Related
The Zen Kernel Project presents...
The ZenDROID Kernel
----------------------
What is it?
* Kernels for Android devices
* If you have suggestions, patches, fixes, whatever - they will be reviewed and implemented like a "community kernel", and you'll get credit if desired
What's so special about it?
* Idk, it's zen - each one will end up being different somehow, but implementing a few things others have done and implementing what others have not done.
---------------------
ZenDROID 2.6.29 For the EVO 4g (based on HTC's official source)
Changes over HTC official source (See gitweb for the most detail
7/13/2010
- BFS v304 backported to 2.6.29 - Stopped here as it was a little bit of work to get to this point, I will continue through 318 later on only taking changes that will have a significant effect.
- BFS is tuned to be less aggressive with the nice levels, this should fix any lag that other bfs kernels experience, although without this adjustment on DC 3.2.3 my system is lightning fast.
- Remove nice from the workqueue and kthread (for both CFS and BFS)
- toast's audio boost (thanks)
7/5/2010
- Add AOSP atmel touch screen driver (from toast's kernel), but implemented differently. Selecting CONFIG_AOSP will use atmel_aosp.c, which will fix the screen for aosp. The Kconfig option is the first one above general setup, may be needed for some other drivers in the future, but for now only changes the touch screen one.
- Undervolt patches (no OC or anything yet, that will be a separate patch or Kconfig option or whatever) - about .1V less, seems what they're going with around the town
- Simple I/O scheduler. Based on deadline, v(r), and no-op perfect for SSDs or flash discs. I expect it to be ideal for the EVO and other android phones.
- Made lowmemorykiller less aggressive by default
- Some CFS adjustments (no bfs til' i start froyo, we need to research android's nice settings)
Branch: evo4g-src-2.6.29
GIT IS UP!
http://git.zen-kernel.org/?p=kernel/zendroid.git;a=summary
Like I said, don't plan anything earth shattering for the .29 kernel, but there is still more I just got started suggestions/patches are welcome.
DOWNLOADABLE BUILDS
*** Not yet availlable, I have a few things I want to implement like the 30fps hack fix (will also be a Kconfig option), 5 point multi touch (Kconfig option), overclocking (may not be in the first build, not really a priority over the other things)
I have tested this already on DC, the system is very fast and no lag or anything is present. The simple I/O scheduler seems to be working great with no flaws, and there is no touch lag - the system actually feels quite a bit faster. - This is with no 30 fps fix!
This sounds like a really good idea, but what else can you bring to the table other than a well-maintained kernel? Look at the Nexus board, it's ALL kernels, most of which are pretty well maintained. I think it's great that you wanna do this, but you'll have to provide something to differentiate yourself from all the other custom kernels.
Making this a community kernel much like CyanogenMod is on the ROM side would be a good way, like allowing users to submit their own patches that you could review and choose to incorporate (or not ) would be a good way IMO.
Just giving some suggestions, best of luck!
Regards,
Jesse C.
i see you are going for BFS cpu scheduler as the default; i know that on the Nexus One, the BFS was very unstable and the CFS worked like a charm; not sure, if there are enough similarities between the N1 and Evo, but will the BFS also be unstable for the most part?
i had the N1 prior to getting the EVO and tried few BFS kernels and it wasn't as stable as the CFS; BFS ran the phone at a higher temperature and few other things that i can't remember now
EDIT: as for kernels, would like to see Overclocked kernel as well as Undervolted kernels (similar to maybe IntersectRaven's kernel on the Nexus One forum, where he uses AVS and undervolted to 925mV, which was best for N1 from experience)
Welcome! look forward to seeing your kernel!
Josh
Sounds good bro. I plan to start working on custom kernels as soon as the holiday is over so if you need any help just let me know.
Welcome, we need more kernels for CyanogenMod 6.0.
Have fun.
The broken BFS must have been poorly implemented or a really old version or something. I don't foresee any problems, if there are I'll work them out with CK.
It's not entirely about being 100% unique, more so being another option. And options are good in this open source community Although I'm totally open to suggestions.
Like I said, I don't expect the BFS problems of those kernels, but we'll certainly work them out if need be. I have not had a bfs-related bug report in quite a long time. (pre 2xx). I'd imagine the benefits could be quite surprising actually
One of our developers made an i/o scheduler, SIO "simple I/O scheduler"
It's based on no-op and deadline PERFECT for SSD or just non-spinning disks. I imagine it would be perfect for phones, although I'd like to do some heavy testing before pushing that because of how young it still is (although it's very simple so any bugs shouldn't be especially hidden like they may be in say, cfq)
Updated first post, GIT is up and also updated the plans
Ok guys, about ready to use although I've done no testing yet. I AM implementing a zen-sched patch which will allow selection of BFS or CFS, and i have a few more tricks for CFS up my sleeve yet.
We'll get to BFS 318 , a few things in between aren't worth porting but otherwise it'll be mostly complete.
I also added compcache 0.6.2, that's all located in drivers/block (where I used to put it on the older kernels, now it's in staging but oh well)
Updated!
Now we have a proper backported BFS, read the OP!
I flashed a test to my phone over the DC rom i'm running and everything works great! Using SIO and BFS v3xx
bbedward said:
Updated!
Now we have a proper backported BFS, read the OP!
I flashed a test to my phone over the DC rom i'm running and everything works great! Using SIO and BFS v3xx
Click to expand...
Click to collapse
I cant wait for the download thanks for all the hard work
Sweet man! Thanks for the dedication to the project. Looking forward to it
hello guys so i was looking at this project for Motorola Droid X:
http://code.google.com/p/milestone-overclock/
since Droid X has similar OMAP3630 chipset i thought we could try it out for this device...
if it has already been tried out by u guys then sorry for dupe post
anyways the sources available (milestone-overclock-module-1.4.8.tar.gz) were not directly compatible with the i9003 kernel sources (GB_CHN_UPDATE1) so i have modified it slightly.... and now the module gets compiled and it gets loaded (by insmod)... the sysfs and proc interface is active... even the app (MilestoneOverclock148.apk) detects the module correctly... but the changes dont work...
i invite all devs to help out with this...
modified sources are posted here:
https://github.com/DooMLoRD/i9003-overclock
noobs kindly dont spam this thread....
P.S.: Droid X has got overclock upto 1.4GHz with help of this so i am sure we can try little more overclock than 1.1GHz
If you look at the Samsung code in arch/arm/mach-omap2 and /plat-omap, and compare it with code seen for example in the nook color sources (OMAP 3630, see here), there are HUGE differences.
Normally the opp frequency table seems to be hard coded and easy to edit. Samsung on the other hand decided to dynamically assemble it in board-latona.c with info from cpufreq34xx.c (if I overlook that correctly). These differences could be the reason the module does not work.
Further, the line Amit and you changed in clock.c could - but I am not quite sure - actually lead to 10% higher clocks on every opp step. Because what you changed
Code:
- mpurate *= 1000000;
+ mpurate *= 1100000;
is a conversion factor from MHz to Hz. The line
Code:
if (mpurate < 1000)
above seems to be a logical check if the desired clock has been given in Hz or MHz, which is expected to be at max 800 for the 3630. For our 3640 the highest frequency is 1000, which would require the check to be
Code:
if (mpurate < 1001)
or similar, but they might have overlooked this change. If the input is below this boundary, it is thought to be in MHz, and is converted to match the internal logic which works with Hz only.
And two more questions: I experimented a lot with the OC code, and even added two new opps (1100/840 and 1200/865) to my tables. I could select them, and everything including cpufreq scaling tables was correct, but the CPU never was actually clocked above 1000MHz. Do you know why? And did you check if it is with your kernel (compare benchmark values, do not trust any other source, they all lie )?
previously i was also having milestone A853 and by the overclock module it can be overclocked to 1.2 ghz
XDA_Bam said:
If you look at the Samsung code in arch/arm/mach-omap2 and /plat-omap, and compare it with code seen for example in the nook color sources (OMAP 3630, see here), there are HUGE differences.
Normally the opp frequency table seems to be hard coded and easy to edit. Samsung on the other hand decided to dynamically assemble it in board-latona.c with info from cpufreq34xx.c (if I overlook that correctly). These differences could be the reason the module does not work.
Further, the line Amit and you changed in clock.c could - but I am not quite sure - actually lead to 10% higher clocks on every opp step. Because what you changed
Code:
- mpurate *= 1000000;
+ mpurate *= 1100000;
is a conversion factor from MHz to Hz. The line
Code:
if (mpurate < 1000)
above seems to be a logical check if the desired clock has been given in Hz or MHz, which is expected to be at max 800 for the 3630. For our 3640 the highest frequency is 1000, which would require the check to be
Code:
if (mpurate < 1001)
or similar, but they might have overlooked this change. If the input is below this boundary, it is thought to be in MHz, and is converted to match the internal logic which works with Hz only.
And two more questions: I experimented a lot with the OC code, and even added two new opps (1100/840 and 1200/865) to my tables. I could select them, and everything including cpufreq scaling tables was correct, but the CPU never was actually clocked above 1000MHz. Do you know why? And did you check if it is with your kernel (compare benchmark values, do not trust any other source, they all lie )?
Click to expand...
Click to collapse
Yes I know the changes in diff omap kernel sources... Spent a few hrs today comparing milestone/droid x and i9003 kernel sources to get this module complied and loading...
I am sure that the change done by amit is correct, because there is a very prominent change in linpack scores ~18 compared to ~16 which is typical of a 10% overclock...
As for ur other two questions I posted this earlier in the other thread
DooMLoRD said:
Not sure... These omap chips seem to have only 4 bins (300/600/800/1000)... We are currently making the 1000 MHz bin run at 1100mhz... I am not sure if we can add extra bins... I tried adding a lower 125MHz bin, it was shown by setcpu but the device never really went below 300mhz... May be we need to investigate it further...
Sent from my R800i using XDA App
Click to expand...
Click to collapse
P.S.: the chip on this phone is OMAP3630
Sent from my R800i using XDA App
DooMLoRD said:
I am sure that the change done by amit is correct, because there is a very prominent change in linpack scores ~18 compared to ~16 which is typical of a 10% overclock...
Click to expand...
Click to collapse
OK, so the overclock is working. Nice Concerning the mpurate, have a look here. The author is working at TI, so I expect him to be familiar with the code. However, that does not mean we can't use the conversion factor for overclock. It's just not "clean".
DooMLoRD said:
P.S.: the chip on this phone is OMAP3630
Click to expand...
Click to collapse
Yep, you're right. Got that wrong
---------- Post added at 10:50 PM ---------- Previous post was at 10:41 PM ----------
And another idea: For the Nook Color, there is a guy who implemented an interface and an app to change the clocks. It is different from droid-overclock, because he implemented a sysfs interface in the kernel sources. Hope this helps.
http://code.google.com/p/milestone-overclock/
sorry its of no use as yr already checked it out
XDA_Bam said:
And another idea: For the Nook Color, there is a guy who implemented an interface and an app to change the clocks. It is different from droid-overclock, because he implemented a sysfs interface in the kernel sources. Hope this helps.
Click to expand...
Click to collapse
That looks very much like the sysfs interface we added for VDD control on QSD8250/MSM7X30... Should work I think...
Sent from my R800i using XDA App
akashsgpgi said:
http://code.google.com/p/milestone-overclock/
sorry its of no use as yr already checked it out
Click to expand...
Click to collapse
U should be BANNED for spamming.... The link u posted is already there in the second line of the main post...
READ!!!!
Sent from my R800i using XDA App
I made progress with the sysfs interface seen on the Nook. Kernel boots, and the correct rates are displayed under /sys/power/mpu_freq_oppX. I was also able to set the hightest opp to 800 MHz, so that the two highest were both at the same frequency. The setting worked (confirmed with Linpack). But 1100 MHz was ignored (stayed at 1000). Looking into this further.
XDA_Bam said:
I made progress with the sysfs interface seen on the Nook. Kernel boots, and the correct rates are displayed under /sys/power/mpu_freq_oppX. I was also able to set the hightest opp to 800 MHz, so that the two highest were both at the same frequency. The setting worked (confirmed with Linpack). But 1100 MHz was ignored (stayed at 1000). Looking into this further.
Click to expand...
Click to collapse
so wht cpu freq table are u using exactly?
wht i think is we should concentrate on this (atleast for now):
just keep the 4 bins as is (300, 600, 800, 1000)
then try n get the access to these via sysfs (or proc)
see if we can modify them via that interface, say change 1000 to 1100 or change 800 to 900
and then do tests if these work...
if possible make a sysfs (or proc) interface for VDD (voltage control) too...
have u pushed the testing changes... i am working on same thing here... might help to speed things up...
I am currently testing on the master branch, so the branch is "wrong", but this is the commit:
Sysfs interface
Because only underclock works as of now, I am tested setting
Code:
if (mpurate < 2000)
but that didn't help. Now I will define 1100 and 1200 MHz steps in board-latona.c and cpufreq34xx.c to see if this helps.
EDIT: Nope, that didn't solve it. CPU does not run at 1100. Not even 900. Stays at 1000 in both cases. 800 can be forced...
some updates on the overclock module:
we need to search in /proc/kallsyms for:
clk_init_cpufreq_table
cpufreq_stats_update
on our kernel (uc-kernel v04) they are at:
Code:
c005a198 T clk_init_cpufreq_table
c03c5aec t cpufreq_stats_update
these may be different on stock kernel we need to use specific address
I just thought this might be helpful since Optimus black has the same hardware.
joelmonty said:
I just thought this might be helpful since Optimus black has the same hardware.
Click to expand...
Click to collapse
its the same module dude...
these are all based on milestone-overclock module
DooMLoRD said:
some updates on the overclock module:
we need to search in /proc/kallsyms for:
clk_init_cpufreq_table
cpufreq_stats_update
on our kernel (uc-kernel v04) they are at:
Code:
c005a198 T clk_init_cpufreq_table
c03c5aec t cpufreq_stats_update
these may be different on stock kernel we need to use specific address
Click to expand...
Click to collapse
Why these two? Cpufreq seems to be quite happy with the frequency tables. All frequencies are correctly listed, and the highest available for hardware and scaling are correct (say 1100, if I set it). But some "mysterious barrier" doesn't let the cpu clock as high as requested by cpufreq.
XDA_Bam said:
Why these two? Cpufreq seems to be quite happy with the frequency tables. All frequencies are correctly listed, and the highest available for hardware and scaling are correct (say 1100, if I set it). But some "mysterious barrier" doesn't let the cpu clock as high as requested by cpufreq.
Click to expand...
Click to collapse
read the sources of the module it explains why we need to look at those values...
DooMLoRD said:
read the sources of the module it explains why we need to look at those values...
Click to expand...
Click to collapse
Looked into it, and got the module to load and change frequencies by manually setting omap2_clk_init_cpufreq_table_addr=0xXXXXXX. I was able to underclock to 800 and back to 1000 MHz. 1200 was set, but not correctly applied - the mpu was still running at 1000 MHz. Further on, it ****ed up the frequency table. Instead of [300,600,800,1000] it was [600, 600, 1000, 1000] after the test. Not good
XDA_Bam said:
Looked into it, and got the module to load and change frequencies by manually setting omap2_clk_init_cpufreq_table_addr=0xXXXXXX. I was able to underclock to 800 and back to 1000 MHz. 1200 was set, but not correctly applied - the mpu was still running at 1000 MHz. Further on, it ****ed up the frequency table. Instead of [300,600,800,1000] it was [600, 600, 1000, 1000] after the test. Not good
Click to expand...
Click to collapse
yups code needs some more patching but i am sure this is the way forward for stable overclock
I've got the basics working with a completely reworked sysfs interface (no module). See GitHub.
The idea is simple: At all times, there shall be no more than 4 OPPs enabled. For each overclock "wish", we disable the currently highest OPP, and enable the overclocked one. If this has no corresponding OPP yet, we create it.
The code works, and has the following problems / features:
Only the highest OPP can be set for now. Consequently, the overclock has to be higher than 800 MHz.
The voltage is not adjusted, yet. All overclock frequencies are run with the 1 GHz stock voltage.
The cpufreq table, policy and stats are not updated, yet
The cpu does not go into deep sleep after the clocks have been adjusted (possibly because the cpufreq table is wrong)
As OPPs are added to the table if necessary, it is theoretically possible to max out the OPP array by defining new frequencies hundreds of times (depending on the maximum array size).
To set the frequency (in MHz), type
Code:
echo "1100" > /sys/power/overclock_max_freq
It would be really cool if you could take a look, DooMLoRD. The only real problem I see right now is the cpufreq table. If this would be correctly updated, the rest would be "easy" I tried some stuff (not in the commit), but nothing worked, yet.
XDA_Bam said:
I've got the basics working with a completely reworked sysfs interface (no module). See GitHub.
The idea is simple: At all times, there shall be no more than 4 OPPs enabled. For each overclock "wish", we disable the currently highest OPP, and enable the overclocked one. If this has no corresponding OPP yet, we create it.
The code works, and has the following problems / features:
Only the highest OPP can be set for now. Consequently, the overclock has to be higher than 800 MHz.
The voltage is not adjusted, yet. All overclock frequencies are run with the 1 GHz stock voltage.
The cpufreq table, policy and stats are not updated, yet
The cpu does not go into deep sleep after the clocks have been adjusted (possibly because the cpufreq table is wrong)
As OPPs are added to the table if necessary, it is theoretically possible to max out the OPP array by defining new frequencies hundreds of times (depending on the maximum array size).
To set the frequency (in MHz), type
Code:
echo "1100" > /sys/power/overclock_max_freq
It would be really cool if you could take a look, DooMLoRD. The only real problem I see right now is the cpufreq table. If this would be correctly updated, the rest would be "easy" I tried some stuff (not in the commit), but nothing worked, yet.
Click to expand...
Click to collapse
this is the same problem as with other overclocks i was playing with...
the cpufreq table doesnt get updated... only the module based way seems to change that table...
anyways we will have to investigate this further...
oh btw i have found patch to overclock GPU...
DooMLoRD said:
oh btw i have found patch to overclock GPU...
Click to expand...
Click to collapse
Nice. Hehe
Finally after someone pm'd me I looked back into GPU Overclocking.
New thread created issues with old thread OP permissions (people seem to be asking the same questions over and over again all information will be kept in OP & DO READ THE THREAD, repetitive questions will now be ignored)
Benefits:
Smoother UI
Handle 2d & 3d core GPU intensive applications & games
Currently only 2D core has been overclocked working on 3D core OC
2D-core original value - 192mhz OC to 245mhz DONE achieved 25% performance boost grp_2d_clk outputs 245760000hz
3D-core original value - 245mhz OC to 300+mhz WIP hoping to achieve 40-50% performance boost
2D-core OC only Download: Coming soon...
Download Links for other devices coming soon...
Works for all HTC Sense/Cm9/Cm10 kernels (Just ask a kernel developer for your device to implement the source code)
Note: Don't have internet on PC so providing 3 main files that need replaced for 2D-core OC to work
Download link to source code: http://d-h.st/wbH
3D-CORE OC TESTS Download: Coming soon...
Do check under sys/kernel/debug/htc_clock/clks/ look for file with all clocks & look for GRP_clks (Graphics clock)
OK so basically today I've been thinking and I've come to the conclusion that I will release the 2D-core OC patch As Soon As Possible, 3D-core Core OC is NOT Impossible but for now I'll give it a break, I will attempt 3D-core OC If/When I can get a hold of a msm7x30/msm8x55 device, as it will make it much easier for both me & users.
So for now you can enjoy the 50mhz increase/bump up, perf boost 25% in 2D-core (will increase performance in both 2D/3D intensive appications as 2D-core is used for 3D AFAIK and increase User Interface performance (Note: This will not take a hit on battery life)), I will also release a couple of fixes on patchas kernel that shouldnt be there/set etc.
(Theres a device available in my area for roughly £90, if anyone wants to contribute towards getting that device or can donate a device , more than welcome and shoot me a PM so I can list you here.)
(This isnt a promise of 3D-Overclock if you donate, if you donate please do so expecting nothing I will only attempt 3D-overclock)
Working device list - all kernel 3.0+ msm7x30/msm8x55 soc devices.
Main thread is in Desire HD Android Development section: http://forum.xda-developers.com/show....php?t=2368497
rk3188[x7010] - Performance Tweaks, Test Journals, Learning diary, and custom ROM nV 1.03 for x7010
First of all, I wanted to say thanks to LEOLAS of FREAKTAB for all the knowledge I've learned and also for OMA for making his roms PUBLICLY ACCESSIBLE AND FREE!!
PLEASE DO NOT COPY/PAST MY/THIS ARTICLE/POST Without Giving the right credit/Link to this post and XDA Forum too!!!!
I'm A WRITER AND I Dont like PLAGIARISM !!!
although It is fine for me if you will rewrite it in your own words
Oh I will introduce my self first, I am From the Country of [>--] Philippines, though very few developer came from our country, this is because of the LACK of INTEREST / PASSION of most of my countrymen.
I am th3f33, Pronounced as DAFFY or read as The Fee, ayt? xD The real name is Demi francisco, I am a Professional DESKTOP and Laptop Tech, and was used to be my previous work / job. Now I am currently focused on Web Developing and Programming (HTML,CSS,PHP,MySQL,JAVASCRIPT,ETC) that's why I am pretty familiar in the way the ANDROID Language goes (C+,Perl,PHyton,java Ayt?). I AM THE MAD PROGRAMMER!! lol XD (Just like hohouin kyouma).
For the sample/preview of my first cooked rom performance, Here is a YOUTUBE Video for it, stress test in HD Games.... http://www.youtube.com/watch?v=mm1ewtDljuk&feature=youtu.be
Vid is 14 minutes long, it would be better to check this first at least the first 5 minutes, and if you are interested, then you can continue to read.
So the device is:
SKYWORTH SKYPAD X7010
(quickly phased out on Skyworth MAIN Upon the release of x7011)
Full SPECS: (Stock)
Rockchip 3188 Quad Core Processore (28nm, 512kb l2 cache, 1.6ghz)
2GB Hynix DDR3 SDRAM 1600(h5tq4g8) Clocked @ 300mhz idle, 400mhz 1.2v
Hynix 8GB Internal Nand FLASH h27ucg8t2atr
Mali 400MP4 GPU @ 600mhz
Wifi and BT By mediatek (MT5931 and MT6622BT)
UP TO 64GB micro sd support
USB OTG
NTFS SUPPORT
HDMI SOCKET
7", 1280 x800 resolution
Audio is Realtek RT3261
NO GPS, No other sensor except accelerometer
Stock OS JB 4.1.1 (Ported/unperfected rom, missing libs, cannot run NBA2k13 on stock rom 4.1 due to the missing lib)
Facebook Group for my Custom Rom with Links to Download - please join my group, im CREATING an Army XD
https://www.facebook.com/groups/187854081403188/
NV v1.03 Custom rom By: Th3f33 (yey its me!~)
Features:
Rooted
Tweaked for MAXIMUM Performance for the price of a little battery timespan.
ANDROID SDK 2.0 4.2.2 (4.3) Based
Incomplete UI (Kitkat Based/Transparency), Not calibrated very well...
Internal storage(system) increased from 1GB to 2GB while internal SD decreased to 4.4gb
Anti-Aliasing x4 Unlocked on Developer Options
Comes with the CWM Recovery (Thanks again to LEOLAS!!! YOU ROCK!! HAPPY NEW YEAR)
Faster Reading And Writing Speed
Nand Flash and RAM Tweaks
GPU Tweaks!
99.7% Fully Functional and Error Free (As far as I tested, until Nov 2013, Im not sure right now, but may not have any problems updating google apps at all)
Bloatware Removed / Reduced
Added Apps:
Root Explorer
Jota+
Android Kitkat 4.4's Office
Android Kitkat 4.4's Google Launcher
Chrome
Rebooter App
PPSSPP (PSP Emulator)
Viper4Android Sound Effects
DOWNLOAD LINK IS SOMEWHERE IN THE MIDDLE of the whole article...
if there are some that i had forgot to mention, please do remind me XD
Instructions.
1.Install Adb Drivers after downloading the zip,
2.enable usb debugging or enter the bootloader mode by turning off the device, then holding both the volume button and power.
3.run the nV flasher, install drivers, if not yet in bootloader mode, reboot now by pressing the reboot on the flasher.
4.clear nandb then flash!
I AM NOT HELD RESPONSIBILITY IF IN ANY CASE YOUR DEVICE GETS BRICKED!
EXCLUSIVELY FOR Skyworth's Skypad x7010 only, but may also work on chuwi v88 [not tested]
I do provide unbricking service for a fee, but if you will read thoroughly this whole post, you will see the keywords to search for to unbrick rockchip rk30 devices!
DETAILS
I do believe that these tweaks / Performance increase / Speedup can be applied to all RK3188 devices, with the same logic / techniques can also be applied to rk3066 to maximize the performance. BUT BEWARE ON THE RAM(MEMORY) CHIP!!!!
Now here are the information on What i Modded and how to Improve the Performance
NOTE: (If the performance of my tablet on the video on youtube is the same as the performance of stock rockchip rk3188 tablets/devices, please do say it to me so that I will be able to change some Description on all of my posts across the INTERNET!! Thank you!)
so.... here is the learning diary and test journal
CHAPTER 1. Full brick at 1st week
The x7010 did not had any kind of update OTA from skyworth, and was replaced quickly by x7011. Since skyworth is a chinese company, i am not really expecting anything from them. but the sad part for me was when i bought the tablet, (About $150 converted currency) and then looked over the internet for supports such as CWM Recovery, CROM, Cyanogenmod or anything else, I FOUND NONE! Great! Not even a Copy of STOCK ROM! Which give me the idea of studying/learning the developing since i had already interested in doing so, and will also make me one of the pioneer custom developer/cooker for our device.... YEY!!... but it was not easy, IT WAS A HARSH PATH....
After my 1st week of trying, I already got a FULL BRICK ON MY DEVICE, But lucky me, im a great actor, so it ended up replaced from the store since it was still covered in the 1week replace 1 year service warranty. (I had softbricked the device by trying to flash Finless rom 1st try, then glassrom 2nd try, then i fully bricked it by flashing with no bootloader & parameter, erase nand db so that the store wont find out that it had been rooted and flashed!)
Chapter 2: Unbricking Success!
So on my 2nd chance, First thing i did was to do a ROMDUMP for backup (NO CWM YET), then I tried to create another CWM Again BUT I FAILED, Where i left it as is with no recovery but working system.... then i posted my 1st post here and in freaktab WHERE I stated that I want to learn and dont want to be spoonfed... Then after a few days, leolas tried to make the CWM!!! but I cant seem to flash it via flash_image, so I used a 1.37 rk tool, to flash the recovery, (Now i know that i messed up the partition offset of the recovery before so it doesnt really boot to recovery!). but by testing out what i understand from the forums, researched alot regarding parameter, mtd partition, the way android system boots and operates, i finally flashed the CWM Successfully on my device! so i backed up trice the device, and since i was so excited, i tried to flash a new rom again, but the excitement and an old mouse that double/triple clicks BRICKED my device by flashing it with NO PARAMETER AND BOOTLOADER, and all other are checked and the flash are already being erased.
No more bootloader mode, no more adb, not detected in PC In any way. then i found out the pin 8& 9 shorting for the nandflash unbrick method, tried it, stripped down my tablet, but i was so sleepy so it resulted in a failure. I dreamed that i will be able to revive the device by retrying it so when i woke up, the first thing i did was to do the PIN SHORTING then i plugged the usb cord, then poof,! Success! MASK ROM MODE FOUND!! then flash my romdump (Backup from cwm with tar and tar.a cannot be flashed ). Although i have one problem, the dump have the system in .tar and i need a .img, so once again, GOOGLE! kept on searching till i stumbled on a forum where an instruction of how to make a System.img flashable from update.zip. Using a Linux mint VM then plus learning basic syntax of linux, got it working ! .
Chapter 3 - BRAVERY!
This scenario made me think that NOTHING CAN STOP ME NOW! since it was the most difficult part for me, especially the unbricking part. so ill make it a little bit short, i flashed many roms, and recorded the differences for every rom,(ALL ROM ARE FOR RK3188, m9, u8gt,mk908, t428, glassrom, m9, etc), some rom displays, some don't, some boots, some doesn't. I also learned to logcat, read logcats, read and debug using dmesg. the rom that had been most compatible for my device is the chuwi rom v88 by oma (Yeah it rocks!) with working wifi, bluetooth, and everything else except for 3 problems, Inverted Camera, Accelerometer not working and NO SOUND! NO AUDIO! WAAAAA but was able to use it still, and checked the performance. run some benchmarks, played some games, although no sound, it was fine, with a few random lagging.... still, i continued to study it and modifying.
Next part was the System UI - framework.res and baksmaling and modfying apks, mostly for the User interface, and those were pretty easy since you can use the TRIAL AND ERROR Technique!. I have already the idea for the configs in build.prop, and just did some trial and error. By doing all of this, continuously, i am learning different syntax & commands too using adb and terminal emulator. so in the end, i maximized the performance of the rom, without touching the kernel. But i still notices the lag, one scenario is this, Fresh boot, Playing NBA2k14, then by some reason, after the camera zoom, the game slows down or get a little lag, about 22 - 28 fps, in order to get back to the max fps gaming, sometime i touch the notification area, sometime pause - resume. and when i am on max fps on any game, then accidentally touched the home / recent button, when i return to the app, it gets so lag! and when playing simultaneous game, it gets really lag.
Start the tweaking
I know this symptom, Applying my knowledge & experience in desktops. I had declared to myself that THIS IS THE EFFECT OF A BOTTLENECK! (Unbalanced specs & performance) or Memory bus width too low therefore bandwidth limits the overall performance so i checked the OC of the device, then found out that it was not overclockable thru software.
Google is my friend, and found out that i need to compile my own kernel that enables the Clock adjustment of the device to exceed the 1.6ghz. I was too naive that I Underestimated RK3188's Power. my initial goal was to OC the CPU itself, but after googling hours and hours, asking my mentor on how to compile a kernel, looking for sources which was the same for our device, eventually i found none xD, then, I stumbled upon.... OVERCLOCKOMATIC / Patchomatic!! a linux app that enables patching kernel's clock settings. And i found out that it can patch ram / gpu / cpu!
Since i was An Experienced Overlclocker (Overclocking on desktop, e5400 (2.7ghz Raised to 3.15ghz), On AIR / STOCK FAN + Exhaust & intake circulation). First i did was a hand made heat spreader (To make this, you need an aluminum foil, fold it into three, with the width of about 1-1.5 inch, then cover most of it with any tape except for the part that will be connected to the target(my target is the cpu), and the other end for the receiver/outer/safe place to transfer heat.) first, my target is my procie and spread it through the sd card slot, and another one from the procie to the aluminum case on the wifi & bt (it was covered!) i moved the 1st end from the sdcard slot to the back cover later on when i realized that i might damage the sd card or the socket itself .
Then I tweaked it and tweaked it for days, checked stability, compared results.
our kernel is 3.0.36 compiled on ubuntu! Yey! at first i was running the device on 1.8ghz, then later on i reverted it back to 1.6ghz, due to heat issues that may damage the device with prolonged use and to save battery spantime too.
But not everything is about the clockspeed, especially when i checked the GPU CLOCK IS ALREADY MAXED, So what would be the cure or solution for the bottleneck....
of course, the logics of tweaking a processor or ram or gpu are the same for the desktop. SO HERE ARE THE MOST IMPORTANT THINGS THAT I LEARNED AND WILL SHARE FOR EVERYONE EVEN IF THEY ALREADY KNEW IT OR HAD IT ON THEIR DEVICE ALREADY OR DOESNT hAVE A CLUE!
Chapter 5: OVERCLOCKING? the correct term to it for me was clock correcting / clock tweaking.
This is the EXACT modification of the kernel clock and stepping on both cpu & gpu and ram on my kernel which IS INCLUDED IN MY CROM Where everyone in our country(philippines) with x7010 tried and was satisfied of the great improvement of performance. and last week, some russian had already visited our facebook group and was testing my rom already . they also shared some roms of their own, although i think they are ported too.
nV 1.03 [first Edition] by th3f33 Download Link:
http://bagongpleion.net/th3f33/nv103.rar
RAM
modifying the RAM (2g DDR3 sdram 1600mhz) to be run @800mhz like a ddr2 instead of the stock @400mhz like ddr1 thus increasing the bandwidth have a huge impact on MALI-400mp4 gpu since the Quad Core Rockchip Processor were fast enough to make up for the lost clock cycle but lacks bandwidth because of the 32 bit bus width. BEWARE! Our ram is 2gb ddr3-1600mhz, which have a normal voltage of 1.5v. Low profile sdram modules have only 1.2 / 1.3v normal running voltage. YOU WILL NEED TO MANUALLY CHECK THE SDRAM MODULES ON THE DEVICE ITSELF TO IDENTIFY What type of ram it is, is it a ddr3 sdram? ddr2? a low profile? and clock timings? Ive used google to fully identify my hynix ram specs and even tried running the RAM @ 800mhz 1.5v and no restarting / corruption issues! it just that it drains the battery really fast. for our device, the exact and sufficient Voltage for the ram to perform well @ 800mhz is around 1.3v - 1.35v although im using 1.3v only. THIS IS NOT APPLICABLE TO ALL RAM and WILL REQUIRE MOST OF THE ATTENTION, EFFORT AND TIME TO APPLY CORRECTLY AND SAFELY!
I cannot guarantee your life or you freedom !
Some Info on ram: Lower clock speed of ram have faster clock cycle (on default clocks of actual ram sticks) Higher Clock Speed of ram have slower clock cycle BUT Bigger Bandwidth.
Low Profile RAMS Uses less energy(VOLTAGE), expect a slight decrease in overall performance....
Some Rams have lower clock cycle and latency in their stock settings compared to other. This are the high profile / OC'ed version of rams..
Anonymous asked: "Why did you use ddr800 against ddr400 config for our ram?"
RK3188 is a quad core processor with 512kb (L2 cache per core i believe) running @ 1.6ghz which are ULTRA FAST and can be compared to an AMD athlon Mobile processor ! it was so fast that before that RAM Cycles to refresh, all the data have been stored in the ram had been processed already, so there was really no need for the quick cycle, PLUS The GPU Is an Onboard/SOC/bundled along with the CPU (like intel's I series and AMD A Series). Now They will need a Bigger bandwidth so the power of the cpu can be utilized more since more data (RAM IS 2gb, can store a lot of data! Really...) can be passed as the bandwidth is higher. therfore GPU have a wider data path for transferring the rendered graphics data to the cpu to process then pass back to the GPU or ram, dpending on situation, without clogging up the I/O Path.
Short Answer: Faster GPU and Read/Write speed at no cost (aside from a little energy drain for the increased voltage of ram, since the cpu can handle that load!)
Chapter 6: Cpu and GPU
CPU
Kernel patched, using rkpatchomatic / overclockomatic! (linux)
I also reduced the Clock stepping and adjust the voltage of the cpu from:
312mhz @ .9v
504mhz @ .925v
816mhz @ 1v
1008mhz @ 1.075v
1200mhz @ 1.150v
1416mhz @ 1.250v
1608mhz @ 1.350v (this may not be right, if i am not mistaken it was only 1.3v @ 1.6ghz on stock)
(6steppings)
to
504mhz @ 1v
1008mhz @ 1.115v
1416mhz @ 1.250v
1608mhz @ 1.350v
(4steppings)
AND the GPU speed increase by loading the driver module with:
insmod /system/lib/modules/mali.ko (removed mali_dvfs & mali_init_clock=133)
and also changing the GPU voltage and stepping FROM:
133mhz @ .975v
200mhz @ 1v
266mhz @ 1.025v
300mhz @ 1.050v
400mhz @ 1.100v
600mhz @ 1.250v
(6 Steppings)
To
200mhz @ 1.05v
400mhz @ 1.15v
533mhz @ 1.2v
600mhz @ 1.35v /1.3v
(4 steppings)
OR
266mhz @ 1.05v
400mhz @ 1.175v
600mhz @ 1.35v
(3 steppings)
The Steppings Reductions Resulted in a VERY VERY Fast Application Switching (Not using a multiwin rom, but able to run about 6-7 HD games with seemingly no lags on switching). BUT Additional battery power may be consumed because of higher idle values for each settings.
My theory, explanation, and strategy was.....
GPU & CPU Steppings logic are the same, while for the GPU the clock increases, so as their texture and pixel fillrate. But gpu power may not be nough @ 1.25v during heavy loads, that is why i've used the maximum allowed voltage for the gpu, and adjust most of the voltages so that the system will never be short on current especially when switching state.
Wanderer asked: "why did you reduced the steppings and what exactly is that?"
Stepping is the term I used for the clock settings / stop point, It came from intel, they use Speedstep and steppings. as you can see, Most of the time, when not playing hd games or not running so many many apps, The cpu (based on my new config) will either run fine at 500mhz, - 1ghz only. (large bandwidth maximizes the power of the quad core of rockchip). so it wont need to adjust itself to 1.4ghz unless it really needs it, and it will not revert back to 500mhz unless all apps and memory are cleared and closed. (Even if you press the recent apps, it will stay @ 1ghz and will not drop to 500mhz). When running a solo HD game, except for 2013 Q4 HD games like CoD & GTA SA, cpu will be fine @ 1ghz, when running other apps, like browser, music player, etc etc, it will be kept in memory, and the cpu will have to speed up itself so it will be stable @ 1.4ghz, when running multiple hd games, it will run @ max 1.6ghz.
The stepping settings helped the cpu to be STABLE @ certain clock during prolonged usage.
Plus, rk3188 is not a very good at clock switching / stepping adjustment. Some users of rk3188 uses a cpu tweaker and set their clock permanently to 1.6ghz right? but it consumes battery faster. I tried different numbers of steppings and clock config and this was the best result for performance. Well in reality, if you are going to overclock a desktop processor, you will disable the speedstep or clock switching feature because it consume power and energy. imagine that the processor needs to consistently monitor the cpu load and adjust it according to its needs. even a human will have a delay of a few seconds during the adjustment and calibration of the strength, the same logic applies to the processors too, plus if you will add the stress of the switching of the GPU, bugs and lags are generated, sometimes, it takes quite a while for the gpu to adjust its clock, from 133, to 200, then lags a little, adjust cpu, then gpu from 200 - 400, then suddenly you pressed the home button while playing, gpu reverts back to 133mhz which is far from the max, ew, where in my config, you start at 233, making everything seems to be so smooooooth, even if you enable AAx4 in the developer options you wouldnt notice the impact of it in performance, then you can disable hw overlay and enable gpu rendering too to maximize performance! if you want to save battery, then choose the 4 steppings on gpu, if you are a hardcore gamer on tablet, then use the 3 steppings on gpu, if you want more, replace the 400mhz config in the 3steppings with the 533 from the 4steppings.
STORAGE
I have Also increased the Speed if reading / writing in the nand flash by mounting mtd partitions with:
ro noatime nodiratime noauto_da_alloc data=writeback barrier=0 nobh
TIP {
instead of using an app to mount the android folder in external sd, you can use a terminal emulator, then this command to mount the folder from exterlan sd - internal, so you can play more games! (Note: please do create the /Android/data and /Android/obb on both internal and external SD
$ su mount -o bind /mnt/external_sd/Android /mnt/sdcard/Android
or
# mount -o bind /mnt/external_sd/Android /mnt/sdcard/Android
to remove the mount/bind
$su umount /mnt/sdcard/Android
or
# umount /mnt/sdcard/Android
The mount / bind is not permanent though, and you will need to enter the command every restart of the device.
}
end of message...
Elp Psy Congroo
XDA:DevDB Information
nV v1.03 [1st revision], a ROM for the Android General
Contributors
th3f33, th3f33
ROM OS Version: 4.2.x Jelly Bean
ROM Kernel: Linux 3.0.x
ROM Firmware Required: ANY x7010 firmware / softbricked
Based On: AOSP
Version Information
Status: Stable
Current Stable Version: v1.03
Stable Release Date: 2013-11-14
Current Beta Version: v1.00
Beta Release Date: 2013-09-09
Created 2014-01-07
Last Updated 2014-01-07
Reserved
ROM Screenshots, And Benchmarks:
Passmark Results:
{
"lightbox_close": "Close",
"lightbox_next": "Next",
"lightbox_previous": "Previous",
"lightbox_error": "The requested content cannot be loaded. Please try again later.",
"lightbox_start_slideshow": "Start slideshow",
"lightbox_stop_slideshow": "Stop slideshow",
"lightbox_full_screen": "Full screen",
"lightbox_thumbnails": "Thumbnails",
"lightbox_download": "Download",
"lightbox_share": "Share",
"lightbox_zoom": "Zoom",
"lightbox_new_window": "New window",
"lightbox_toggle_sidebar": "Toggle sidebar"
}
Antutu:
Rom:
Reserved
about:
if there are any question, or criticism, or bugs, or anything else, feel free to post, i will entertain anyone :good:
th3f33 said:
if there are any question, or criticism, or bugs, or anything else, feel free to post, i will entertain anyone :good:
Click to expand...
Click to collapse
can you provide a version that have a default Android icon on the NavBar? [Back][Home][Switch]
Hey TH3F33! Its me Patrick! Hehehehe, nice you're already starting to spread your name and our tab, I hope this becomes successful and well known, will always have your support!
Edit: They really should make a forum for our tab.
Sent from my XT890 using Tapatalk
samplermonhar said:
can you provide a version that have a default Android icon on the NavBar? [Back][Home][Switch]
Click to expand...
Click to collapse
Yes, maybe I will do that maybe tomorrow, I just have to finish another task from my work.
The nV 1.04 will be Philippine themed
re
Jabbzz said:
Hey TH3F33! Its me Patrick! Hehehehe, nice you're already starting to spread your name and our tab, I hope this becomes successful and well known, will always have your support!
Edit: They really should make a forum for our tab.
Sent from my XT890 using Tapatalk
Click to expand...
Click to collapse
Tnx
I was already creating our own forum, just be patient, its a virtue. :good:
Why are you using xt890 ? Lol
th3f33 said:
Tnx
I was already creating our own forum, just be patient, its a virtue. :good:
Why are you using xt890 ? Lol
Click to expand...
Click to collapse
No I mean a category here in XDA..hehehe..
XT890 is my smartphone, the one I use for texting, blah2. Motorola RAZR i
Sent from my XT890 using Tapatalk
Jabbzz said:
No I mean a category here in XDA..hehehe..
XT890 is my smartphone, the one I use for texting, blah2. Motorola RAZR i
Sent from my XT890 using Tapatalk
Click to expand...
Click to collapse
we alreay got a place, check facebook group -.-
Looks like only a few skyworth users visits xda....
Sent from my x7010 using Tapatalk
Wait a moment, is there a 4.3 rockchip rom? The blobs it is using are from 4.2?
Edit: sorry, I just saw this: ANDROID SDK 2.0 4.2.2 (4.3) Based
jsevi83 said:
Wait a moment, is there a 4.3 rockchip rom? The blobs it is using are from 4.2?
Edit: sorry, I just saw this: ANDROID SDK 2.0 4.2.2 (4.3) Based
Click to expand...
Click to collapse
yep it is a 4.2.2 based on 4.3
two thumbs way up for you kabayan! I'm no coder, but I love to read and tweak usually consolidating tweaks/scripts for my myPhone Vortex.
Good thing its a fairly popular phone in India, else I would have been stuck with the stock rom.
I'm curious if your rom can be applied to the recently released tablet from Starmobile : Engage 8 quad+
It sports the same processor as your Skyworth but with 1G of RAM only, 8G internal. Been looking around and I've read through
the list of roms you've tried (PIPO, M9, GLASS) and its a good thing I found your thread. I would have risked to root and flash
my tablet with any of those roms.
I'm currently reading on rooting my tablet using this method below, however I'm now hesitant to proceed. Would you have another method that
would work? or at least give me a little more confidence of root success. Thereafter, its CWM, then flashing time! woot
http://www.slatedroid.com/topic/88850-rooting-rk3188-tablets/
btw, I've found the manufacturer of this tablet, guess Engage 8 is a rebrand of this unit.
http://www.google.com.ph/url?q=http...Q4BrZw&usg=AFQjCNEsT9Dux2uMCLEZykSB5umoY8nyOQ
{
"lightbox_close": "Close",
"lightbox_next": "Next",
"lightbox_previous": "Previous",
"lightbox_error": "The requested content cannot be loaded. Please try again later.",
"lightbox_start_slideshow": "Start slideshow",
"lightbox_stop_slideshow": "Stop slideshow",
"lightbox_full_screen": "Full screen",
"lightbox_thumbnails": "Thumbnails",
"lightbox_download": "Download",
"lightbox_share": "Share",
"lightbox_zoom": "Zoom",
"lightbox_new_window": "New window",
"lightbox_toggle_sidebar": "Toggle sidebar"
}
Floppy is kernel for all custom rom based on LineageOS 14.1, support the Moto G 2014 LTE (4G)
Code:
#include <disclaimer.h>
* Your warranty is now void. *
* No guaranteed this'll explode your device, be careful, follow the instructions ! *
*don't point at me if your alarm dies after flashing this kernel ! *
Based on Linux Kernel 3.4.113
Based on LineageOS Kernel
GCC Linaro 7.2.1
Governors: yankdemand, yankactive, smartmax, lionheart, impulse, dancedance, BioShock, zzmoove, intellidemand, intelliactive, Interactive, Conservative, Ondemand, Powersave and Performance.
I/O Schedulers: Zen, Noop, Sio, Sioplus, Tripndroid, vr, bfq, row, fiops and deadline
Compiled with -Os
Full Linaro Optimizations ON
Optimized Parameters from L Speed @Paget96
Overclock CPU 1593600Hz
Overclock RAM to 566MHz
Reclaim RAM from loggers
Google Snappy Algorithm
ARM RWSEM Algorithm
S2w and Dt2w - credits : showp1984
USB Fast Charge Mode
Spectrum support
Kernel Samepage Merging (KSM)
Dynamic File Sync and Asynchronous FIle Sync - credits : andip71 | Francisco Franco
Frandom - Entropy Generator
USB OTG NTFS r/w Support
SYSInterface Gentle Fair Sleepers
Performance enhancements
Intelliplug 4.0, Alucard Hotplug, autosmp
OC Slimbus frequencies to provide better sound quality
Arch Power saving - credits : kirananto | Cl3Kener
Adreno Idler- credits : arter97 | yarpiin
Sound Control
Support f2fs, exfat, ntfs filesystem
Quick WakeUp Driver and power suspend 1.7 - credits : Paul Reioux | Jean-Pierre Rasquin | Francisco Franco
Support cpu boost
MDSS color control - credits : savoca
Sysfs interface to Enable/Disable CRC
Intelli-thermal v2
Kexec-Hardboot - credits : hurtsky
Sysfs interface wakelocks - credits : Francisco Franco
TCP Congestion Algorithms: Westwood (default), Reno, Bic, Highspeed, Hybla, HTCP, Vegas, Scalable, Lp, Yeah and Illinois.
And much more!
First time:
Reboot to recovery
Flash FloppyKernel.zip
Wipe cache e dalvik cache
Reboot
Update the kernel:
Reboot to recovery
Flash FloppyKernel.zip
Wipe cache e dalvik cache
Reboot
Coming from another kernel:
Reboot to recovery
Flash your ROM
Flash FloppyKernel.zip
Wipe cache e dalvik cache
Reboot
Floppykernel
LineageOS builds for Thea
LineageOS
Mi5Devs
@GreekDragon
XDA:DevDB Information
FloppyKernel, Kernel for the Moto G 2014 LTE
Contributors
Jonas Cardoso, Caio99BR, InsideMuz, Artur Macedo Costa
Source Code: https://github.com/JonasCardoso/android_kernel_motorola_msm8226
Kernel Special Features: A magical support, I am all day here, reading the publications of the posts, ready to help in any doubt (unless you have not read the OP that contains the information that you need).
Version Information
Status: Stable
Current Stable Version: RC1
Stable Release Date: 2017-12-19
Created 2017-12-19
Last Updated 2017-12-19
RC1 - 19/12/2017
Initial release
Frequently asked questions
Q: Will this conflict with any mod I'm using?
A: As long as it doesn't have any entropy related stuff, its okay. Since the kernel already comes with an inbuilt entropy generator, it will conflict with the ones from mod, causing a poor, stuttered performance!
Q: Deep Sleep isnt working for me, Google Services causing wake-locks, what to do?
A: Go to settings, Privacy, Privacy Guard, All apps, long press Google Services, and deny/disable wake-locks! (Thanks to @Paget96 for the solution)
Q: Is your kernel secure?
A: You MAY face some issues, but it's safe! My cell phone has not exploded yet.
Q: WiFi doesnt work for me.
A: Reflash ROM and Kernel.
Q: What application can I use to configure the kernel?
A: Use Spectrum, and if you have enough knowledge of what to modify or do in the application use together Kernel Adiutor.
Q: Can I use the kernel source in my ROM?
A: If you want to use my kernel in your ROM, just put my name in the credit, just like I did.
Q: Does your kernel support writing to the NTFS file system?
A: Yeeh, you can use flash drives and HD with NTFS via OTG.
Q: My cpu clock is locked with any governor.
A: Just disable the thermal control in the kernel temperature settings.
Q: I have problems with your kernel.
A: Take a logcat and kmesg, this website and this video should help you
Q: Why does my magisk and root disappear after installing your kernel?
A: The kernel script makes changes that directly interfere with both, reinstall the magisk that it will function normally, follow this installation line always: FloppyKernel -> Magisk -> Reboot system.
Q: What is the "Battery" profile in the spectrum?
A: Based on Project Zhana & X.A.N.A was made for trying to be as efficient as possible without compromising the performance. Thanks to the discovery on Voltages used by our SD820 I could make a profile that utilizes more the BIG cluster. My first attempt was to move ALL the load to the BIG cluster, it was against anything that I have learned about little. BIG CPU but I tried, it turned really good at first, low usage was just amazing... But on more usage things didn't go that good. So what I did it was to give a twist to things, but not such twist. It kind of works like Nameless and Shotgun, Mostafa Wael and I worked the same idea. #LITTLE cluster will be running on low-mid frequencies but delivers full performance when is needed. # BIG Cluster will help LITTLE ease Load, but in a great way than before. Things can be done faster and more efficient giving us profit in form of battery. It should run smooth on all scenarios, but you guys tell me. You can easily get around 7~9 hours SOT for 24 hours even more If you don't play game. This will, of course, may vary, but it will definitely be better.
Q: What is the "Balanced" profile in the spectrum?
A: Based on GhostPepper was made to uses a quantized, frequency-aligned parametric curve to influence low core clock rates while providing extremely smooth transitions from each clock rate and exceptional battery life. Really well-defined clock speed escalation for a good battery and performance.
Q: What is the "Performance" profile in the spectrum?
A: Based on FusionP this profile brings a more "agressive" way to scale frequencies. It will scale frequencies sooner(regarding load values from cpu) and will use more the big cluster to provide more performance doing the tasks on hands. Battery should not take a big hit but using heavy demanding tasks you can expect a higher drain.
Q: What is the "Gaming" profile in the spectrum?
A: Based on Burnout PR5, this profile's main goal is to provide absolute maximum performance when needed, but lies low quite well. You see, some times I see the CPU hitting some of the high freqs of 1190 while making something as simple as watching YouTube or playing some spotify tracks needlessly, and sometimes I feel it is a bit slower than it should be in things like brute multitasking. Most profiles do aim at bringing a mid-ground to rest on. But with the sensible words of one of my favourite idols Jeremy Clarkson, he says 'Power solves everything' and right he is. By eliminating the mid freqs of the Little cluster and efficiently using some specific frequency steps, you get some power savings by staying on lower freqs as much as possible, and even on brutal tasks you get the job done faster as well, since the CPU will give you all the power to finish the job as fast as it can, and hit back to idle faster.
Q: I want more information about your spectrum profiles.
A: Read this thread.
Hi there.
Is it possible to run this kernel on the LOS 15.1?
Did read that it is for the 14.1, but you know, why not try to brick it?
Have you tested it? Cause the android go version of oreo with a good kernel could really make thea a brand new entry-level phone. With google camera 5 it could even compete with some budget phones from 2017.
Thanks for the kernel, Just flashed it on AOKP Nougat and it's running smoothly.
I discovered Spectrum with your post, amazing app to don't have an headache with kernel tweaking.
However does Spectrum activates a hotplug or does I have to activate it with kernel adiutor or else ?
PS: Spectrum profile's values can't be seen in Kernel Adiutor, I just see Lineage profile values... Also by looking at your spectrum init script I see that you assume that Thea have BIG and small clusters. when I'm trying to `cat` the value for some lines they do not exist.
Thanks a lot for keeping our good old Thea alive.
One more thing.
Could you make the kernel compatible with the oneoplus camera from this post ?
What camera app are you using?
BridouZ said:
Thanks for the kernel, Just flashed it on AOKP Nougat and it's running smoothly.
I discovered Spectrum with your post, amazing app to don't have an headache with kernel tweaking.
However does Spectrum activates a hotplug or does I have to activate it with kernel adiutor or else ?
PS: Spectrum profile's values can't be seen in Kernel Adiutor, I just see Lineage profile values... Also by looking at your spectrum init script I see that you assume that Thea have BIG and small clusters. when I'm trying to `cat` the value for some lines they do not exist.
Thanks a lot for keeping our good old Thea alive.
Click to expand...
Click to collapse
Yes, I got this script from the source of my Mi5, some lines do not change it, because the kernel moto g 2014 does not have the resources
Breno Spangler said:
One more thing.
Could you make the kernel compatible with the oneoplus camera from this post ?
What camera app are you using?
Click to expand...
Click to collapse
For this i need source kernel to put this feature
Breno Spangler said:
Hi there.
Is it possible to run this kernel on the LOS 15.1?
Did read that it is for the 14.1, but you know, why not try to brick it?
Have you tested it? Cause the android go version of oreo with a good kernel could really make thea a brand new entry-level phone. With google camera 5 it could even compete with some budget phones from 2017.
Click to expand...
Click to collapse
In the build that I made available probably does not work because of the need Oreo changes
s2w for thea? Amazing
zitronenmelissa said:
s2w for thea? Amazing
Click to expand...
Click to collapse
So far it only works on titan
Any can test in LOS 15.1?
Unfortunately not booting on my phone
Jonas Cardoso said:
Any can test in LOS 15.1?
Click to expand...
Click to collapse
Tried flashing your new build...dirty flashed and then, because it wasn't working, clean flashed on 28/12 and 29/12 build of LOS 15.1 with TWRP 3.2.1-0. But unfortunately the phone (thea - XT1072) was stuck on boot logo every time. I really would like to see the kernel working on LOS 15.1, because it worked really good on LOS 14.1.
A few questions mate:
So the cpu is clocked to 1.6 Ghz right? Cpu z shows it reaching 1.2 only. Perhaps on performance profile?
How do i enable dt2w?
Got some lags in some animations, for example in settings, going to display, the animation in between is kinda sluggish. Whats funny is, overall the device is smoother! Some settings must be changed perhaps?
Is the permissive Selinux a big deal?
Im using Los 14.1 with your balanced profile. Thanks, happy new year!
Arcline said:
A few questions mate:
So the cpu is clocked to 1.6 Ghz right? Cpu z shows it reaching 1.2 only. Perhaps on performance profile?
How do i enable dt2w?
Got some lags in some animations, for example in settings, going to display, the animation in between is kinda sluggish. Whats funny is, overall the device is smoother! Some settings must be changed perhaps?
Is the permissive Selinux a big deal?
Im using Los 14.1 with your balanced profile. Thanks, happy new year!
Click to expand...
Click to collapse
Use kernel adiutor, cpu z is not very good to get info, dt2w dont work on thea, only in titan, it is normal to have some lags, after all it is an old device already, it also depends on its use, selinux dont will stop the lags, if that's what you're thinking
I'm using in titan, your kernel doesn't lag but battery gets horny (bad joke) I mean that the battery duration got worst, you maybe consider undervolting a little and upgrading your lmk parameters because it does make apps don't lag but it's a little aggressive, it does FC my keyboard (google keyboard) and I don't wanna be annoying but I have a request, can you bake voltage control into the kernel?
---------- Post added at 09:39 PM ---------- Previous post was at 09:26 PM ----------
I can provide sources for voltage control if you want and, where I send logs?
---------- Post added at 10:21 PM ---------- Previous post was at 09:39 PM ----------
The fix for laggy animations when opening recents and for almost the animations is disabling the adreno idler, or tweaking it
If you have any other test builds for 8.1 just pm me so I can test it
I am using Thea. I like your kernel but had to remove it and restore a backup because the battery usage settings was hanging. When I unplugged after charging it still showed up as charging. That I could fix with a restart but then it wouldn't show what apps was used since last charge. Didn't note any significant improvement in battery duration the days I tried out your kernel.
BTW is there a way to remove your kernel without having to install a backup?
Are you still here?
Is this kernel compatible with Android 8.1? I am running LOS 15.1 but I want to use FloppyKernel because I loved it when I was running Nougat.
However, there is a problem. I can't afford to flash it because I am currently using my device a lot and there is no boot.img in your script. I tried to manually repack a boot image with your Zimage and hot-boot it from Fastboot but it stays stuck in the Motorola logo when I do it. I dumped LuK's Oreo kernel and it hot-boots fine. I'll also try to build your kernel from source.
I really hope that you didn't abandon this device too, but if you did then I will try to become a developer for this nice old phone.
Wattsensi said:
Is this kernel compatible with Android 8.1? I am running LOS 15.1 but I want to use FloppyKernel because I loved it when I was running Nougat.
However, there is a problem. I can't afford to flash it because I am currently using my device a lot and there is no boot.img in your script. I tried to manually repack a boot image with your Zimage and hot-boot it from Fastboot but it stays stuck in the Motorola logo when I do it. I dumped LuK's Oreo kernel and it hot-boots fine. I'll also try to build your kernel from source.
I really hope that you didn't abandon this device too, but if you did then I will try to become a developer for this nice old phone.
Click to expand...
Click to collapse
No item didnt work it causes Brock on 15.1 or anthther but 14.1