On all roms there is a banding problem due to our screens 65 k colors.
But launcherpro fixes the banding problems on widgets But on some gingerbread roms it does not.
I have not made this fix. But I have tested it on some roms. Im am just sharing this. credit goes to Tytung who took his time and fixed this in his latest 2.5 rom and GPC + other developers that may have found this already.
I have tried it on Tyweens cm7 rom, Tytungs rom and some other. But it "maybe" works on others to? Like gpc's one?
= When using launcher pro with this fix. The launcher will have no banding problems on widgets and on the homescreen. But other banding problems will still be there like it was before.
So without fix ,the clock widget will look like crap in launcher pro in tyweens cm7 rom. But with the fix it will look smooth.
The zip changes the libsurfaceflinger.so.
Flash in cwm.
Thank God, at last! Great stuff
works on tyween cm7.
thank you! this is the main reason i switch some roms lol
will this work on other HD2 NAND roms using launcherpro? like using hyperdroid CM7?
The source code
If you are a developer of source code maker, the following is my fix for banding issue. (NOT Tytung, i don't know how he fix it )
The source code path :
/frameworks/base/services/surfaceflinger/SurfaceFlinger.cpp
Function position:
Code:
sp<Layer> SurfaceFlinger::createNormalSurface(
const sp<Client>& client, DisplayID display,
uint32_t w, uint32_t h, uint32_t flags,
PixelFormat& format)
{
// initialize the surfaces
switch (format) { // TODO: take h/w into account
case PIXEL_FORMAT_TRANSPARENT:
case PIXEL_FORMAT_TRANSLUCENT:
format = PIXEL_FORMAT_RGBA_8888;
break;
case PIXEL_FORMAT_OPAQUE:
//format = PIXEL_FORMAT_RGBX_8888;
format = PIXEL_FORMAT_RGB_565;
break;
}
sp<Layer> layer = new Layer(this, display, client);
status_t err = layer->setBuffers(w, h, format, flags);
if (LIKELY(err != NO_ERROR)) {
LOGE("createNormalSurfaceLocked() failed (%s)", strerror(-err));
layer.clear();
}
return layer;
}
Modified code:
Code:
case PIXEL_FORMAT_OPAQUE:
//format = PIXEL_FORMAT_RGBX_8888;
format = PIXEL_FORMAT_RGB_565;
break;
Reason:
It is caused by the display driver is not match with HD2. The gingerbread has fix it on driver level with new pixel format RGBX_8888, however it is still not working for HD2. So we have to force the default pixel format into RGB565
iamgpc said:
If you are a developer of source code maker, the following is my fix for banding issue. (NOT Tytung, i don't know how he fix it )
skip...
Click to expand...
Click to collapse
I think I need to clarify something.
Gpc sent a private message and asked me how I fixed the color banding issue on 27th March 2011, 09:50 PM.
P.S.: Admin can confirm this.
Because I was busy at that time, I replied his question by only giving him this link including libsurfaceflinger.so.
Then maybe he fixed this issue from the source code by himself, but he really got hints from me.
Let us go back about 20 days ago.
In fact, I tried to solve this color banding issue since 13th March 2011, and I shared my possible method in my thread.
http://forum.xda-developers.com/showpost.php?p=12049180&postcount=2487
When I had time, I released a modified/recompiled libsurfaceflinger.so on 21st March 2011, and requested other users to test it.
http://forum.xda-developers.com/showpost.php?p=12241393&postcount=2846
Then I included this .so file in my v2.5 ROM.
Gpc saw my change log and sent me a private message, and ...
That's the whole story.
what stops this being implemented throughout the UI (i.e. notification bar, buttons etc)?
I'm not getting 3g, only EDGE on this ROM. My APN settings are right and data is definitely on so I don't know what the problem is. I have a T-Mobile HD2 and T-Mobile. Please help.
Tried this fix, and i am truly impressed! No longer banding using ADW Ex and even XDA App! It does not seem to "fix" only launcher pro to me!
EDIT : copilot still has banding though, but not a big deal given the improvements elsewhere!
tytung said:
When I had time, I released a modified/recompiled libsurfaceflinger.so on 21st March 2011, and requested other users to test it.
http://forum.xda-developers.com/showpost.php?p=12241393&postcount=2846.
Click to expand...
Click to collapse
humm,
Do you modify the libsurfaceflinger like mine? Do you compile with BOARD_NO_RGBX_8888 := true to fix this isse?
Because my libsurfaceflinger is changed for my personal using. I also modify somewhere else for my testing, so that I cannot use your libsurfaceflinger.
I don't know what you modify in your source code, so I have to find out the root cause.
According to Google search on "android color banding", you can find a JAVA solution on color banding.
http://stuffthathappens.com/blog/2010/06/04/android-color-banding/
It says the PixelFormat will take effect on banding issue. Take a look on surfaceflinger source code, you can find out the PixelFormat.OPAQUE and PixelFormat.RGBA_8888 will cause the different result.
Finally, you just have to modify the PixelFormat.OPAQUE from RGBX_8888 to RGB_565
iamgpc said:
humm,
Do you modify the libsurfaceflinger like mine? Do you compile with BOARD_NO_RGBX_8888 := true to fix this isse?
Because my libsurfaceflinger is changed for my personal using. I also modify somewhere else for my testing, so that I cannot use your libsurfaceflinger.
I don't know what you modify in your source code, so I have to find out the root cause.
According to Google search on "android color banding", you can find a JAVA solution on color banding.
http://stuffthathappens.com/blog/2010/06/04/android-color-banding/
It says the PixelFormat will take effect on banding issue. Take a look on surfaceflinger source code, you can find out the PixelFormat.OPAQUE and PixelFormat.RGBA_8888 will cause the different result.
Finally, you just have to modify the PixelFormat.OPAQUE from RGBX_8888 to RGB_565
Click to expand...
Click to collapse
Yes, I compiled with BOARD_NO_RGBX_8888 := true to fix this issue.
BTW, you didn't know how I modified my source code because you didn't say that you want to see my modified source code in the private message.
Anyway, I only changed the color support of the config file from RGBX_8888 to RGB_565.
And you don't need to modify SurfaceFlinger.cpp because setting NO_RGBX_8888 in config file is enough.
Below is the detail.
Before Google released Gingerbread OTA 2.3.3, we could only use Froyo OTA's proprietary files to work with Gingerbread source code when we compiled AOSP Gingerbread 2.3.2 or 2.3.1.
And Froyo's proprietary files are not compatible with Gingerbread source code, so we needed to fix the RGB_565 problem by applying the following two commits at that time.
https://github.com/CyanogenMod/andr...mmit/d06cf2371b2db46ab4ecedea832e4c17f3591165
https://github.com/CyanogenMod/andr...mmit/baa078471f4198bd73819794f2713a845305a227
After Google released Gingerbread OTA 2.3.3, we used Gingerbread OTA's proprietary files to work with Gingerbread source code when we compiled AOSP Gingerbread 2.3.3.
At the same time, I removed these two commits above when compiling a new GB 2.3.3 ROM.
That's why hg3atintin (Thanks to him.) told me that my GB 2.3.3 ROM had the color banding issue but my GB 2.3.2 ROM didn't.
Finally, I added back these two commits and compiled a new surfaceflinger.so.
Then the color banding issue is improved.
it fixed my launcherpro , calculator had severe banding but its fixed too
nice job mate
How do I apply this fix to my AmericAndroid SD build? Any help would be appreciated...
P.S. I searched this thread and others and cant figure it out, please point me in the right direx... What is CWM btw because it says to flash with CWM?
How does this banding look like?
edit: sorry wrong thread
nice! gonna try this out
ludetekniq said:
How do I apply this fix to my AmericAndroid SD build? Any help would be appreciated...
P.S. I searched this thread and others and cant figure it out, please point me in the right direx... What is CWM btw because it says to flash with CWM?
Click to expand...
Click to collapse
Hi ludetekniq.
You're in the wrong place. This is NAND development forum and not SD development. You can only install this fix if you have a NAND Rom in your HD2.
EDIT: CWM = Clockworkmod Recovery. It's a tool that we use, to help flashing and recover Roms on NAND.
silly question..lol
what is the banding problem? )
alaaassem said:
silly question..lol
what is the banding problem? )
Click to expand...
Click to collapse
Well... from what I understood, its an issue related to the color depth of HD2 screens. Our HD2 Android is displaying 16 bit color insted of 32 bit because of a driver's issue... it loses quality because it's displaying less colors. With 32 bit you will see a smoother color gradient.
Please devs correct me if i'm wrong!
Cheers from Portugal!
Related
Please read this entire post!
<<<This is a link to the... FAQ Click it!!>>>
I have created a complete bundle of GingerBread with the newest kernel from GIT (Feb 28 / 1276), and the newest rootfs from GIT (Feb 24).
Please, feel free to DONATE to the XDAndroid project!
Every little bit helps!
Directions:
1. Download the full bundle (zip). (Updated Mar. 11 2011)
If instead you just want the system.ext2 (zip) file (updated Mar. 25 2011) by itself... Don't download this unless you know you want to update!!
2. Extract it. You’ll see a folder, GBX0A, copy its contents to the root (base) of your SD card. If you want to run Android from a folder instead of all the files on the root of the card, follow the steps below.
3. Go into the STARTUPS folder. Grab the appropriate startup.txt for your device (if you don't know what device you have, you should read the FAQ), and move it to the root of the card (or where you run haret.exe from. If you want to change the location of the build, put a rel_path= statement in the cmdline section of the startup.txt. Mine is located two folders deep on the SD, so my rel_path=Androids/TP2Ref)
4. Download the ts-calibration file from this post (should work for RAPH/DIAM) and extract the .zip to where you run haret.exe from - typically the root or 'base' folder of the SD card. Not within any folders, unless you run haret.exe from a folder!
**If you have issues with the above file, you must calibrate manually. To start manual calibration, just delete any ts-calibration files and boot. Sometimes this doesn't work, and you have to use an older kernel from GIT (I found 1225 works well) and generate the ts-calibration file by booting and pressing the 5 zones. Once you have a good calibration file hold on to it (make 15 copies if it's a really good one ), reboot & go back to the newest kernel from the autobuild service! Thanks**
6. Run haret.exe... Profit!
Let it settle out on the first boot. Many have reported they had to reboot basically because it was so slow - if you let it sit for about 10 mins or so the media scanner can go thru everything, etc. If you want adb in and watch the processes via top, you'll see why the phone seems so slow - there's lots of background processes cranking because this is the first boot .
Troubleshooting:
Please read the... FAQ
If you have any issues with the kernel, feel free to change it:
There are some devices that are having issues with the newest kernels. Please see the kernel autobuild service to get newer or archived kernels. Once you download a replacement kernel, go to where you run haret.exe from - remove your old zImage/modules-xxxxx.tar.gz. Take the new zImage/modules-xxxxx.tar.gz and replace the old ones, same folder - where you run haret.exe from. Make sure the ‘zImage’ is named just that. Do not rename the modules file, do not extract it - should be in .tar.gz format.
In addition, I am using the newest rootfs from GIT - Feb. 10. There are other rootfs options, see F22’s thread if you want to use more keys or move around the home button. Of course, there's always the roofs autobuild service if you want "stock" rootfs images - F22 builds his off of these and modifies them .
See Incremental Updates for more information on this topic of updating!!
Random issues can often be solved by forcing the system to create a new data.img. If you're worried about losing data, Titanium Backup works quite well. If you wish, you can rename the data.img to something else, and let the system create a new one - just to see if it resolves your problem.
Similarly, if you wish try formatting your SD card - I prefer to use the HP Tool - do a full format, FAT32.
Obviously this build is Alpha, and there are going to be problems. The next post will address issues particular to this build - PLEASE READ THESE before asking questions! Feel free to post questions in this thread, I will do my best to address them. Big thanks to stinebd for releasing the system image, and of course the other developers for their hard work on making these kernels available.
stinebd's original post HERE
Mar. 11 2011 Changelog
Things broken in this build:
libgps - explains itself, no GPS and probably no netloc (cell-tower based location)
HW3D is either not working, or just not preset - I ran neocore and let's just say a lot of textures are missing. Got a ridiculous score, but again - no textures, not much to render haha.
DIAM users have reported the g-sensor is flipped -
emc02 said:
ok, the solution for DIAM100 is:
change "gsensor_axis=2,-1,3" to "gsensor_axis=-2,1,3"
works perfect!
Click to expand...
Click to collapse
This line should be in the startup.txt file if that wasn't obvious .
Thanks to emc02 for that solution!
The keyboard should be fine, but I've had some Blackstone users complain - if you want, switch the keyboard out with another. mr_grisha's original post.
Anything that's broken in Froyo, is probably broken in this build.
Changelog March 25, 2011
Further tuned keyboard sizing to be useful on VGA (also made the key labels bigger)
World-phone settings support (very important for future RIL enhancements especially on CDMA)
Enabled HSPA icon (someone on GSM please let me know if it actually works)
Strange, I never saw the calib. screen. Probably because I just replaced the system.ext2 file and let it run it's course? I do recall there being calib. "issues" with 1253 because the text would continue to scroll even while you were attempting to tap the points.
[R^7Z EDIT]I don't suppose that would work if you were re-working this from the beginning or starting over from scratch (doh!). Lucky me, I save just about every new/update regardless of kernel package or rootfs update[/R^7Z EDIT]
R^7Z said:
Strange, I never saw the calib. screen. Probably because I just replaced the system.ext2 file and let it run it's course? I do recall there being calib. "issues" with 1253 because the text would continue to scroll even while you were attempting to tap the points.
Click to expand...
Click to collapse
If you already have a ts-calibration file, there's no need to create a new one .
A note to all troubled users
And additionally, to those who are reporting issues, it would behoove you to help the devs out by not making them guess what you are using (device) to include but not limited to, which rootfs and kernel package (date of release and commit ID might be helpful). That being said, be aware of what you are talking about (app discussion should not fall into this thread) and please keep talks of neopeek variants to a minimum (I know a little bit about these builds and will help out where I can, PM me if you really think I can help and if you really need some assistance)
arrrghhh said:
[CUT]
HW3D is either not working, or just not preset - I ran neocore and let's just say a lot of textures are missing. Got a ridiculous score, but again - no textures, not much to render haha.
I noticed the (virtual) keyboard is comically tiny. Seems to work fine tho...
[CUT]
Click to expand...
Click to collapse
Noticed the same thing in neocore (30fps without textures) but quadrant 3D tests looked good. Probably we need to test more apps.
Virtual keyboard is unusable on Blackstone. All buttons are tiny and kind of stacked together at the bottom of the screen.
BTW. I can't calibrate the screen since .35 FB commit (17 Dec). I think it's because of FB debug spam - this should be easy to fix.
on ralph100 i've been stuck on "busy" loop for about 20mins now.
loops saying "INFO task rpcrouter: 15 blocked out for more than 120 seconds"
thebranded said:
on ralph100 i've been stuck on "busy" loop for about 20mins now.
loops saying "INFO task rpcrouter: 15 blocked out for more than 120 seconds"
Click to expand...
Click to collapse
Hrm. I'm going to go with format your SD card using the HP Tool. Full format, FAT32.
Also, might want to try kernel 1258. Let me know if neither of those things improve it. Is it creating a data.img...?
g3rm0 said:
Noticed the same thing in neocore (30fps without textures) but quadrant 3D tests looked good. Probably we need to test more apps.
Virtual keyboard is unusable on Blackstone. All buttons are tiny and kind of stacked together at the bottom of the screen.
BTW. I can't calibrate the screen since .35 FB commit (17 Dec). I think it's because of FB debug spam - this should be easy to fix.
Click to expand...
Click to collapse
Sorry forgot to respond to this. Two things, there's an app called "gingberbread keyboard". Find it on the market, see if it works better for you.
Second, I've heard the new 1258 kernel is fixed - not sure how, but it seems the clocks commit from jb's tree did it.
works with 1258, did reformat SD card.
Still had issue with not seeing the touchscreen calibration though, just copied over a old TS file.
thebranded said:
works with 1258, did reformat SD card.
Still had issue with not seeing the touchscreen calibration though, just copied over a old TS file.
Click to expand...
Click to collapse
Glad it's working for you. I think we've narrowed down the calib issue, so hopefully we'll get that fixed for all devices soon. Not sure why but RHOD's don't seem to have the issue any longer - no matter, we will get it resolved for all devices. Sometimes just takes longer for the old devices unfortunately...
[xda Link] Gingerbread keyboard
For those of you who need it, check THIS thread out!
n-Joie! (Enjoy)
R^7Z said:
For those of you who need it, check THIS thread out!
n-Joie! (Enjoy)
Click to expand...
Click to collapse
I've heard it doesn't work... Per another user, I have added a different recommendation:
arrrghhh said:
I noticed the (virtual) keyboard is comically tiny. Seems to work fine tho... If it does bother you, user "webxplore" said the app "gingerbread keyboard" (on the Market) fixes the issue. Thanks webxplore!
Click to expand...
Click to collapse
Gingerbread keyboard (market version)
Will see about getting the market version upped for those who are not able to access a wifi or have the awesome pull-out keyboard the RAPH/TP has
[R^7Z EDIT]Keep in mind, I have a RAPH110/TP/at&t Fuze
Free for all!
n-Joie! (Enjoy)
[/R^7Z EDIT]
Please note, we now have a source fix for the keyboard size and it will be integrated in the next testing image.
Not sure why WiFi works better in GB than FroYo but I just disabled and re-enabled like several times, without fail/error. I'm using the packaged build found in this thread and using Glemsom's kernel package for the 14th. If needed, I can supply screenshots.
R^7Z said:
Not sure why WiFi works better in GB than FroYo but I just disabled and re-enabled like several times, without fail/error. I'm using the packaged build found in this thread and using Glemsom's kernel package for the 14th. If needed, I can supply screenshots.
Click to expand...
Click to collapse
No need. I'll let stinebd know, I'm not sure what the difference would be... if he can backport something to Froyo, or if we are just going to be moving forward and focusing on GB. Thanks again for testing!
Thanks alot man this build is great! seems to boot and run faster and smoother then froyo ever did for me...only issues i ran into were not being able to see the calibration which if youtap on the screen it glitches and shows part of it.....but the main thing is i seem to have a problem staying connected to the internet and have to soft reset to get it working again....also airplane mode seems to get stuck once clicked and only a soft reset fixes it
m using this one is there anyway to chnge keyboard cx buttons are to small and my phone doesnot rotate any help
xtremists said:
m using this one is there anyway to chnge keyboard cx buttons are to small and my phone doesnot rotate any help
Click to expand...
Click to collapse
Read the second post. Gsensor doesn't work yet, but there's a fix in the pipeline. Keyboard fix as well, in the meantime... read the second post!!!!
Links and rom updated:
ICS Sources build Image (Not SDK anymore, version 4.0.1_r1)
BIG, HUGE, MASSIVE, THANKS TO PAWITP FROM CM TEAM!!! (specially camera.aries.so, great job!)
FUGU kernel, linux 3.0, integrated voodoo sound and color (big thanks to renaudallard)
Here is the most recent image BETA 2:
Changelog on the second post.
We moved to google code:
http://code.google.com/p/ice-cream-sandwich-sgs/downloads/list
(ATENTION: image comes with a kernel for i9000b, if you have a i9000 international or vibrant, download the correct kernel)
Thanks to ytt3r for the vibrant kernel
Agreat tutorial made by ybinnenweg for installation and common issues (Thanks!):
http://forum.xda-developers.com/showthread.php?p=19563068#post19563068
How to root the rom:
http://forum.xda-developers.com/showpost.php?p=19175375&postcount=628
Github's:
http://github.com/onecosmic
http://github.com/syndtr
http://github.com/koaaN
http://github.com/pawitp
http://github.com/teamhacksung
http://github.com/cyanogenmod
IRC channel:
http://webchat.freenode.net/
##icssgs <- yes, double ## (we switched channel)
What works:
Both cellphones:
Hardware Acceleration
Audio
calls, sms, 3g - no more script needed.
Touch
Wifi
SDCards (internal and external)
Market
Contacts sync
Calendar sync
camera
panorama mode
face unlock
Gps
data usage
Partition schemma is now including datadata back to avoid issues with slow movinand devices
usb support
bluetooth
What does not work:
Camera (video recording)
Phone encryption
I'll be posting updates as soon as i have them.
CREDITS:
Pawitp: ril fixes, hw acc , camera lib hack(big android guy ).
teamhacksung
Cyanogenmod team
Syntr: kernel patches
Galnet : First Release
kooaN: i9000 kernel update
Thearc77 : google apps
WarDrake: mirroring
ytt3r: kernel patching, vibrant and captivate kernels
SGS CM team: ICS patches
Motafoca 3g switch apk
And all the other people that is helping to make this a reality
Cheers
Changelog:
Beta 2:
Adb now works at boot (previously it only worked after the phone full boot was complete)
PIN lock fixed (it should work, i cannot test it since i dont have pinlocked sim)
adb reboot recovery now reboot the phone to the recovery instead of performing a normal reboot
Recovery now allows you to make nandroid backups
Fixed Market download problem
Minor fixes (code cleanup)
Thank you Onecosmic
Enviado desde mi GT-I9000B usando Tapatalk
Subscribed! Thank you for your efforts and for everyone involved.
Sent via my GT-I9000 running CM7 (GB 2.3.7) using XDA premium
super cool onecosmic
*subscribed*
so that I'm not left in the dark about the latest changes via email
great work onecosmic !
keep it up
thanks !
subscribed, thank you for your efforts onecosmic! Wish you all the best
Subscribed. Gr8 work onecosmic.
mix tweaks-http://forum.xda-developers.com/showthread.php?t=1334818
+1, subscribed;
any chance of updating first post with link to github sources?
thanks
Really nice job, thanks a lot for this
Awesome job! No chance of doing a captivate version is there?
support you onecosmic
dunno a lot about kernel works but I can help about the system part.
anyway thank you
Is bluetooth working?
Is it safe to install this image. I just want to see how ICS is.. I may even report any bugs if I find, if that helps. Or should I wait for a stable release. I can't wait to see ICS, that's why
Please be gentle, thank you..
Sent from my GT-I9000 using XDA App
Guys, a couple of clarifications for those who ask by private and here:
This is an ALPHA release, means testing, developers, unstable. Not recommended for people whitout a minimum modding/development/rom playing experience.
I cannot work on other phones for a simple reason, i dont have one to test and the trial and error method is not an option (too many things can went wrong..)
All working features will be added to the first post (if is not there, is not working)
donate link is in the left side below my name (usual location on XDA)
Hope this covers all the questions.
Cheers
The colors problem is fixed? Or still we have the orange pattern?
Looking forward to re-igniting my (now)dormant passion for the good old SGS!
All the best!
I just had a little fun experimenting.
1) can you check the version of the kernel for I9000. Neither touchscreen ril worked when I installed. Swiping didn't work and my sim was never accessed.
2) I would not just try this as it might not work for everyone. I was able to install the system as an update to my cm7 installation. It has a nice update system where it converts contacts and calendar events to the ics standard. After it finished my phone froze and needed a restart. However after booting again, all my apps were still installed and angry birds worked. Looked a bit weird in those colours and swiping wasn't working at that point.
Oh and the colour issue seems to be like an overlay more than anything else. Menus that flicker show correct colours behind them. (If you bring up the power menu, the desktop behind it as it flickers is in the correct colouring). Or maybe the colouring and flickering are linked and it only appears like an overlay. (Fyi, hw acceleration may not help this. I used non hw accelerated roms in the past and there was no flickering on them like there is in this port).
Anyway its pretty exciting to be playing around with ics.
Edit: anyone wanting to test this should do a nandroid backup. I restored mine after testing and its as thought I never had ics installed (aka completely successful). Backup the nandroid on your pc tho. I am sure there will be work done on the sdcard which in theory could get wiped. (And many backups are a great idea)
Sent via my GT-I9000 running CM7 (GB 2.3.7) using XDA premium
Is safe, only do a nandroid of your actual rom/mod.
You must have your actual stock rom (odin one) with your pit, for unistall this rom.
Ps: really sorry for my basic english
richiebm said:
Is it safe to install this image. I just want to see how ICS is.. I may even report any bugs if I find, if that helps. Or should I wait for a stable release. I can't wait to see ICS, that's why
Please be gentle, thank you..
Sent from my GT-I9000 using XDA App
Click to expand...
Click to collapse
drakenabarion said:
I just had a little fun experimenting.
1) can you check the version of the kernel for I9000. Neither touchscreen ril worked when I installed. Swiping didn't work and my sim was never accessed.
2) I would not just try this as it might not work for everyone. I was able to install the system as an update to my cm7 installation. It has a nice update system where it converts contacts and calendar events to the ics standard. After it finished my phone froze and needed a restart. However after booting again, all my apps were still installed and angry birds worked. Looked a bit weird in those colours and swiping wasn't working at that point.
Oh and the colour issue seems to be like an overlay more than anything else. Menus that flicker show correct colours behind them. (If you bring up the power menu, the desktop behind it as it flickers is in the correct colouring). Or maybe the colouring and flickering are linked and it only appears like an overlay. (Fyi, hw acceleration may not help this. I used non hw accelerated roms in the past and there was no flickering on them like there is in this port).
Anyway its pretty exciting to be playing around with ics.
Sent via my GT-I9000 running CM7 (GB 2.3.7) using XDA premium
Click to expand...
Click to collapse
Kernel for i9000 does not have the ril inside, ril is on the last image i've uploaded, you need to download and install the lastest image (first post) and then the i9000 kernel.
Its true that the negative color its a androidGLES problem, but the only working (so far) software GLES is the one that cames with the emulator, so enabling the hardware video chip will fix the color (since it is going to be rendered on real hardware and not in the softGLES)
Cheers
Ok guys, quick and sweet. Pure aosp build with insecure boot, root, busybox, and my CDMATools app. Radios are 100%. SMS/MMS send/receive win. Everything wins afaik.
Other than that have fun. Oh feel free to mirror, tweak, theme, fix, mod, do whatever. If you start a thread with it, gimme a holla.
Note: Once we get our repo right, i'll push the theme engine source. I need some more eyeballs on ActivityThread as the theme change is not applying to running processes.
Version 3 Change Log:
Tmobile Theme Engine! REQUIRES REBOOT FOR THEME TO FULLY APPLY. THAT'S JUST HOW IT IS FOR NOW!
3 way reboot menu in Global Actions
Panel Tile Mods from my buddy kevdliu
His thread is here and his git is here
Google insists December does not exist and i agree for now
Maybe next build December will exist again.
Version 2 Change Log:
Added all the "experimental" 4.2 AOSP binaries. Not sure if there's any real difference in feel.
Excluded SprintExtension.apk as it was fc'ing on boot.
Added Google Location fix credit to Craig Gomez and mmuzzy for shooting it over here.
Cleaned up build props and keep 4.1.1 fingerprint for app and service compatibility.
I think that's it. This will likely conclude my participation here. The Eos project should be up and running soon. Peace.
My personal gapps. Photosphere, swipe kbd, facelock, videochat, I can't find anything that don't work. Built from slim_gapps and misc other sources.
BRD 4.2 Gapps
VERSION 3 Theme Engine AOSP Toroplus
VERSION 2 AOSP 4.2 Toroplus
Download AOSP 4.2 toroplus
Patches and Misc.
Stock AOSP kernel
Version 2 Tether Fix (use old SecPhone)
How to fix radios for 4.2 Toroplus
Ok guys. The root of the problem is libsec-ril_lte.so. This library was compiled against either 4.1.1 or 4.1.2. Even the library on AOSP for 4.2 was compiled against 4.1.1 or 4.1.2. I found the problem when running a radio log at boot. Turns out the system library linker could not load the library as it was missing a linked symbol. That missing symbol was actually a ScreenshotClient call to SurfaceComposerClient. So I diffed the 4.2 version against the 4.1.2 version. Low and behold, one of the constructors was removed in 4.2. So, we just add the missing constructor back!
Code:
frameworks/native/libs/gui/SurfaceComposerClient.cpp
at about line 591 or so, right under the
Code:
ScreenshotClient::ScreenshotClient()
: mWidth(0), mHeight(0), mFormat(PIXEL_FORMAT_NONE) {
}
class declaration, add this method...
Code:
status_t ScreenshotClient::update() {
sp<ISurfaceComposer> s(ComposerService::getComposerService());
if (s == NULL) return NO_INIT;
mHeap = 0;
return s->captureScreen(0, &mHeap,
&mWidth, &mHeight, &mFormat, 0, 0,
0, -1UL);
}
Also, we have to declare the method in the header as well...
We move to
Code:
frameworks/native/include/gui/SurfaceComposerClient.h
At about line 166 under the class declaration
Code:
class ScreenshotClient
{
sp<IMemoryHeap> mHeap;
uint32_t mWidth;
uint32_t mHeight;
PixelFormat mFormat;
public:
ScreenshotClient();
// frees the previous screenshot and capture a new one
include the following method
Code:
status_t update();
Now run a fresh build. Your libsec-ril_lte.so will load up and your radio will light up like a Christmas Tree. I think the real question here is, "Why is a radio library require linking to the screenshot call?" I don't know, but I know that is what the problem was.
Thanks as always :good:
Great job BRD!
Sent from my Galaxy Nexus using Tapatalk 2
OK. Good job man. Giving it a whirl
Sent from my Galaxy Nexus using xda premium
YES Finally i can stop scratching my arm like pookie .......Ahhhhhh relief .....lol
I'll wait until someone reports back. So please post your results!
Sent from my Galaxy Nexus using XDA Premium HD app
Win!
I need to go try this once I get home!
flash flash flash!! data works!
edit/ only issue ive ran into is it seems gps is not working correctly. Google now says "location service is off" even though it is on. im thinking its a google now issue because maps gets a lock on location.
edit2/ my issues have been resolved with version 2. thank you.
Just double checking, included GApps or no?
Sent from my Galaxy Nexus using Tapatalk 2
Mirror
Mirror if anyone needs: http://www.multiup.org/fichiers/download/d6a7d19e86f44fbde7c4c276f582a16e_full_toroplus-ota-eng.bigrushdog.zip
Hopefully it downloads faster than it uploads.
backup up now so I can flash...Will this have the gapps included?
nacho945 said:
backup up now so I can flash...Will this have the gapps included?
Click to expand...
Click to collapse
Doesn't look like it.
THIS IS WHY I BOUGHT A NEXUS!!!
Hell to the yes!!
Already on the 3rd upgrade
jesusice said:
Doesn't look like it.
Click to expand...
Click to collapse
Where can I find a set of Gapps that should work with this rom?
nacho945 said:
Where can I find a set of Gapps that should work with this rom?
Click to expand...
Click to collapse
I haven't got a chance to flash this yet but I found these which are supposed to be for 4.2. http://d-h.st/wBX
EDIT: Icebluemale30 says they suck, go down a couple more posts and try the other link.
nacho945 said:
backup up now so I can flash...Will this have the gapps included?
Click to expand...
Click to collapse
GApps not included. Just flashed and found that out lol. Ah well.
Super Duper!!! I'll give them a shot, In the process of flashing the rom right now.
jesusice said:
I haven't got a chance to flash this yet but I found these which are supposed to be for 4.2. http://d-h.st/wBX
Click to expand...
Click to collapse
I used the ones in this link, simply because this is what the Team SyEnce build had linked to: http://d-h.st/SrU
I have a weird issue with many custom roms
There is a big black horizontal overlay bar that fickers where I put my thumb
And both bottom and top side of screen have a halo melted cheese effect if I push to far up or down
Here is a video about it (sorry for quality, just shot it before going to work this morning)
http://youtu.be/1AQQ5XLTHQM
Roms with issue:
CodeName Lungo
Mokee
CM11
Nameless
Rom Ok
Team EOS Revival
Each was installed like this
4wipe=>Install Rom=>Factory Reset=>Reboot
Recovery is TWRP 2.8.0.1
Any idea how can I fix this ?
Thx in advance
Same issue here too. CM11 M8 snapshot is fine and so is AICP Sept 9 Release version. Problem seems to be after the sept 20 cm nightly. Issue occurs on all cm based roms. I have not seen a fix for this yet. Besides i have decided tonot use nightlies anymore. Will wait for next aicp release.
Sent from my Find7 using Tapatalk
Not a fix, but a workaround for this bug :
Go to Developer options and check "disable hardware overlays".
Works, but you have to re-check it every time you restart the phone.
I had these problems too, but this worked for me, permanent (unless you update, because then the build.prop get's overwritten). But be careful with your build.prop, so make a backup first.
jee'sgalaxy said:
There are time when people get screen flicker problems when flashing a new ROM. This either has to do with the ROM itself, for the kernel. Until full support is givien, I have a temporary fix. This fix is basically disabling HW overlay, but disabling HW overlay in the settings does not stick on reboot and this has been really annoying.
For those who want to know what Disabling HW overlay is, it is basically making the phone's gpu render the graphics instead of the cpu. This allows for better graphical rendering of everything. This is not a proven statement, but theoretically, disabling HW overlay, may increase battery consumption... not by much though.
Finally, I am not responsible for anything that goes wrong with your phone when using my method. Please don't whine that it did not work or I messed your phone up. You are doing this at your own risk!!
My fix
Step 1 - Download a build.prop editor app from the play store (I recommend JRummy's one) and run the app. MAKE A BACKUP OF ORIGINAL build.prop!!
Step 2 - Locate these four lines
a) debug.mdpcomp.maxlayer
b) debug.composition.type
c) dev.pm.dyn_samplingrate or dev.pm.cpu_samplingrate (you may not find these values; however, if you do, change the "dyn" or "cpu" to gpu) --> dev.pm.gpu_samplingrate
**If you cannot find this value add "dev.pm.gpu_samplingrate"
d) debug.sf.hw
e) debug.egl.hw
Step 3 - Edit the values of each one
a) debug.mdpcomp.maxlayer = 0 or 1 or 2 --> READ EDIT BELOW
EDIT* --> This value can be either 0 or 1 or 2. I put mine on 0 right now and I feel like I'm getting better performance. Before, I put it at 1 or 2; so, you can try it with 2 first.
b) debug.composition.type = gpu
c) dev.pm.gpu_samplingrate = 1
d) debug.sf.hw = 1
e) debug.egl.hw = 1
f ) If you are running CM10, go to setting --> performance --> surface improvement --> disable dithering (AOKP has no option for this)
Make sure to save after each edit. The saved edit will not appear changed until you reboot your device.
After rebooting, make sure to see the changed values.
Thanks to everyone who has given me info of their device. Thanks to davietr for his input on including the "dev.pm.gpu_samplingrate" value
If this helped and work, Thanks would be appreciated.
EDIT **If it's not too much of a hassle and my fix methods worked for you, please post what ROM/Kernel you are using on this page to help others!!**
Click to expand...
Click to collapse
Seems like CM broke something, and everyone that tracks CM got broken too.
The four projects I know of that are not CM-based are:
EOS (as you've mentioned)
Gummy (they track CM for a lot of stuff, but not everything, dhacker did his own kernel rebase for example)
Omni (We track CAF on our own without doing CM's AOSP+CAF frankendisplay crap, and did our own kernel rebase)
Dirty Unicorns (uses Omni's hardware support stuff)
Temporary workaround for this is to enable developer options and remove the check boxes from Force GPU Rendering as well as Deactivate Hardware Overlay.
Anyone with UART or USB access to u-boot willing to help with a dev project?
Devs only ATM, too many serious bugs
And there is the possibility of screen damage
If it hardlocks with papyrus ic on
test build #1 pulled due to poweroff/screen burnin bug
Test build #1: http://forum.xda-developers.com/showpost.php?p=60336716&postcount=14
Test build #2: http://forum.xda-developers.com/showpost.php?p=60453540&postcount=74
Test build #3: http://forum.xda-developers.com/showpost.php?p=60677266&postcount=136
Kernel Tree: https://github.com/kfazz/android_kernel_bn_zoom2/tree/nook_nst_3.0
zoom2 device tree: https://github.com/kfazz/android_device_bn_zoom2
vendor tree: https://github.com/kfazz/propietary_vendor_bn.git
Local manifest:
Code:
<?xml version="1.0" encoding="UTF-8"?>
<manifest>
<remove-project name="CyanogenMod/android_build" />
<remove-project name="CyanogenMod/android_frameworks_base" />
<remove-project name="CyanogenMod/android_frameworks_native" />
<remove-project name="CyanogenMod/android_bootable_recovery" />
<project path="frameworks/base" name="kfazz/android_frameworks_base" revision="nook"/>
<project path="frameworks/native" name="kfazz/android_frameworks_native" revision="nook"/>
<project path="bootable/recovery" name="TeamWin/Team-Win-Recovery-Project" revision="android-5.1"/>
<project path="build" name="kfazz/android_build" revision="nook"/>
<project path="device/bn/zoom2" name="kfazz/android_device_bn_zoom2" revision="master"/>
<project path="kernel/bn/zoom2" name="kfazz/android_kernel_bn_zoom2" revision="nook_nst_3.0"/>
<project path="vendor/bn" name="kfazz/propietary_vendor_bn" revision="master"/>
</manifest>
Done:
Kernel 3.0.8 boots
Zoom2 devicetree
Cm11 boots
Twrp boots and installs ROM
Wip:
Ghosting
Ram usage
Suspend
To do:
WiFi sleep of death,
USB(works, but kinda slow) fixed adb gets ~4mb sec mtp gets ~2.8
Integrate epdcontroller in framework (partially done)
Pngcrush grayscale all images done, by adding back custom squisher to android/build
Fastmode,
multitouch, done, using 1.3.1 zforce driver + changes based on http://lwn.net/Articles/572311/
USB host not sure what the limit is but a 100mA usb mouse works without a powered hub, and a 200mA usb stick doesn't
Themeing
Gapps don't work (but too much lowram lag to be practical)
kfazz said:
Anyone with UART or USB access to u-boot willing to help with a dev project?
Devs only ATM, too many serious bugs
And there is the possibility of screen damage
If it hardlocks with papyrus ic on
Done:
Kernel 3.0.8 boots
Zoom2 devicetree
Cm11 boots
Twrp boots and installs ROM
Wip:
Ghosting
Ram usage
Suspend
To do:
WiFi sleep of death, USB, fastmode
Integrate epdcontroller in framework
Pngcrush grayscale all images
Fastmode, multitouch, USB host, etc
Themeing
Gapps don't work
Click to expand...
Click to collapse
Woah, this looks sick! I'd be glad to help you in any way I can. I've got a Nook Touch Glow running a customized version of the 1337 ROM, a modded framework-res.apk, and a modded Settings.apk (see my sig for full specifications). Let me know how I can assist!
By the way, I'm inviting you to my Nook Touch Devs group now, so we can talk in there.
ИΘΘK¡€ said:
Woah, this looks sick! I'd be glad to help you in any way I can. I've got a Nook Touch Glow running a customized version of the 1337 ROM, a modded framework-res.apk, and a modded Settings.apk (see my sig for full specifications). Let me know how I can assist!
By the way, I'm inviting you to my Nook Touch Devs group now, so we can talk in there.
Click to expand...
Click to collapse
if you'd be willing to try compiling it, javifo has a u-boot tree that looks like it supports ttyUSB.
https://github.com/javifo/NST/tree/master/u-boot
I'm not sure if my kernel supports usb because my usb port is broken, but if you can get u-boot access over usb
i'll try to build a kernel and bootscript that will be debuggable over usb.
My current setup is completely self contained on the sdcard and doesn't touch stock partitions except reading device specific data from /rom
kfazz said:
if you'd be willing to try compiling it, javifo has a u-boot tree that looks like it supports ttyUSB.
https://github.com/javifo/NST/tree/master/u-boot
I'm not sure if my kernel supports usb because my usb port is broken, but if you can get u-boot access over usb
i'll try to build a kernel and bootscript that will be debuggable over usb.
My current setup is completely self contained on the sdcard and doesn't touch stock partitions except reading device specific data from /rom
Click to expand...
Click to collapse
I'll give it a go on the weekend when I have some free time.
rendering from userspace
omap3pfb drawing modes:
pageflip mode: (non deferred)
turned on by writing a 1 to
/sys/devices/platform/omap3epfb.0/graphics/fb0/pgflip_refresh
apps that are unaware of bn's eink setup work fine.
this is how stock recovery uses the display, and also bootanimation on stock rom (i think).
Downside: ghosting, no control over waveform used to draw ( that i know of )
'command' mode.
display update is controlled by various private ioctl commands to the /dev/graphics/fb0 file.
ioctl -> * -> EpdService function -> EpdController function
OMAP3EPFB_IO_SET_REGION <- * <- configureRegion <- setRegion
OMAP3EPFB_IO_GET_REGION -> * -> UNUSED
OMAP3EPFB_IO_RESET_REGION <- * <- resetRegion <- resetRegion
OMAP3EPFB_IO_FILL_REGION < * <- fillRegion <- fillRegion
* is jni function calls in libandroidruntime.so
EpdController.epdRefresh writes to /sys/devices/platform/omap3epfb.0/graphics/fb0/epd_refresh
1 = Gray Clear
0 = update
epdDisable writes to /sys/class/graphics/fb0/epd_disable
value is timeout period in milliseconds
by themselves these writes and ioctls do nothing. the magic is in libui.so
stock rom calls OMAP3EPFB_IO_REQ_AREA_UPDATE in FramebufferNativeWindow::queueBuffer()
to update the display. No idea what logic the is used to decide when or whether to update.
Deferred Rendering:
this is turned off in kernel config for stock rom, and theres a typo in one of the ifdefed clauses which
leads me to believe this code path was abandoned early in development of the driver
i got it to work, but it doesn't seem to have many advantages over regular pageflip mode, and it's slow.
more info on deferred io: http://elinux.org/images/2/25/E_paper_Displays.pdf
anyone see any errors or have any insights as to how to implement this on Kitkat's graphics stack?
Oh Wow !!!
is that really a cm11(KITKAT) lock screen on a NST G ???
thanks for trying @kfazz following this thread, look forward to what will materialize out of this GREAT effort. kitkat is known to support low ram devices, hopefully it will accommodate our Nook Simple Touch activity which is only static read centric operations.
thanks for attempting
Any chance of a video? I'm scared that this is just a late April fool's joke! Been waiting for this for so long!
coolius said:
Any chance of a video? I'm scared that this is just a late April fool's joke! Been waiting for this for so long!
Click to expand...
Click to collapse
I to can't believe it. but this is cool. I hope it will be usefull since I thought the cm7 would have the highest version of cyanogenmod to work well on our devices.
coolius said:
Any chance of a video? I'm scared that this is just a late April fool's joke! Been waiting for this for so long!
Click to expand...
Click to collapse
dorfd1 said:
I to can't believe it. but this is cool. I hope it will be usefull since I thought the cm7 would have the highest version of cyanogenmod to work well on our devices.
Click to expand...
Click to collapse
I mean, the source to compile is right there, and he last updated it about an hour ago... Id compile a version to test for myself, but my server that I've used in the past for compiling android has yet to be set up again(new install )
Im honestly amazing that kfazz got it working, major props to him and hopefully this would allow more funtionality to come out of my old nook(s)
Edit: Oh, and this would also open up a wider range of usb devices to plug into the nook, as well as better compatibility for any new apps
---------- Post added at 10:54 PM ---------- Previous post was at 10:46 PM ----------
Also, to kfazz, Ive got an extra nook that doesnt matter if the screen gets damages(a new one is fairly cheap anyways), Id love to help you test out this project. I can also get access to the uart if needed, ill just have to pop open the case and solder on a few points(right? I last read about the access to the uart on the nook many months/year ago and I may be mistaken)
Yes, yes, and yes. UART access isn't strictly necessary, but there are a few issues I want to resolve before release. Mainly the circle animation android draws on loading screens, and auto partitioning sdcard. PS delete system/core/ uncrypt/android.mk and it should build. (Twrp conflict)
Wow, unbelievable we have WIP and moreover kitkat of all, Can't wait to try it out.
switching off all possible animation is best don't want to lose out on valuable battery time.
thanks Guys.
aiamuzz said:
Wow, unbelievable we have WIP and moreover kitkat of all, Can't wait to try it out.
switching off all possible animation is best don't want to lose out on valuable battery time.
thanks Guys.
Click to expand...
Click to collapse
I think the boot animation would have to be disabled.
Trying to be optimistic ...
A LG optimus Pro C660 with 256 RAM and less than 200MB internal storage for apps.
Came across this Lollipop
I believe Kitkat was also attemtpted for this device. but the ROM file is purged and no longer available.
If this device can operate as a full fledged phone, Will our device hold any promise for static reading purposes only ?
super buggy test build #1
super buggy test build #1:
update: pulled due to serious bugs. poweroff may fail with the screen on, buring in an image.
First up, to the best of my knowledge this rom won't melt your eink screen, wipe your sd, or cause your nook to spontaneously combust, but NO WARRANTIES.
With that out of the way, here is a test build. I'm not publishing it because it's done, but rather to encourage others to sync the source and try bulding it.
backup contents of sdcard first
write sd.img to a sdcard in the usual way. (sd_img.zip is attached to this post)
nook_nst_3.0 kernel: (this still has wifi sleep of death bug)
Same thing built with encore_merge kernelno wifi sleep bug, may have other problems, probably usb)
copy rom zip and md5 to nook's media partition
boot from sdcard, mount media
install rom (make sure signature verification is on)
reboot
if it doesn't boot hard reboot by pressing power and home for 9 seconds
"Where's the source?"
See the first post.,
Specifically the kernel in this rom was built from commit 7c944768270ec14736c7977b876a107e0a30b87f (current tip of nook_nst_3.0 branch)
except arch/arm/mach-omap2/board-3621-gossamer.c which was checked out from commit fbd1b8ccbebcb483eed158c5b6d51a4f44bbd2c4 (current tip of encore_merge branch)
"How can I help?"
sync my code and send me patches
sync my code and build it yourself
make a cm11 theme and test it
patch apps to use EpdController
What isn't helpful:
it's buggy. it's slow.
feature x is broken / app x doesn't work.
it's ugly. etc.
"Where're the gapps? I really want gapps, even though i know they will run all the time stealing my precious ram"
try the pico ones, or pico(uni). you probably won't be able to run browser after they're installed
http://www.androidrootz.com/2015/03/download-android-44x-kitkat-pa-gapps.html
Alternative: https://f-droid.org/ has a market app for free/foss software, and lost of reading apps
I'm very excited to see where this is going. I have an app that requires 2.3+ and would love to get it working. I can put up with no Gapps, slow system, etc. Just wondering about a few things:
Ghosting: Does the e-ink screen not properly refresh?
Wifi Sleep of Death: What does this mean?
Cheers!
Kralik said:
I'm very excited to see where this is going. I have an app that requires 2.3+ and would love to get it working. I can put up with no Gapps, slow system, etc. Just wondering about a few things:
Ghosting: Does the e-ink screen not properly refresh?
Wifi Sleep of Death: What does this mean?
Cheers!
Click to expand...
Click to collapse
For Ghosting, its basically the same as the regular OS, theres issues after a while of use, but I havent seen much from a few hours of trying out the build.
And also to kfazz, this build is awesome! Ive been trying to compile the OS, but have been having issues compiling(last time I compiled an android zip was like 2 years ago for my Nook Color), Although I am not done trying yet
Also: I found a small bug(no idea where to go to fix it, but I remember it being like a bug a while back for the CM10 or 11 development build for the Nook Color) where if you you try to go into landscape(by an app forcing it), the x and y values for the touchscreen are flipped, it shouldnt be too hard to change.
kfazz said:
super buggy test build #1:
First up, to the best of my knowledge this rom won't melt your eink screen, wipe your sd, or cause your nook to spontaneously combust, but NO WARRANTIES.
With that out of the way, here is a test build. I'm not publishing it because it's done, but rather to encourage others to sync the source and try bulding it.
backup contents of sdcard first
;
;
;
Click to expand...
Click to collapse
Hey this is really great !!!
but too early for me to try ... not a techie who understands all this stuff.
We have CWM recovery installed and using yattaro&nookie's ROM
Can't this be flashed from CWM ?
Twrp Recovery a must ?
dombeef said:
For Ghosting, its basically the same as the regular OS, theres issues after a while of use, but I havent seen much from a few hours of trying out the build.
And also to kfazz, this build is awesome! Ive been trying to compile the OS, but have been having issues compiling(last time I compiled an android zip was like 2 years ago for my Nook Color), Although I am not done trying yet
Also: I found a small bug(no idea where to go to fix it, but I remember it being like a bug a while back for the CM10 or 11 development build for the Nook Color) where if you you try to go into landscape(by an app forcing it), the x and y values for the touchscreen are flipped, it shouldnt be too hard to change.
Click to expand...
Click to collapse
Hey dombeef, great to hear you complied it and have flashed it on your device.
Looking forward to your feedback on its performance.
Excited !!!
Just remembered -- I have the Glowlight model. Is turning the light on/off and adjusting the level supported? Is it possible to customize the left/right side buttons?
Kralik said:
Just remembered -- I have the Glowlight model. Is turning the light on/off and adjusting the level supported? Is it possible to customize the left/right side buttons?
Click to expand...
Click to collapse
Well @Kralik as you can see its a little too early for that ... i believe this is not even a beta build !!!
am i right @kfazz ?
dombeef said:
...
Also: I found a small bug(no idea where to go to fix it, but I remember it being like a bug a while back for the CM10 or 11 development build for the Nook Color) where if you you try to go into landscape(by an app forcing it), the x and y values for the touchscreen are flipped, it shouldnt be too hard to change.
Click to expand...
Click to collapse
change touch.orientationAware = 0 to touch.orientationAware = 1 in /system/usr/idc/zForce_Touchscreen.idc (device/bn/zoom2/prebuilt/usr/idc/zforce_ts.idc)
Fix tested and pushed
Kralik said:
I'm very excited to see where this is going. I have an app that requires 2.3+ and would love to get it working. I can put up with no Gapps, slow system, etc. Just wondering about a few things:
Ghosting: Does the e-ink screen not properly refresh?
Wifi Sleep of Death: What does this mean?
Cheers!
Click to expand...
Click to collapse
Ghosting: Screen updates in GU mode most of the time, so previous contents of screen are faintly visible.
Wifi Sleep of Death: sometimes after the nook is in suspend for a while, the wifi chip doesn't power back on and toggling wifi from the android settings will cause the nook to lockup in a bad state.