Differences between SO-01B and X10i 2.1 kernels - XPERIA X10 Q&A, Help & Troubleshooting

Although the hardware is believed to be identical, there are some small differences I noted between the 2.1 global generic and the firmware I flashed on my SO-01B with SEUS:
- LED during mid-charge is not orange but yellow like in 1.6.
- Once charge reaches 100%, the bright green LED goes off. → The behavior is identical to 1.6
Build fingerprint: 'docomo/SO-01B_1233-7397/SO-01B/es209ra:2.1-update1/2.0.B.0.138/ihBA:user/release-keys'
The differences are big enough to make the JIT method not work.
hxxp://forum.xda-developers.com/showpost.php?p=9162192&postcount=576

It's not a difference in kernel.
In fact it just looks like the framework is odexed, or your core.jar is missing?
E/dalvikvm( 1790): /system/framework/core.jar odex has stale dependencies
I/dalvikvm( 1790): Zip is good, but no classes.dex inside, and no valid .odex file in the same directory
Click to expand...
Click to collapse

VirtualXDriver said:
Although the hardware is believed to be identical, there are some small differences I noted between the 2.1 global generic and the firmware I flashed on my SO-01B with SEUS:
- LED during mid-charge is not orange but yellow like in 1.6.
- Once charge reaches 100%, the bright green LED goes off. → The behavior is identical to 1.6
Build fingerprint: 'docomo/SO-01B_1233-7397/SO-01B/es209ra:2.1-update1/2.0.B.0.138/ihBA:user/release-keys'
The differences are big enough to make the JIT method not work.
hxxp://forum.xda-developers.com/showpost.php?p=9162192&postcount=576
Click to expand...
Click to collapse
It's a slightly different build, I assume a newer one, if only by a week or two.. Everyone else has a "A" build, we have a "B" build.

Related

Custom Kernel Deep Sleep Bug Hunt

** Notice **
There are 2 new kernels out K011 & K012 that are supposed to fix the deep sleep issue. If you are using a ROM that does not include one of these new Kernels please do not continue reporting deep sleep issues on them. Contact your ROM developer and get them to update the ROM with the new kernel. You could try another ROM with the new kernel's included or try the fixes mentioned further down in this post.
Mainly interested in now is reports of deep sleep issue with K011 and K012. Thanks out to everyone for their help with this bug hunt.
**********
Time to find out what those of us who have no deep sleep when running the custom kernels have in common. Even if you don't have any problems with Deep Sleep, your input would be appreciated as it could help eliminate items. If you can think of anything else we can add let me know and I'll update this Post. If possible please attach the same information as I have below as it would help greatly in determining what phones that absolutely can not get deep sleep with the FreeXperia Custom Kernel have in common. When posting possible solutions please indicate if you have had problems getting deep sleep before or not.
Update - I've added a count at the front of each line to count the number of submittal's for that line. As Items are eliminated from confirmed working deep sleep users I'll change the font color to red.
Update 2 - Thread has grown beyond my ability to keep up so I'm Just going to but a synopys at the bottom. If you believe that any of my assumptions there are incorrect, please PM me with the details so I can keep this information as accurate as possible
When reporting deep sleep problem or working include the following information
-------------------------------------------------------------
Phone Information from under battery
Sony Ericsson 10a
Type: AAD-3880074-AV AAAB
FCC ID PY7A388074 CB5
IC: 4170B-A3880074 10W20
Carrier Information
1 Rogers Wireless
GSM/GPRS/EDGE on the 850 and 1900 MHz bands
UMTS/HSPA on the 850 and 1900 MHz bands
ROMS tried using custom kernel with deep sleep issues
All FreeXperia ROMs from V014 to V021
[ROM][MAY24]BlackEditionHD_1.4_SE[Gingerbread 2.3.4]
Base ROM's tried
Rogers 2.1 stock ROM
Generic X10a build 2.0.A.0.504
Generic X10i build 2.0.A.0.504
Baseband's tried
Base .52
J's custom .52
Base .49
Base .46
Mini-Boot versions tried
V1
V2
V3
Factory Unlocked No
-----------------------------------------------------------------------
Here is what I've been able to determine so far
Majority of X10a have Deep Sleep problems
Majority of X10i are able to obtain Deep Sleep in some way.
So far everyone reporting unlocked phones have deep sleep (Incorrect, some don't have deep sleep)
Possible solutions if you are having deep sleep issues
1. Install V2 or V3 of the mini-loader from the custom kernel thread (DoomLoRD) Link
2. Try the Screen Off Fix (9Lukas5) Link
3. Manually select your network operator. (The Gingerbread Man) Link
4. After fully charging the phone to 100%, reboot after disconnecting (DoomLoRD) Link
5. Un-check automatic updating in date & time settings. (Temoi19) Link
6. (My_Immortal) has a solution If it's the dialer keeping your phone awake. Link
#5 is what finally worked for me. If any of these fixes have worked for you make sure to thank whomever came up with it by following the link to their original post.
** BREAKING NEWS **
J has just posted in his CM7 thread.
small progress
weems that we have found the deep sleep issue
not shure for now but we will provide an baseband togherther with J003
this combination seems to fix deepsleep for all testers
ETA 24h to be shure
Click to expand...
Click to collapse
The new Rom release is now available in the second post here. The Kernel included does not include overclock so if you want overclock you'll have to download K012 from the first post.
Here are the known issues posted by J in his thread. Please note the items for Deep Sleep and make sure to apply them before reporting any deep sleep issues.
KNOWN BUGS
-after reboot USB mass storage is not working if USB in plugged (unplug/replug USB to fix)
-if you still have deepsleep bug disable autotime update and automatic network
-deepsleep appers only when device is not connected to USB and corect baseband is used
Click to expand...
Click to collapse
Nice one buddy, great idea, i do not suffer from deep sleep so will edit my post later with my results, do you want the date the phone was manufactured.
I ALWAYS HAVE HAD DEEP SLEEP.
*Sony x10i
*Type AAd-3880056-bv aaac
*FCC ID PY7A3880056 CB5
*IC 41708_A3880056 10w13
*manufacture date year 2010 week 13
Using 3UK, HSPA, i have tried only Jerpelea's roms and had deep sleep on every one, my generic rom used is Sony Ericsson X10i_2.1.A.0.435_X10i and i use J's bb52 but it works also for me on bb49/sony bb52.
My only other point is that my phone was factory unlocked which may be another good point.
For those that don't know how to tell the manufacture date its under the battery and is shown like mine as 10W12, 10 being the year 2010 and W12 is week 12 of that year so April.
Apps that are very active in background prevent my device from going to deepsleep. Like MSN Talk Pro, PowerAMP and idk what more. I dont think its hardware related.
Sent from my X10i using Tapatalk
smartgardens said:
Nice one buddy, great idea, i do not suffer from deep sleep so will edit my post later with my results, do you want the date the phone was manufactured.
Click to expand...
Click to collapse
Great idea!! Now just where would I find that info?
PlayGunsta said:
Apps that are very active in background prevent my device from going to deepsleep. Like MSN Talk Pro, PowerAMP and idk what more. I dont think its hardware related.
Sent from my X10i using Tapatalk
Click to expand...
Click to collapse
I've actually tried this from a fresh install, with only CPU Spy installed and still I didn't have deep sleep. I'm saying it's hardware related as it could just as well be provider related.
GreatBigDog said:
Great idea!! Now just where would I find that info?
Click to expand...
Click to collapse
No idea sorry, Doomlord should know, it is under the battery somewhere, iv'e looked but can not make it out.
Hi guys, no deep sleep, no matter what
Edit: i received from Tim-br...
X10a Sensuous Black
Mobile Phone GSM/GPRS/EDGE 850/1800/1900
UMTS/HSPA band 1/2/5/6
Type: AAD-3880074-AV AAAB
FCC ID PY7A388074 BY9
IC: 4170B-A3880074 10W21
This is my post in J's thread...
I'm trying to fix deep sleep issue since J release v21 final, but no success.
My steps:
* usual (wipe everything, install update.zip, BB from package.)
* v2 e v3 miniloader (one in each clean install)
* repair with seus (after that clean install of bb from package and rom)
*flash many se 2.1 firmware (after that clean install of bb and rom)
* clean install and bb .49.
*battery calibration (full charge "off line", wipe battery stats)
*applied patch "when screen off "
*no apps installed (in each installation, JUST CPU spy from market, without gapps.
*another formatted sdcard
*wifi set to "when screen off " or wifi disabled.
*GPS and bt disabled.
*tried adw or arc launcher
No DEEP SLEEP. (Just work with airplane mode active)
Sent from my X10 using XDA App
michaeljordam said:
Hi guys, no deep sleep, no matter what
This is my post in J's thread...
I'm trying to fix deep sleep issue since J release v21 final, but no success.
My steps:
* usual (wipe everything, install update.zip, BB from package.)
* v2 e v3 miniloader (one in each clean install)
* repair with seus (after that clean install of bb from package and rom)
*flash many se 2.1 firmware (after that clean install of bb and rom)
* clean install and bb .49.
*battery calibration (full charge "off line", wipe battery stats)
*applied patch "when screen off "
*no apps installed (in each installation, JUST CPU spy from market, without gapps.
*another formatted sdcard
*wifi set to "when screen off " or wifi disabled.
*GPS and bt disabled.
*tried adw or arc launcher
No DEEP SLEEP. (Just work with airplane mode active)
Sent from my X10 using XDA App
Click to expand...
Click to collapse
Can you edit your post please to add the information asked for in the first post. We are trying to find something in common between all of our phones.
GreatBigDog said:
I've actually tried this from a fresh install, with only CPU Spy installed and still I didn't have deep sleep. I'm saying it's hardware related as it could just as well be provider related.
Click to expand...
Click to collapse
Yes I tried the same earlier, its just a matter of finding something we have in common, what it is I don't know.
Ok guys just done some testing cm7.03 and disappointing results, atleast for me.
Flashed bb.49, wiped everything in recovery. Fresh install of v21 and flashed v2 miniloader (no reboot in between) and signed into Google, allowed it to sync contacts no Google restore, no gmail, skipped install of Google apps. I have only installed CPU spy through market. Nothing restored through titanium.
My SD card is being a SOB and for some reason thinks I have no pics on my card, so possible issue there so it wont allow me to upload the screenies I just took but I managed 9 seconds of deep sleep after leaving the phone alone after my install as described for maybe about 2 - 2.5 hours. Baseband.49 with v2 miniloader isn't the answer I'm affraid atleast for me. Tomorrow will try same kind of install but using v3 mini loader.
Edit; no luck sorry
Sent from my X10 using XDA Premium App
GreatBigDog said:
Can you edit your post please to add the information asked for in the first post. We are trying to find something in common between all of our phones.
Click to expand...
Click to collapse
Edited, my last post. I see small diferences between you and me.
i don´t remember what bb came with phone, but i gues was .46.
I´ll try that now....
.49 don´t fix deep sleep.
How come we havn't asked those with deep sleep working to make a nandroid backup from xrecovery and upload it so we all can try it out?
SkyStars said:
How come we havn't asked those with deep sleep working to make a nandroid backup from xrecovery and upload it so we all can try it out?
Click to expand...
Click to collapse
Good idea, worth a try, i will do one tomorrow if nobody has done it before, think i will just check with Jerpelea first if that's ok by him.
Edit .... I will just pm a few with it rather upload it.
Sent from my X10 using XDA Premium App
Great! let us know how that goes.
I guess I'm a bit behind on what this even is.. What is the deep sleep? I know sleep mode and hibernation on a PC, but how do I know if my phone has it or does it?
CPU spy from market
highaltitude said:
I guess I'm a bit behind on what this even is.. What is the deep sleep? I know sleep mode and hibernation on a PC, but how do I know if my phone has it or does it?
Click to expand...
Click to collapse
Sent from my X10i using XDA App
Deep Sleep is sort of like Stand-by mode on your PC, it will wake up by its own to sync mail and do whatever and then go back to sleep. Without deepsleep you will lose 5-10% battery an hour just idling instead of just 1% for every 5-10 hours.
GreatBigDog said:
I've actually tried this from a fresh install, with only CPU Spy installed and still I didn't have deep sleep. I'm saying it's hardware related as it could just as well be provider related.
Click to expand...
Click to collapse
Ok just tired another fresh install on bb.49 this time using miniloader v3. Same testing.g environment as mentioned in cm7 thread. It seems impossible for me.
Ok guys just done some testing cm7.03 and disappointing results, atleast for me.
Flashed bb.49, wiped everything in recovery. Fresh install of v21 and flashed v2 miniloader (no reboot in between) and signed into Google, allowed it to sync contacts no Google restore, no gmail, skipped install of Google apps. I have only installed CPU spy through market. Nothing restored through titanium.
My SD card is being a SOB and for some reason thinks I have no pics on my card, so possible issue there so it wont allow me to upload the screenies I just took but I managed 9 seconds of deep sleep after leaving the phone alone after my install as described for maybe about 2 - 2.5 hours. Baseband.49 with v2 or v3 miniloader isn't the answer.
Sent from my X10 using XDA Premium App
Does anybody have an unlocked phone that they could try a sim card from another provider with? I've tried a 2nd one from the same provider but my phone is locked so I can't try a different one.
cm 7.0.3 - bypass v3
I tried a different simcard today and got deepsleep on it but I also had deepsleep with my usual simcard until it hit 15 % battery. Maybe losing deepsleep had to do with coming out of airplane mode on my device. I couldn't restore it and tried everything written in the other thread.
I gave the hardware - software cause some thoughts. I don't think it is hardware related. Sony would never upgrade to GB if it would be hardware related but I remember they said that they had some trouble with GB and the kernel.
Anyways I use the same apps on J's GB that I used on freex10 and i have absolutely no problems with deepsleep. The problem must reside within the Kernel, bypass or GB since lots of people even get the no deepsleep on a fresh install or lose it after some time.
Also drivers are not complete like camera and proximity. I would guess that there are still lots of incompatibilities that could also be the cause for something like that. Maybe some piece of hardware doesn't get addressed right or something. Figuring out what would be the cause exactly is way out of my league.
Just some 5 cents from me.
Regards
No deepsleep here for the past 3 or 4 versions of J's kernel (didn't try any before that).
Phone info:
Code:
Sony Ericsson X10a
Type: AAD-3880074-AV ACAM
FCC ID: PY7A3880074 CB5
IC: 4170B-A3880074
MoP 1233-8309.10 SI 1237-4604
Manufacture Date: 10W33 (August 2010)
Made in China
Network info:
AT&T
HSPA
WCDMA preferred
(Sorry, don't know where to find the same network info you posted).
ROMs:
Only use Z's FreeX10 B4 - Hotfix 1 (with CM settings)
Basebands:
.49
.52
J's .52
Bootloader bypass versions:
V1
V2
V3
Running services:
Code:
XPERIA Flashlight; Process: jp.co.telemarks.ledlightx10
LedLightWidgetService
Digital clock; Process: com.sonyericsson.digitalclockwidget
DigitalClockService
Market; Process: com.android.vending
AlarmService
HTC_IME mod; Process: jonasl.ime
HTC_IME mod
Misc:
FreeXperia Recovery
I keep Wifi/3G turned off unless I am using them, so they shouldn't be causing an issues.
Not using any SetCPU profiles/task killers/OC APPS/UV APPS/UV modules
I do get deep sleep in airplane mode (only tested for 2 minutes, but that was 2 minutes more than I've gotten any other time).
Perhaps a radio is keeping the CPU running?

[Q] [REQ]Pls Help - ACTIVE ST17 Random Reboots

I am writing in desperation.
My Xperia Active is about 3 months old.
It is the ST17a version but that has never made much difference to anything else.
It has been unlocked via testpoint method so warranty is not valid (plus they don't sell this phone in my country anyway).
Phone randomly reboots when it is not plugged in to power, especially when trying to bring the phone out of lockscreen standby, after using any GPS function or just unlocking the phone
The fact that this WILL NOT happen when plugged in to power is telling me perhaps it is a power management issue.
Things I have tried so far: COMPLETE REFLASH formatting all available mounts from recovery mode + format SD card and start fresh.
Combinations:
1.Spartan Rom 1.9.5 + non-oc kernel
2. Spartan Rom + Rage Kernel
3. Different baseband versions including original (56??), .66 and .62 (currently have .62 installed)
4. EX MIUI + ST17 satsuma kernel
Reboots still happen with all of the above configs
The problem started suddenly whilst using GPS SportyPal pro + Bluetooth music I have previously not had any problems in this area but now they will not stop.
I did not experience any reboots with CM9 but I did not stay in that rom very long as there was way too much stuff not working on my handset - it probably would have happened eventually. Sometimes I can get over half a day of usage without a reboot but then will get like 5 in a row.
I absolutely LOVE this phone but I bought it primarily for sports tracking and it keeps crashing and losing my workouts so I'm getting pretty pissed off.
If anybody can tell me what to look at to try and track this down I would really appreciate it. Checking logcats are no good because it will never reboot whilst plugged in to USB. Are there any other logs I should check???
OK - if not any help to identify the cause - can anyone comment on repair cost @ sony service center out of warranty?
Try fully stock 4.0.2.A.0.62 firmware, some other phones had issues with restarting all the time also, this firmware is known to fix this.
SOLVED - have carried out 5 days of testing.
I can confirm the latest version of MIUIex has addressed my reboot woes.
Have been running reboot logger 1.9 on the phone since then and have not had a single unexpected reboot. I am a happy camper.
Not sure if perhaps there is/ was a common driver or something that was causing these problems between a few of the up to date roms - I am just happy the issue seems to have passed for now.
And it's back. I switched back to Telstra, that means back to 850mhz band, reboots are back with vengeance. Might need to try and find a st17a specific Rom and radio.
bandario said:
And it's back. I switched back to Telstra, that means back to 850mhz band, reboots are back with vengeance. Might need to try and find a st17a specific Rom and radio.
Click to expand...
Click to collapse
Flash this via Flashtools: http://www.mediafire.com/?yvjnt0qctun9t6j
Is ST17a Specific ICS Firwmare. It contains ST17a specific baseband(radio).
Hope it helps
pastillero said:
Flash this via Flashtools: http://www.mediafire.com/?yvjnt0qctun9t6j
Is ST17a Specific ICS Firwmare. It contains ST17a specific baseband(radio).
Hope it helps
Click to expand...
Click to collapse
Wow, thanks!
Seems to have done the trick but now is back to vanilla flavour - seems the rooting process has gotten ridiculous lately, I wonder if it would be easier to pull this ftf to pieces and insert the radio detials etc into another rom file.
For now I guess I will use stock phone unrooted, at least it won't reboot.
THanks for your help!
i have also had that problem. I flashed the sti7a ics version and it is gone, after that you can use any rom without any prob.
bandario said:
Wow, thanks!
Seems to have done the trick but now is back to vanilla flavour - seems the rooting process has gotten ridiculous lately, I wonder if it would be easier to pull this ftf to pieces and insert the radio detials etc into another rom file.
For now I guess I will use stock phone unrooted, at least it won't reboot.
THanks for your help!
Click to expand...
Click to collapse
No prob..
You can flash any custom ROM you want, if it start rebooting again then get back to Stock..
Cheers
I had the same problem, it used to happen when signal was scarce or absent, i solved by getting a new sim card

[HACK][SOURCE] Forced touchscreen fw update

This concerns kernel devs and people with touchscreen problems.
As I saw android 4.4 updates didn't fix the touchscreen for some. In addition it made things worse, like in my case.
So I tried to force firmware updating the touchscreen module and try many different firmwares.
To ease that process I modified the already existing code of the drivers (ektf3k.c) to enable firmware update by demand and using a filepath.
In the package below you will find the modified driver source file for custom kernel developing, all firmware updates until now and a help file for how to flash them.
All fw were modified to a version number of 99, to lock them in place and not let future auto updates (unlock fw included).
Kernel driver source + Firmware files
EDIT: Added patch, check below
EDIT2: Corrected the typo in patch. Thanks @Tk-Glitch
After creating the custom kernel, you can use terminal or adb or a text editor to echo or edit /sys/devices/i2c-3/3-0010/update_fw and write the fw filepath. You can also check fw version by checking /sys/devices/i2c-3/3-0010/vendor (the last numbers are your fw version).
Check readme for more
To all kernel devs:
The modified drivers do not break anything. They just create a special sysfs file. The code is activated only by writing a path in that sysfs file.
To all others with touchscreen problems:
This needs a custom kernel with this modified driver. Otherwise you can't use the firmware files
Kernels with touchscreen force update patch:
Glitch
SaberMod franco.Kernel
X-Flo
If you incorporated the patch into your kernel please inform me to update the list.
Changelog of touchscreen firmwares:
Code:
A011:
A011 is the lastest touch firmware and used for factory image.
firmware update history:
A009:
Update boot-code Ver: 60.47 (Do not turn off WDT in IAP process).
Add Self Check Mechanism to enter suspend mode.
Modification Algorithm, and Digital Parameters for ER Module.
a. Improve "double fat finger (22 phi copper)" performance.
b. Improve "fast drawing straight line" performance.
c. Improve "multi-finger in co-axis (8 phi copper)" performance.
d. Improve "under table lamp" performance.
A010:
Modify Calibration Mechanism.
A011:
Modify Open/Short Calibration Mechanism for ESD problem.
Improve Stylus Performance
=================================================================
A012:
1.Fix the Sensitivity Problem That Suspend and Wake up When Charging.
2.Modification for Algorithm, and Digital Parameters for ER Module.
a.Improve "double fat finger (22 phi copper)" performance when charging.
=================================================================
A021 (A015):
Fix:
8717455: Holding touch panel edge by one finger,the touch action will be affected.
8768005: Holding Flo along the left edge of the device reveals false touch problems.
8840158: Pinch zoom in and out on chrome and maps doesnt seem as accurate / responsive as on nakasi.
8798827: Touch can not be correct clicked.
customer report: pin unlock will lose touch report.
Modification for AC De-noise Mechanism.
=================================================================
A025:
Based on A023 and remove the grip information.
1.[Issue 8840158] [FLO] pinch zoom doesnt seem as responsive / reliable as on nakasi.
2.[Issue 8858019] [FLO] Poor touch resolution.
Need calibration to fix these two issue with ER1/ER2 device:
3.[Issue 9052815] [FLO] Dead "Bands" on the touch screen.
4.[Issue 8930391] Spurious touchscreen taps.
=================================================================
A028:
Improve hovering like Nakasi
=================================================================
A029:
Need calibration for below Issue:
[Issue 9261425] New touch firmware is needed to apply the new parameters.
=================================================================
A031:
Fix following issues:
1.Latency on resume
2.Edge swipes lost (notification panel)
3.[Issue 9284036][FLO]ABS_MT_TOUCH_MAJOR values change
base on number of pointers down
=================================================================
A032:
1.firmware to A032.
2.modify driver to update power source when use slim port to hdmi.
A032 fix issue:
9471817:Hover touches.
9366206:Device show multiple-ghost touch after connect HDMI and tilt device to 40~65 degree.
9471956:Unclean lift offs.
9471962:Edge swipes causing hooks or missing touch points.
Noise problem.
Add ESD Protection Mechanism.
=================================================================
A033:
Fix issue:
1.Seeing lost touches.
2.Magic disappearing square.
3.Need to clean up small pointer dancing.
4.Pointer splitting.
5.Jumps between two large touches.
6.9521124:Poor performace in handwritten input method.
Bug: 9472154
Bug: 9471828
Bug: 9521124
Bug: 9471817
=================================================================
A034:
improvement:
9323363:Scrolling not smooth.
=================================================================
A036:
Fix issue:
962875:Need a new touch firmware to pass ESD test.
=================================================================
A038:
Fix/Improve issue:
1.[Issue 9692415] Touch screen on flo is broken for games
2.[Issue 9323363] [FLO/DEB] [Touch] Scrolling not smooth
3.[Issue 9471817] [FLO/DEB] [Touch] Hover touches
=================================================================
A040:
Fix/Improve issue:
1.[Issue 9692415] Touch screen on flo is broken for games.
2.Broken line.
3.Two Fingers Tapping Easy Connectivity.
4.co-axis problem with two finger.
=================================================================
A042:
Fix/Improve issue:
1.ghost touch probelm that caused by lcm slef-noise.
2.broken line in particular area.
=================================================================
A043:
Bug: 10388772
Bug: 10391124
Bug: 10541253
=================================================================
A044:
Fix issue:
1. Improve stability of two finger co-axis.
2. Ghost point problem of two fingers.
I've never had a touch screen problem
I have had touch issues. Not so much with 4.4.4
Sent from my Nexus 7 2013 using Tapatalk
CTCaer said:
This concerns kernel devs and people with touchscreen problems.
As I saw android 4.4 updates didn't fix the touchscreen for some. In addition it made things worse, like in my case.
So I tried to force firmware updating the touchscreen module and try many different firmwares.
To ease that process I modified the already existing code of the drivers (ektf3k.c) to enable firmware update by demand and using a filepath.
In the package below you will find the modified driver source file for custom kernel developing, all firmware updates until now and a help file for how to flash them.
All fw were modified to a version number of 99, to lock them in place and not let future auto updates (unlock fw included).
Kernel driver source + Firmware files
EDIT: Added patch, check below
After creating the custom kernel, you can use terminal or adb or a text editor to echo or edit /sys/devices/i2c-3/3-0010/update_fw and write the fw filepath. You can also check fw version by checking /sys/devices/i2c-3/3-0010/vendor (the last numbers are your fw version).
Check readme for more
To all kernel devs:
The modified drivers do not break anything. They just create a special sysfs file. The code is activated only by writing a path in that sysfs file.
To all others with touchscreen problems:
This needs a custom kernel with this modified driver. Otherwise you can't use the firmware files
Changelog of touchscreen firmwares:
Code:
A011:
A011 is the lastest touch firmware and used for factory image.
firmware update history:
A009:
Update boot-code Ver: 60.47 (Do not turn off WDT in IAP process).
Add Self Check Mechanism to enter suspend mode.
Modification Algorithm, and Digital Parameters for ER Module.
a. Improve "double fat finger (22 phi copper)" performance.
b. Improve "fast drawing straight line" performance.
c. Improve "multi-finger in co-axis (8 phi copper)" performance.
d. Improve "under table lamp" performance.
A010:
Modify Calibration Mechanism.
A011:
Modify Open/Short Calibration Mechanism for ESD problem.
Improve Stylus Performance
=================================================================
A012:
1.Fix the Sensitivity Problem That Suspend and Wake up When Charging.
2.Modification for Algorithm, and Digital Parameters for ER Module.
a.Improve "double fat finger (22 phi copper)" performance when charging.
=================================================================
A021 (A015):
Fix:
8717455: Holding touch panel edge by one finger,the touch action will be affected.
8768005: Holding Flo along the left edge of the device reveals false touch problems.
8840158: Pinch zoom in and out on chrome and maps doesnt seem as accurate / responsive as on nakasi.
8798827: Touch can not be correct clicked.
customer report: pin unlock will lose touch report.
Modification for AC De-noise Mechanism.
=================================================================
A025:
Based on A023 and remove the grip information.
1.[Issue 8840158] [FLO] pinch zoom doesnt seem as responsive / reliable as on nakasi.
2.[Issue 8858019] [FLO] Poor touch resolution.
Need calibration to fix these two issue with ER1/ER2 device:
3.[Issue 9052815] [FLO] Dead "Bands" on the touch screen.
4.[Issue 8930391] Spurious touchscreen taps.
=================================================================
A028:
Improve hovering like Nakasi
=================================================================
A029:
Need calibration for below Issue:
[Issue 9261425] New touch firmware is needed to apply the new parameters.
=================================================================
A031:
Fix following issues:
1.Latency on resume
2.Edge swipes lost (notification panel)
3.[Issue 9284036][FLO]ABS_MT_TOUCH_MAJOR values change
base on number of pointers down
=================================================================
A032:
1.firmware to A032.
2.modify driver to update power source when use slim port to hdmi.
A032 fix issue:
9471817:Hover touches.
9366206:Device show multiple-ghost touch after connect HDMI and tilt device to 40~65 degree.
9471956:Unclean lift offs.
9471962:Edge swipes causing hooks or missing touch points.
Noise problem.
Add ESD Protection Mechanism.
=================================================================
A033:
Fix issue:
1.Seeing lost touches.
2.Magic disappearing square.
3.Need to clean up small pointer dancing.
4.Pointer splitting.
5.Jumps between two large touches.
6.9521124:Poor performace in handwritten input method.
Bug: 9472154
Bug: 9471828
Bug: 9521124
Bug: 9471817
=================================================================
A034:
improvement:
9323363:Scrolling not smooth.
=================================================================
A036:
Fix issue:
962875:Need a new touch firmware to pass ESD test.
=================================================================
A038:
Fix/Improve issue:
1.[Issue 9692415] Touch screen on flo is broken for games
2.[Issue 9323363] [FLO/DEB] [Touch] Scrolling not smooth
3.[Issue 9471817] [FLO/DEB] [Touch] Hover touches
=================================================================
A040:
Fix/Improve issue:
1.[Issue 9692415] Touch screen on flo is broken for games.
2.Broken line.
3.Two Fingers Tapping Easy Connectivity.
4.co-axis problem with two finger.
=================================================================
A042:
Fix/Improve issue:
1.ghost touch probelm that caused by lcm slef-noise.
2.broken line in particular area.
=================================================================
A043:
Bug: 10388772
Bug: 10391124
Bug: 10541253
=================================================================
A044:
Fix issue:
1. Improve stability of two finger co-axis.
2. Ghost point problem of two fingers.
Click to expand...
Click to collapse
Been looking for something like this that embeds all the firmware variants in one package. Guess it's not usable unless integrated in some custom kernel... Will have to wait for aroma installer then. But thanks.
Well many have problems and were forced to flash custom kernels with different firmwares.
This can ease up a lot the process and help many people that are using custom kernels to test different fw.
For example many fixed their issues with A011 (aka ts10). Because like in my case, the problem comes from the LCD's electromagnetic interference on ELANtech's touch panel(tested with Nexus 7 Nexus 7 II Tester).
Well it's up to kernel dev's to incorporate the driver mod if they want.
I may make a stock based custom kernel, touch fw tester edition, when I get back home and refix my ubuntu vm.
So 1 flash, try firmwares, find the personal best, reflash previous kernel.
Hello, thanks for the patch, i'm using Glitch Kernel which implemented your patch and i'm going to try and fix my touchscreen but i want to know which version of the driver is the latest, is it the A011 or the A044?
And another question, i've just tested A011 and A044 but at vendor both show ELAN_KTF3K_xa064, are they both supposed to show the same version?
Trevomader said:
Hello, thanks for the patch, i'm using Glitch Kernel which implemented your patch and i'm going to try and fix my touchscreen but i want to know which version of the driver is the latest, is it the A011 or the A044?
And another question, i've just tested A011 and A044 but at vendor both show ELAN_KTF3K_xa064, are they both supposed to show the same version?
Click to expand...
Click to collapse
Yeah both of them show the same version to avoid updating to new one.
I may release an archive with incremental (bigger than a044) sometime.
The latest is A044 since months.
Also thank you for testing this mod.
Good luck mitigating the problem.
EDIT: Checked @Tk-Glitch kernel commits and corrected the typo to the above patch. Thanks
For the longest time I had crazy touch problems especially when I set the device down on the bed, etc, not unlike many others..but a week ago I pulled my screen protector off and all problems have completely vanished, put another one on, problems return immediately, can anyone explain this? And also is there a screen protector that I can use i.e, tempered glass, for anyone that found the same problem
trident76 said:
For the longest time I had crazy touch problems especially when I set the device down on the bed, etc, not unlike many others..but a week ago I pulled my screen protector off and all problems have completely vanished, put another one on, problems return immediately, can anyone explain this? And also is there a screen protector that I can use i.e, tempered glass, for anyone that found the same problem
Click to expand...
Click to collapse
Some have reported the same problem with screen protectors. I had similar issue with 2012 Nexus 7 when the screen got dirty with finger prints. I now use Nova screen. Lag almost unnoticeable. Matter of fact, it's mostly grounding issue. Too bad there still isn't a way to flash the OP batch, though, according to other threads, the multi-touch fix isn't for that. YAMTT records no problem, but if I leave it flat on a table, some areas don't register touches at all. I've used EMI tape as much as I can see and yet, no improvements. I guess it's between the LCD and the touchscreen. That being the case, no cure.
Hi @CTCaer,
I have touchscreen problems on my N7 Flo :crying:. Can you say me how I can test this patch? Or which kernel I must flash? It's there one, which have your patch already intergrated?
Thanks! :victory:
raXoN91 said:
....Can you say me how I can test this patch? Or which kernel I must flash? It's there one, which have your patch already intergrated? ....
Click to expand...
Click to collapse
Hey, check 1st post.
I added a list that I'll try to maintain
CTCaer said:
Yeah both of them show the same version to avoid updating to new one.
I may release an archive with incremental (bigger than a044) sometime.
The latest is A044 since months.
Also thank you for testing this mod.
Good luck mitigating the problem.
EDIT: Checked @Tk-Glitch kernel commits and corrected the typo to the above patch. Thanks
Click to expand...
Click to collapse
Do you know if there was a change from android 4.4 to 4.4.2 in the driver? From memory 4.4 seemed to be perfect and around 4.4.2 I started getting problems again (ghost touches)
CTCaer said:
Hey, check 1st post.
I added a list that I'll try to maintain
Click to expand...
Click to collapse
Hey, I fashed the Glitch kernel r151 and the issue is not fully fixed. It's Vetter! But ff the tablet get hot, the issue come again. Is there a special Version with the patch? Thanks :victory:
godutch said:
Do you know if there was a change from android 4.4 to 4.4.2 in the driver? ..
Click to expand...
Click to collapse
No it's A044. But when I updated 4.4.3->4.4.4, I checked logs and the touchscreen driver wrote something with calibration (can't remember exactly).
raXoN91 said:
Hey, I fashed the Glitch kernel r151 and the issue is not fully fixed. It's Vetter! But ff the tablet get hot, the issue come again. Is there a special Version with the patch? Thanks :victory:
Click to expand...
Click to collapse
The firmwares can't completely fix the problem, only can make the situation better. The device getting hot means that it uses a lot of power and thus more EMI for the touchscreen panel.
You have to test different fw because almost every fw has different internal calibration methods.
Also I learned that the correct way to use the forced calibration (as I mentioned in the other touchscreen problems thread, echo "FF" > /proc/ektf_dbg), is to put the device on an insulated surface, away from power sources, chargers and do it by using terminal emulator and not adb with wifi or cable. And also in flight mode.
CTCaer said:
No it's A044. But when I updated 4.4.3->4.4.4, I checked logs and the touchscreen driver wrote something with calibration (can't remember exactly).
The firmwares can't completely fix the problem, only can make the situation better. The device getting hot means that it uses a lot of power and thus more EMI for the touchscreen panel.
You have to test different fw because almost every fw has different internal calibration methods.
Also I learned that the correct way to use the forced calibration (as I mentioned in the other touchscreen problems thread, echo "FF" > /proc/ektf_dbg), is to put the device on an insulated surface, away from power sources, chargers and do it by using terminal emulator and not adb with wifi or cable. And also in flight mode.
Click to expand...
Click to collapse
I've noticed that Franco's kernel with embedded firmware comes in 2 versions: AOSP and CM. Should I switch between the two, would it be necessary to unlock the patch and reflash it again with the version that corresponds with the new rom? Thank you.
graphdarnell said:
I've noticed that Franco's kernel with embedded firmware comes in 2 versions: AOSP and CM. Should I switch between the two, would it be necessary to unlock the patch and reflash it again with the version that corresponds with the new rom? Thank you.
Click to expand...
Click to collapse
AOSP and CM are only the versions that are supposed to match with your rom. AOSP are the majority of the roms you'll probably use. (Stock, PA) CM are roms mostly derived from CM (basically CM, PAC)
In that case you only need to flash the rom if not already done. Then you need to flash the corresponding kernel version and you can use the said patch.
Envoyé de mon Nexus 7 en utilisant Tapatalk
zaclimon said:
AOSP and CM are only the versions that are supposed to match with your rom. AOSP are the majority of the roms you'll probably use. (Stock, PA) CM are roms mostly derived from CM (basically CM, PAC)
In that case you only need to flash the rom if not already done. Then you need to flash the corresponding kernel version and you can use the said patch.
Envoyé de mon Nexus 7 en utilisant Tapatalk
Click to expand...
Click to collapse
Thanks for the reply. Guess I wasn't very clear. I'm now on CM-derived rom. So I flashed CM-based patched kernel. But if I decide to switch to an AOSP-based rom, do I need to unlock the CM firmware and flash the AOSP firmware instead? Or is it the case that once you flash it, it would stick until you unlock again, no matter what rom you install in between?
It locks it. Doesn't matter the kernel. It will only update it if the touchscreen goes into recovery mode.
To unlock it, use A044-unlock.b. This is untouched, so it will show _xA044 in the vendor sysfs and not _xA064.
Hello again, yesterday i've tried to "unlock" the firmware, but even after applying A044-unlock.b multiple times it's still showing version 100 (x64) on vendor.
I'm still on glitch kernel, so i don't know if the patch is working correctly on this kernel.
Try the attached file. It's stock A011.
If it succeeds and you reboot the driver should then flash the kernel's stock which is A044 (it should show _xa044).

[Q] Doogee DG2014 and 1% battery problem

hi all folks
first of all, sorry for my bad engilsh.
i have the phone in the subject (MTK based phone), and as many other owner of this phone i have problem with KK firmware and battery level indicator.
the problem is: every time phone is turned off and power on again, it start at 1% battery level.
it is a fake battery level indication, because battery voltage is read correctly by any app with this feature.
i found who is the guilty: rom module LK.BIN (uboot)
as many DG2014 owner i must use LK.BIN from JB rom to have display working, with LK.BIN from KK rom i have only random lines on display.
i done some experiment and at the end i'm sure with LK.BIN from the original KK rom i don't have the 1% battery problem, because there isn't the sound alarm on startup, and beacuse using program MyPhoneExplorer, after connecting phone to PC i can see on PC the correct battery level. with the same proggram i see 1% battery level when bad LK.BIN is running on phone.
after these test i'm searcing for a solution, but is hard to me, so i'm now here to ask your help.
with a binary editor i see inside LK.BIN.
>> In KK version there are 3 LCD driver
1- hct_nt35521_dsi_vdo_hd_jdi
2- hct_nt35521_dsi_vdo_hd_lg
3- otm1283a_dsi_vdo_hd_boe
>> in JB versione only 2 LCD driver
1- hct_nt35521_dsi_vdo_hd_jdi
2- hct_nt35590_dsi_cmd_hd_auo
correct driver for my phone is the number 1 in either LK.BIN versions, but i don't know why, when KK version is on phone, it choice the driver number 2, as reported by MtkDroidTool.
there is a way to force off autodetection and manually specify the display driver to use ?
I answer by myself.
Here my modified LK.BIN:
http://www.needrom.com/download/1-battery-solution/
Hope this can be useful for other people.

THL 4400 touchscreen not calibrated after flashing new FW

I tried to change the internal partitions because by default only 1GB is available for apps, this failed horrible with the provided app (HK Rom Editor) and caused a bootloop.
Afterwards I re-flashed the phone with firmwares which I found on Needrom, that worked fine but the touchscreen is totally misaligned doesn't matter which one I use. If I press for example the top part the phone thinks that I'm touching the middle, it does work fine its just totally wrong calibrated. I can't check if anything else isn't working.
Model: THL 4440, 4GB
Android: 4.2.2
Has someone any idea how I can fix this issue? Do I need a different stock firmware?
Someone told me that I deleted/overwrote my touchscreen calibration files, is there any way how I can identify the chip or get proper files for the THL 4400?

Categories

Resources