:
I agree with phammichael that the real issue here would be jittering, but my symptom is different.
This is what I found so far.
- Pinch out to zoom in something (in gallery or web browser) and hold the 2 fingers on the screen.
- If my fingers are on the right side of the screen surface (portrait mode), the screen jitters. You can see the touch values changing fast and continuously by tiny values.
- If my fingers are on the left side of the screen surface, the screen won't jitter and stay calm.
Am I the only one experiencing this issue?
BTW, this is before I applied verygreen's touch screen patch. I will report back after applying it.
Click to expand...
Click to collapse
The jittery right hand side of the screen in portrait mode is the problem - I have this on my HD+ (which was new) and my Dad has it on his (which was new) as does my friend (again a new unit)
There have also been lots of others reporting the same thing on here.
Put the HD+ into landscape mode and the issue disappears.
I've just applied verygreen's 'fix' zip and I have to say that touch response is loads better - clicking links etc works far more reliably (but right hand side of the screen in portrait mode jitters still exist)
Great work as usual verygreen

Categories

Resources