So I've been traipsing through some of the other Qualcomm repositories searching for tidbits that might be of use for N1 kernels. As far as I can tell, none of these changes have been merged in to the aosp or cyanogen kernels, although unfortunately the underlying kernels have diverged and these won't apply cleanly but it should be possible to manually merge with some effort.
First up, this patch reduces AXI (internal bus) speed when the apps CPU is running at lower clock speeds and increases it when at higher clock speeds. Memory would otherwise be the bottleneck for many applications.
https://www.codeaurora.org/gitweb/q...it;h=6caa6d84d8c687b5f66f5b5ea281183eae8947a8
Code:
msm: acpuclock-8x50: Couple CPU freq and AXI freq.
The memory throughput is directly proportional to the AXI frequency. The
AXI freq is coupled with the CPU freq to prevent the memory from being
the bottleneck when the CPU is running at one of its higher frequencies.
This will cause an increase in power consumption when the CPU is running
at higher frequencies, but will give a better performance/power ratio.
This patch adds core support for AXI clock changes.
https://www.codeaurora.org/gitweb/q...it;h=009d997b1439edf1991e181206c74ee3e943787e
Code:
msm: clock: Add SoC/board independent APIs to set/get max AXI frequency.
Some drivers need to set the AXI frequency to the maximum frequency
supported by the board. Add a MSM_AXI_MAX_FREQ_KHZ magic value that
allows them to achieve that in a SoC/board independent manner.
The following two patches drop the AXI speed when the processor is idle but the screen is on, down to the minimum that will allow the framebuffer to keep driving the screen. Claims to save ~20 mA which would be somewhere around 10-20% at least. Unfortunately these will probably be harder to merge in, as the Chromium kernel uses quite a different mdp/framebuffer architecture. Still, worth trying!
https://www.codeaurora.org/gitweb/q...it;h=fdb01a9945f2ac3a4cc76c507ed0abf5dd5cfb57
Code:
msm_fb: Reduce AXI bus frequency to 62 Mhz from 128 Mhz
Reduced AXI bus frequency to 62 Mhx to save power during idle
screen mode/limited sleep mode.
https://www.codeaurora.org/gitweb/q...it;h=dee586811e34cb24dedc1d0587f3e31f1ba656a6
Code:
msm_fb: Reduce AXI bus frequency to 58 Mhz from 64 Mhz.
Reduce AXI bus frequency further to 58 Mhz for lcdc panels to save
~20mA power during idle screen mode/limited sleep mode.
I can't wait to see what Kmobz, Intersect, and pershoot do with this!
interesting stuff...hopefully one of the many great kernel devs can use this info and find a way to get it to be useful on the N1 and save more battery life
For those interested, if you have a look at pm.c, a lot of the logic seems to be already there:
Code:
/* axi 128 screen on, 61mhz screen off */
static void axi_early_suspend(struct early_suspend *handler) {
axi_rate = 0;
clk_set_rate(axi_clk, axi_rate);
}
The only thing that is missing is the setting of the axi rates along with the cpu frequency. However, I would think that it is going to take a special amount of trial and error to get the rates correct for the nexus one, since they are, after all, different pieces of hardware.
We could try to further reduce the frequency of the AXI bus while the screen is off. AOSP currently does 61MHz, right?
coolbho3000 said:
We could try to further reduce the frequency of the AXI bus while the screen is off. AOSP currently does 61MHz, right?
Click to expand...
Click to collapse
It should be possible to go with Code Aurora's 58 I think?
Yes we could reduce it further. I dont know what impact it will have on stability yet.
I have provided a very quick (and very dirty) patch for scaling the axi rates according to cpu frequency. THIS IS NOT INTENDED TO BE USED UNLESS YOU KNOW WHAT YOU ARE DOING.
It is just a preliminary patch which compiles into a zImage. I do not know if it works (i'll leave that to the experts). Please, always (!!), check the patches and code before running the kernel image.
I tried to follow the convention already existing in acpuclock-scorpion as much as possible. One caveat: I do not know what the return value for a successful clk_set_rate is, and I am only assuming based on the code aura sources (pm.c does not use the return value).
Edit: forgot to add that I think it is a little too much for pm.c AND acpuclock-scorpion to be setting the axi value (if it works at all), and might cause stability issues. Maybe someone might try removing the pm.c code and use acpu-clock only, I dont know....
jazzor said:
Yes we could reduce it further. I dont know what impact it will have on stability yet.
I have provided a very quick (and very dirty) patch for scaling the axi rates according to cpu frequency. THIS IS NOT INTENDED TO BE USED UNLESS YOU KNOW WHAT YOU ARE DOING.
It is just a preliminary patch which compiles into a zImage. I do not know if it works (i'll leave that to the experts). Please, always (!!), check the patches and code before running the kernel image.
I tried to follow the convention already existing in acpuclock-scorpion as much as possible. One caveat: I do not know what the return value for a successful clk_set_rate is, and I am only assuming based on the code aura sources (pm.c does not use the return value).
Edit: forgot to add that I think it is a little too much for pm.c AND acpuclock-scorpion to be setting the axi value (if it works at all), and might cause stability issues. Maybe someone might try removing the pm.c code and use acpu-clock only, I dont know....
Click to expand...
Click to collapse
I've made too many changes to acpuclock to apply this directly. I'll take a look at it later though.
jazzor said:
For those interested, if you have a look at pm.c, a lot of the logic seems to be already there:
Code:
/* axi 128 screen on, 61mhz screen off */
static void axi_early_suspend(struct early_suspend *handler) {
axi_rate = 0;
clk_set_rate(axi_clk, axi_rate);
}
Click to expand...
Click to collapse
True, but the comment doesn't agree with the code itself:
Code:
/* axi 128 screen on, 61mhz screen off */
<snip>
static void axi_late_resume(struct early_suspend *handler) {
axi_rate = 128000000;
sleep_axi_rate = 120000000; // <- 120 MHz...
clk_set_rate(axi_clk, axi_rate);
}
I wonder if we could drop the sleep_axi_rate = 120000000 to something much lower?
Also, the OP patches were for dropping the AXI rate when the screen is on but the processor is idling. (And it's not clear what screen resolution the code in the OP patches was for, but if it's for ChromeOS I presume it's higher than our 800x480, so bus bandwidth for the framebuffer shouldn't be a problem). So no, unless I'm missing something most of the logic in these patches isn't there already as you claim. (If I am missing something, please show me where... I believe it would have to be in the MDP driver somewhere).
hugonz said:
True, but the comment doesn't agree with the code itself:
Code:
/* axi 128 screen on, 61mhz screen off */
<snip>
static void axi_late_resume(struct early_suspend *handler) {
axi_rate = 128000000;
sleep_axi_rate = 120000000; // <- 120 MHz...
clk_set_rate(axi_clk, axi_rate);
}
I wonder if we could drop the sleep_axi_rate = 120000000 to something much lower?
Also, the OP patches were for dropping the AXI rate when the screen is on but the processor is idling. (And it's not clear what screen resolution the code in the OP patches was for, but if it's for ChromeOS I presume it's higher than our 800x480, so bus bandwidth for the framebuffer shouldn't be a problem). So no, unless I'm missing something most of the logic in these patches isn't there already as you claim. (If I am missing something, please show me where... I believe it would have to be in the MDP driver somewhere).
Click to expand...
Click to collapse
I might be wrong, but this is how I understand it. axi_early_suspend is called when the nexus is going under suspend mode, which is like suspend to ram. This happens when there are no active tasks running and the screen is off. As the code says, the rate is set to 0 implying it is going to set it as low as the limits allow (assuming that is 61mhz). axi_late_resume is for the case when the nexus is NOT in suspend. There are still 2 states in this mode, one where the machine is being actively used (browsing web) and one where the machine is idle (such as when we are listening to music or any background task that may or may not be using the screen), but the machine cannot suspend to ram due tasks still actively running.
In any case, experimenting with lower values of axi rates should be done to see if it indeed saves battery.
oooh great thread. I will definitely be looking into this stuff when I get back in town! thanks for the links!
https://www.codeaurora.org/gitweb/q...it;h=8a68421a300878d729991c1db609a541910d2c70
Here is another patch (more up to date -> perhaps easier to merge).
I just implemented the AXI frequency tweak for N1 - I noticed it is also present in HTC Desire kernel
I am testing it now - so far it seems stable.
Attached acpuclock_scorpion.c implementing the AXI tweaks. Basically, AXI frequency now scales with CPU frequency.
Ivan Dimkovic said:
I just implemented the AXI frequency tweak for N1 - I noticed it is also present in HTC Desire kernel
I am testing it now - so far it seems stable.
Attached acpuclock_scorpion.c implementing the AXI tweaks. Basically, AXI frequency now scales with CPU frequency.
Click to expand...
Click to collapse
Do you run this along with a custom kernel and just flash in recovery like an update.zip?
I usually use Cyanogen's kernel as base for my experiments - so this should definitely compile with Cyanogen's 2.6.34 kernels. This acpuclk-scorpion.c has lowered voltages and also support for frequencies up to 1190 MHz
Although you will also need to change few call prototypes in acpuclock.h - it is quite trivial just replace old acpuclk_set_rate() and acpuclk_power_collapse() prototypes with these:
enum setrate_reason {
SETRATE_CPUFREQ = 0,
SETRATE_SWFI,
SETRATE_PC,
SETRATE_PC_IDLE,
};
int acpuclk_set_rate(unsigned long rate, enum setrate_reason reason);
unsigned long acpuclk_power_collapse(int from_idle);
Click to expand...
Click to collapse
That should do it.
Markdental said:
Do you run this along with a custom kernel and just flash in recovery like an update.zip?
Click to expand...
Click to collapse
No, this will have to compiled into the kernel, which I'm doing right now. Thank you so much for doing this Ivan!
Nice to see someone finally got this to work! I've been too lax as of late.
Getting AXI scaling to work was pretty easy - makes me wonder why is it not in the AOSP kernel for N1.
It is definitely in the HTC Desire kernel, and I am sure HTC knows what is good for the hardware they built
Ivan Dimkovic said:
Getting AXI scaling to work was pretty easy - makes me wonder why is it not in the AOSP kernel for N1.
It is definitely in the HTC Desire kernel, and I am sure HTC knows what is good for the hardware they built
Click to expand...
Click to collapse
what change to pm.c if any?
pm.c already has everything in place needed for AXI clock switching.
BIG DISCLAIMER WARNING:
This thread is for all GINGERBREAD ROM's (SENSE), and their RESPECTIVE KERNELS.
THESE KERNELS DO NOT WORK ON ICE CREAM SANDWICH (Or rather, not tested but please don't try)
THE ZIPS in the ATTACHMENTS section are also for GINGERBREAD, and not ICE CREAM SANDWICH
PLEASE READ THIS OPENING POST BEFORE FLASHING!!
PLEASE DONT BRICK YOUR PHONE
Hello everyone,
This is the GSM kernel comparisons thread where I have done some comparisons between each kernel and jotted down their differences.
Basically I ran some tests for gaming, antutu and will do more later such as battery life during idle. I also examine default clock speeds and their voltage, to provide a show difference between kernel voltages (affect battery drain?)
NEW: I added more tests by requests, and also narrow down to better performing kernels in terms of their various departments such as Performance, Battery, Graphics and Overall
The device
HTC EVO 3D [GSM] http://www.gsmarena.com/htc_evo_3d-3901.php
HBOOT 1.49.00.11
Radio 10.55.9020.00U
S-ON (So I use FlashImageGUI.apk) Thanks joeykrim http://forum.xda-developers.com/showthread.php?t=1192527
Click to expand...
Click to collapse
The ROM:
★ ROMeOS3D ★ UNSENSLATED [04/04 [v1.3.4b-FINAL] (-SENSE,+RTL,+BEATS,+BATT) [100% BUGS-FREE] by monx®
Many Thanks! I use your ROM daily.
Click to expand...
Click to collapse
The games
Aftermath XHD (Lighting, Multitouch)
Shadowgun Not THD!(Just intense graphics, FPS)
Where is my water?(Portrait base game, Physics)
NFS Shift 3D *Not on market =.= (3D, Intense Racing)
Click to expand...
Click to collapse
The benchmark apps :
Antutu 2.6
Nenamark 2.2
Smartbench 2012 v1.0.0
Click to expand...
Click to collapse
The sequence of test;
Reboot
Flash Kernel
Clear Cache
Clear Dalvik
Boot up
Settle
System Tuner
Governor Set as Performance (CPU Adjust Min 384Mhz)
System Tuner (CPU Adjust Max 1.51Ghz )
Antutu Benchmark without SD card
System Tuner (CPU Adjust Max 1.18Ghz)
Antutu Benchmark without SD card
-------------------------------------------------------
Added test for Nenamark 2
Added test for Nenamark 2 (Second run)
Added test for Smartbench 2012
The battery test:
Charge to 100%
Leave phone idle completely
Edge is ON
Auto Sync is ON
No mail syncs
Widgets are manually updated, except Weather every 12 hours
Left alone for 5-6 hours (During my sleep)
No special SCREEN OFF profiles or governors or CPU speed changes
Interactive governor 384-1.2Ghz
Click to expand...
Click to collapse
The MOVIE test:
Charge to 100%
Set CPU speed to 384-1180
Set Governor to INTERACTIVE
Ensure mobile data is turned ON
Ensure EDGE is turned ON. (2G) So we still have some sort of daily connectivity.
Ensure there is no UNDERVOLTAGE
Browse to HD movie : FAST 5 (Yes our ROM is Fast and Furious 5x )
Check start time
Open using default Gallery App
Play for 1 hour (check battery drain, if im awake but usually not)
Play for 2 hours & 10 minutes (complete movie)
Check end time
Check logs for battery drain %
Click to expand...
Click to collapse
GPU Drivers
Main page talk can be found here ; http://forum.xda-developers.com/showthread.php?t=1567033
The GPU drivers that "supposedly" increases performance.
Please do a COMPLETE BACKUP before installing. I am not responsible for this
GPU drivers - http://forum.xda-developers.com/attachment.php?attachmentid=967741&d=1332842958
Old GPU drivers - http://forum.xda-developers.com/attachment.php?attachmentid=967742&d=1332842958
There are the GPU drivers from the main page, and not the SAMSUNG-EGL drivers by Leedroid.
13 April 2012
I attached the SAMSUNG-EGL drivers by Leedroid as well, and its so far -
Better performance
Better UI experience
Better FPS in benchmarks
No disadvantages yet... so looks good for all of us
Works on the following Kernels
Anryl 10.07
Leedroid 5.4.0
Unity3b
RLS004
Click to expand...
Click to collapse
FAST CHARGE
Fast charge link ;
http://forum.xda-developers.com/showthread.php?t=1580553 by chad0989
All credits to the developer chad0989, and people who tested, and the people who made the kernels possible
Flash kernel of your choice
Unity3b - http://forum.xda-developers.com/showpost.php?p=24684226&postcount=29 by chad0989
LEEDROID 5.4 kernel - http://forum.xda-developers.com/showpost.php?p=24795813&postcount=48 by ozkaya (S-ON users seem to break WIFI, unable to find out how to fix this)
Anryl 10.08 - http://forum.xda-developers.com/showthread.php?p=24849681 by chad0989 & anryl
Anthrax RLS005 - Infected FORUM
Install Incredicontrol
Turn on [v] FORCE FAST CHARGE
Connect USB cable
Results from me,
http://forum.xda-developers.com/showpost.php?p=24786155&postcount=47
Click to expand...
Click to collapse
Notes
"My ROM is faster than yours in arguement". Yes, I agree all ROM's out there have different readings and probably different glitches, performance, battery life and is also based on what apps you have installed. This is just the baseline because I use the same ROM for all benchmarks, but different kernels.
Numbers might vary , thats why I didnt include SD card test. Class 2,4,6 10 across different SD card sizes, and cluster block size varies. Antutu uses a 300MB file to test your speeds. For me, thats not concrete or consistent.
Give or take , I ran the test multiple amounts of times using same configuration only to see a variance of 50 points on total score, which is negligible
CORE2 is not enabled via Leedroid Tweaks(I prefer battery life). I am convinced explicitly turning on second CORE will greatly improve Antutu benchmarks though. Perhaps?
Voltages affect battery life, indirectly. Adjusting your voltage lower could make it unstable, however save battery. As you can see, the stock kernel voltages are the SAFEST HTC configurations. Lower at your own risk.
By saying glitches in gaming, it usually means it has problems rendering, supporting multi touch, artifacts or just any deformation of 2D/3D. It can be minor, and can be major. I have yet to specify to what degree. NEW: A reboot usually cures this problem and yet to fully test it. Another cause of glitches can be caused by Go Locker. Freezing or uninstalling this app does remove graphic glitches.
I linked direct downloads only if the developer and releaser direct linked it. Its best to visit the thread of discussion and read there for more information. Links might die and get replaced, removed and I will monitor as much as I can
Slide to wake kernel sources were finally implemented in Anryl 10.03 above, and leedroid 5.x.x above. Use Ponury Gesture Mod from [email protected] to get it.
I disregarded Smartbench 2012 (Gaming Index) because they are all similiar, with a variance of 100 points at most between each kernel, and each clock speed whether 1.2Ghz or 1.5Ghz
Battery voltage is between 3.2v (0%) to 4.335mv (100%). HTC seems to report voltage correctly, but not -ma drain. For example, at 4.1mv battery is at 90% and 4.0mv, battery is 80% (consistently)
Click to expand...
Click to collapse
How to tell proper battery %?
This is during drain, and not charging. Charging should always be additional 200mv
4335mv 100%
4000mv 80%
3800mv 60%
3600mv 40%
3500mv 20%
3400mv 0% (near empty)
Click to expand...
Click to collapse
My configuration for daily usage
ENABLE ALWAYS ON MOBILE DATA = OFF
WIFI = OFF
MOBILE DATA = ON
NETWORK MODE = GSM ONLY
LOCATION WIRELESS NETWORK = OFF
LOCATION GPS = OFF
BACKGROUND DATA = ON
AUTO SYNC = OFF (On when you need it)
POWER EFFICIENCY = ON
POWER SAVER = OFF
BLUETOOTH = OFF
DATE & TIME = MANUAL
PHONE FINDER = OFF
BACKUP MY SETTINGS = OFF
AUTOMATIC RESTORE = OFF
ANIMATIONS = ON
DISPLAY TIMEOUT = 60 sec (Personal preference between 30-60 seconds)
BRIGHTNESS = 20 %
MONITORING TOOLS = Mine is Usage Timelines free, CPU Usage Monitor Paid
CPU DAEMON = NO
LATITUDE = OFF
WIFI NETWORK NOTIFICATION = OFF
BEST WIFI PERFORMANCE = OFF Under Settings > Wireless > Wifi Settings > PRESS MENU BUTTON > Advanced
Click to expand...
Click to collapse
MY VDD LEVELS
Yes this is my voltages, I use on every kernel, These voltatges were NOT USED during tests, but after im done with tests and stability checks, I apply these voltages for my daily use
Apply at your own risk.
Is generally
-50uv from anryl 10.07 for 192-1180 (stable for me)
-25uv from anryl 10.07 for 1180-1510 (stable for me)
-0 uv from unity3b 1510 to 1782 (i left it at stock, this is overclocking already and i dont support it, neither wish to lower voltages here to maintain stability)
and imported to Leedroid and Unity3b
I saved this profile, and applied to every kernel and seems fine for me
Apply at your risk!
192000: 750000
310500: 775000
384000: 775000
432000: 800000
486000: 800000
540000: 825000
594000: 825000
648000: 850000
702000: 850000
756000: 875000
810000: 875000
864000: 900000
918000: 900000
972000: 900000
1026000: 925000
1080000: 950000
1134000: 950000
1188000: 975000
1242000: 1025000
1296000: 1050000
1350000: 1075000
1404000: 1075000
1458000: 1075000
1512000: 1100000
1566000: 1200000
1620000: 1225000
1674000: 1250000
1728000: 1275000
1782000: 1325000
Click to expand...
Click to collapse
HTC default kernel
Monx has created a STOCK flashable kernel on his webpage but I am unsure whether its flashable to other ROM's.
I used his package, where he claimed it was STOCK kernel
You would have to register @ http://romeos.monx.biz/ though
Downloads, Android, Romeos Shooter_U, Romeos addons, ROMeOS3D_STOCK_DEFAULT_Kernel_unsecured_signed.zip
CPU SPEED: 1.5GHZ VOLTAGE: 1250mv
CPU SPEED: 1.2GHZ VOLTAGE: 1025mv
Battery: Battery doesn't have % jumps and has proper readings.
Battery Test : 1% for every 60 minutes (average)
Estimated ma : -12ma to -22ma
Movie Test : 100% to 68%
Performance : Good
Gaming: Gaming is OK. No glitches.
Antutu Benchmark at 1.2GHZ without SD
RAM 677
CPU INTEGER 1560
CPU FLOAT POINT 692
2D 294
3D 1106
DATABASE I/O 275
Total (4604)
Antutu Benchmark at 1.5GHZ without SD(n/a)
Stock Kernel cannot be "overclocked" to 1.5Ghz
RAM N/A
CPU INTEGER N/A
CPU FLOAT POINT N/A
2D N/A
3D N/A
DATABASE I/O N/A
Total (n/a)
Click to expand...
Click to collapse
Leedroid 5.4.0
by Leedroid
Download: http://leedroid.ftp.rshost.eu/Shooter-Kernels/LeeDrOiD_Evo-3D-GSM_V5.4-FINAL-GB-KERNEL.zip provided by Leedroid
CPU SPEED: 1.5GHZ VOLTAGE: 1175mv
CPU SPEED: 1.2GHZ VOLTAGE: 1050mv
Battery: 99-90% seems to drain quickly, then 90 below to 0 drains slowly and normal. However, during use, the system is really responsive, and battery drain is slow being consistent. I noticed that the system reports the wrong % of battery as well, making occasional jumps between %. I evaluated use for a whole 12 hours (9am-9pm) and battery was at about 30% remaining. Pretty decent since it was out most of the time replying whats app
Battery Test : 1% every 55 minutes during idle.
Estimated ma: -12ma-29ma
Movie Test : 100% to 64%
Performance : Feels really as the fastest, exiting and entering apps. Probably the fastest in my observation and gut feel. However for this performance, I believe some battery life was compromised. Im not sure but its definitely apparent.
Gaming : So far no issues
Issues : Setting 1.89Ghz immediately causes freeze. Well thats expected
Antutu Benchmark at 1.2GHZ without SD
RAM 790
CPU INTEGER 1810
CPU FLOAT POINT 789
2D 294
3D 930
DATABASE I/O 225
Total (4838)
Nenamark 2 : 25.3 FPS
Nenamark 2 w/Samsung-EGL drivers : 36.9 FPS
Smartbench 2012 : 1183
Antutu Benchmark at 1.5GHZ without SD
RAM 861
CPU INTEGER 2007
CPU FLOAT POINT 869
2D 296
3D 1224
DATABASE I/O 325
Total (5582)
Nenamark 2 : 31.6 ~ 31.7 FPS
Nenamark 2 w/Samsung-EGL drivers : 39.9 FPS
Smartbench 2012 : 1964
Click to expand...
Click to collapse
Unity3b
by M-deejay
Download : http://www.multiupload.com/NX8BV82PQW
or on his thread posted here ; http://forum.xda-developers.com/showpost.php?p=22560256&postcount=984
Used to be very famous on HD2 as well! Many thanks!
User input : -75mV stable by neol23, here
User configuration for Unity3b by neol23 : http://forum.xda-developers.com/showpost.php?p=21880484&postcount=4514
CPU SPEED: 1.5GHZ VOLTAGE: 1200mv
CPU SPEED: 1.2GHZ VOLTAGE: 1087.5mv
Battery: Seems to be great, but only had 2 days to test.
Battery Test : 1% every 80 minutes during idle.
Estimated ma: -9 to -15 ma drain. Second tests proves this is very consistent, and good. Kernel enters "deep sleep" rapidly
Movie test : 100% to 69%
Performance : Decent and responsive. Other kernels are a tad faster, but its not even noticeable
Gaming: Slight bugs in games. Glitches. Artifacts occurred. Multitouch issues.
Antutu Benchmark at 1.2GHZ without SD
RAM 726
CPU INTEGER 1590
CPU FLOAT POINT 689
2D 295
3D 1107
DATABASE I/O 320
Total (4727)
Nenamark 2 : 25.2 FPS ~ 25.3 FPS
Smartbench 2012 : 1422
Antutu Benchmark at 1.5GHZ without SD
RAM 807
CPU INTEGER 1791
CPU FLOAT POINT 790
2D 295
3D 1092
DATABASE I/O 350
Total (5125)
Nenamark 2 : 26.5 FPS ~ 26.6 FPS
Smartbench 2012 : 792-802 (I ran this a few times, and yet got this low score) Puzzling...
Click to expand...
Click to collapse
Anryl 10.07 and Anryl 10.08
by Anryl
Download 10.07 : http://www.4shared.com/zip/sOw0Fwbx/akernel_v1007.html
or
Download 10.08 : http://www.4shared.com/zip/wV5CKc2u/akernel_v1008.html
He is continuing the work of Kamma
Recompile and uses latest sources.
10.08 is an extension to 10.07 with FAST CHARGE.
For Kamma's kernel results, look at Post 3 (The Vault)
For Anryl kernel old results, look at Post 3 (The Vault)
CPU SPEED: 1.5GHZ VOLTAGE: 1125 mv
CPU SPEED: 1.2GHZ VOLTAGE: 1025 mv
Battery: Seems to be amazing, without Ponury Gesture during light/medium use. No erratic battery drops. Definitely can last a full day on EDGE. Observed that the kernel gets stuck at MIN CPU speed for 7 seconds after waking up from DEEP SLEEP. OC DAEMON is supposed to fix this.
Battery Test : 1% every 60 minutes during idle. (Updated and second test done)
Estimated ma : -14 to -19ma estimated drain
Movie test : 100% to 67%
Performance : Feels really fast but comes in 2nd compared to Leedroid 5.3.8
Gaming : OK, better than anryl 10.06. Smooth and decent all around.
Issues : CPU speed gets stuck at MIN FREQ for about 7-10 seconds after waking up from deep-sleep. This is where OC-daemon is supposed to fix, hence its a known issue in most kernels. Only Smartassv2 fixes this 50% of the time, and Smartassv2 is not stable for our EVO3D
Recommended Configuration
-50mv by Jcsy (stable 1.2 and 1.5)
http://forum.xda-developers.com/editpost.php?do=editpost&p=23570246
-62.5mv by Travex_
http://forum.xda-developers.com/showpost.php?p=23520287&postcount=1024
Battery can last thru 12 hours easily, pushing 20 hours as well.
Antutu Benchmark at 1.2GHZ without SD
RAM 692
CPU INTEGER 1594
CPU FLOAT POINT 692
2D 296
3D 1153
DATABASE I/O 215
Total (4642) down 4 points from 10.05 kernel
Nenamark 2 : 28.0 FPS ~ 28.1 FPS
Nenamark 2 w/GPU drivers : 33.7 FPS
Nenamark 2 w/Samsung-EGL drivers : 35.6 FPS
Smartbench 2012 : 1226 (Up from 1199 on anryl 10.05)
Antutu Benchmark at 1.5GHZ without SD
RAM 789
CPU INTEGER 1813
CPU FLOAT POINT 798
2D 296
3D 1208
DATABASE I/O 360
Total (5630)
Nenamark 2 : 28.1 FPS ~ 28.3 FPS
Nenamark 2 w/GPU drivers : 36.1 FPS
Nenamark 2 w/Samsung-EGL drivers : 35.6 FPS
Smartbench 2012 : 1399
Click to expand...
Click to collapse
GSM-SENSE-AnthraX-RLS005
by chad.goodman
Wrongly accused of stealing peoples code for other HTC devices. Has decided to move to another forum, and only allow grown ups to use his kernel.
chad.goodman's disclaimer
http://forum.xda-developers.com/showpost.php?p=25287750&postcount=241
Register and look for chad.goodman's latest kernel download for our device Problems or bugs, let him know at his thread. Not here.
Code:
This thread is intended to be viewable by members of infectedrom.com only. If there is a security flaw that allows you to read this, and you are not a member of infectedrom.com, you are not authorized to use the "software package" and should leave this page immediately. Thank You.
[B]http://forums.infectedrom.com/showthread.php/1162-KERNEL-Private-Kernel-Download-Thread[/B]
CPU SPEED: 1.5GHZ VOLTAGE: x mv
CPU SPEED: 1.2GHZ VOLTAGE: x mv
Battery: In testing the Idle drain was HIGH. Make no mistake, idle for 2-3 hours easily drained 10% which is about 3% per hour. This was during sleep!
Battery Test : x% every x minutes during idle
Estimated ma : -x ma to -x ma
Movie Test : x% to x%
Performance : Feels slower than Leedroid 5.4.0, faster than Anryl 10.08
Gaming : After reboot, games seems to work fine. Multitouch issues go away after proper reboot.
Issues : Secondary core seems to hit 1.53GHZ even when set to lower frequencies, and INTERACTIVE. In some situations, CORE0 is at 384mhz, but CORE1 is at 1530mhz. While this is not always, you also want to always have your secondary CORE at the same speed as what you configured the MAX.
Its the only kernel with SMARTASSV2, and its own INTELLIANTHRAX governor which previously were reported to be unstable as well.
Sweep to wake causes a battery pull in any case
Antutu Benchmark at 1.2GHZ without SD
RAM 807
CPU INTEGER 1850
CPU FLOAT POINT 805
2D 256
3D 1066
DATABASE I/O 235
Total (5019)
Nenamark 2 : x.x FPS ~ xx.x FPS
Nenamark 2 w/Samsung-EGL drivers : 34.7 FPS
Smartbench 2012 : 1937 (VERY HIGH FOR OUR DEVICE)
Antutu Benchmark at 1.5GHZ without SD
RAM 911
CPU INTEGER 2075
CPU FLOAT POINT 912
2D 296
3D 1152
DATABASE I/O 325
Total (5671)
Nenamark 2 : x.x FPS ~ x.x FPS
Nenamark 2 w/Samsung-EGL drivers : 35.0 FPS
Smartbench 2012 : 2167 (really highest for us now)
Click to expand...
Click to collapse
Personal Summary
Since every kernel seems to be similar, below are my observations between all. These are the KINGS in the following department
BEST OVERALL PERFORMANCE - LEEDROID 5.4.0
BEST OVERALL BATTERY - MEDEEJAY UNITY3B
BEST OVERALL STABILITY - ANRYL 10.07/10.08
BEST OVERALL GRAPHIC - ANTHRAX GSM-SENSE-RLS004
Click to expand...
Click to collapse
The ToDoList
[DONE]Clean up formatting
[DONE]EXCEL method comparison
[DONE]Nenamark2
[DONE]Smartbench 2012
[DONE]Battery Life during idle (this is going to be a tough one)
[DONE]Links to kernels, kernel owners
[DONE]Thanks and appreciation to all kernel developers
[DONE]Thanks and appreciation to ROM developer
[DONE]New kernels
[DONE]New tests on new kernels
You tell me
The disclaimer
I will follow up the thread and update as I have time, or as long as I have the same ROM (which I dont usually flash too often) or same phone
If you dislike my post, cross-reference, findings, let me know, especially if you have been affected by this post I will do the necessary to adjust it. My apologies if I have done so in advance.
If you disagree with my information above, let me know and I will add,change or remove it.
If you found something interesting about any of the kernels, discuss and let me know. I am sure others will want to know as well
I am not responsible for anything above, and damage, warranty,legal and etc is on your own self doing. I will remove them immediately when it affects any of the mentioned above.
I dont claim these are 100% accurate, and variance might be there. Use at own discretion, especially voltages.
Undervolting below HTC stock mV, and overclocking above 1.5Ghz is at your own peril. By default, our CPU are clocked at 1.5Ghz. HTC just downclocked it to 1.2Ghz. Anything above that, and you are on your own.
384mhz seem to be the most stable low point. 192mhz caused random reboots or bugs during lockscreen, in call etc.
Sony XLOUD is really LOUD. However, restoring using the uninstall.zip breaks all AUDIO from Music apps, or beats related.
Reserved For change logs / updates
26 April 2012 v1.34
NEW: Added gaming observations to RLS5
NEW: Added issues to RLS5 regarding sweep to wake
NEW: Added issues to Anryl 10.07/10.08, which is known WAKE UP STUCK AT MIN FREQ
Added Anthrax RLS005 results section. This is probably a BETA kernel. Use at your own risk
Added Anthrax Antutu, Nenamark 2 and Smartbench 2012 results for 1.2GHZ. Note the smart bench results are quite high, falling slightly short of Leedroid's 5.4.0 clocked at 1.5GHZ.
Added Anthrax Antutu, Nenamark 2 and Smartbench 2012 results for 1.5GHZ.
Added issues discovered for RLS005
Added battery observations for RLS005
Post regarding xLOUD on the final disclaimer
Fixed typo
15 April 2012 v1.33
Fixed typo as 3.4v as 0%.
Added 10.08 kernel description
Added 10.08 kernel into FAST CHARGE
Added 10.07/10.08 observations regarding stuck at MIN CPU speed
Added Unity3b as supported under Samsung-EGL drivers, and GPU drivers
13 April 2012 v1.32
Added Samsung-EGL results to Anryl 10.07, 1.2ghz and 1.5ghz had same results.
Added Samsung-EGL results to Leedroid 5.4.0, 1.2ghz and 1.5ghz had better results.
Added Samsung-EGL results to Antrhax RLS004, 1.2ghz and 1.5ghz also had good results.
Added Samsung-EGL driver description and credit to Leedroid
Added FORCE FAST CHARGE section
Added results observed by FAST CHARGE
Added my voltage configuration which I deem stable, that runs on any kernel tested (Leedroid 5.4.0, Unity3b, RLS004 and Anryl 10.07)
Please thank the people who created, tested and developed this for us (Hence I linked full posts). USB cables are now FULL chargers!
8 April 2012 v1.31
Added RLS004 kernel battery observations
Added RLS004 battery drain estimates, and battery drain per hour
Added RLS004 movie test results
Made changelogs into Code, past April 2012, to respect your mouse scroll
7 April 2012 v1.30
Added attachments to front page. *If you think its wrong, let me know, I will remove them*
5 April 2012 v1.29
Added movie test results for STOCK KERNEL
Added GPU drivers section
Added results for NEW GPU drivers on Anryl 10.07 for Nenamark 2
Tested all the games, Aftermath, NFS, Shadowgun and Where is my water. Doesn't seem to have performance gain.
Camera functional
Removed some typo
1 April 2012 v1.28
Added Anthrax RLS04
Moved older Anthrax kernel to The Vault
Added results for RLS04 for Nenamark
Added results for RLS04 for Antutu
Added results for RLS04 for Smartbench 2012
Code:
29 March 2012 v1.27
[LIST]
[*]Added movie test results to Unity3b
[*]Added performance results to Unity3b
[*]Fixed typo
[*]To test cile381 kernel next
[/LIST]
28 March 2012 v1.26
[LIST]
[*]Added movie test results to Kamma v1.0 but remains in The Vault
[*]Added Issue list to Anthrax GSM-SENSE-AnthraX-2.6.38.6-032212_1830 for 3D camera (LAG)
[*]Edited main post for more typo's
[/LIST]
27 March 2012 v1.25
[LIST]
[*]Added results for leedroid 5.3.8 movie test
[*]Created "PERSONAL SUMMARY" at the bottom of the tests
[*]Moved Excel Comparison to The Vault (too old and not going to be updated, pointless)
[*]Changed Unityv3b to Unity3b (typo)
[/LIST]
26 March 2012 v1.24
[LIST]
[*]Added results for anryl 10.07 movie test
[*]Adjust typo
[*]Adjust some explanations
[*]Redid Leedroid 5.3.8 tests
[*]Corrected some Leedroid 5.3.8 kernel results
[/LIST]
25 March 2012 v1.23
[LIST]
[*]Anthrax RLS001a by chad.goodman moved to The Vault Post 3
[*]Added THE MOVIE TEST
[*]Begin tests on GSM-SENSE-AnthraX-2.6.38.6-032212_1830.zip
[*]Added results for Movie Test on GSM-SENSE-AnthraX-2.6.38.6-032212_1830.zip
[*]Added results for Antutu, Nenamark, Smartbench on GSM-SENSE-AnthraX-2.6.38.6-032212_1830.zip
[*]Next to observe battery usage during idle, drains.
[*]Notice some scores didnt really increase, instead decreased when using 1.5GHZ.
[*]Converted Leedroid 5.3.5 to 5.3.8, similiar results
[*]Fixed some typo
[*]Fixed and explained some settings
[*]Included a PERFORMANCE: observation for Leedroid 5.3.8
[*]Included a PERFORMANCE: observation for GSM-SENSE-AnthraX-2.6.38.6-032212_1830
[*]Included a PERFORMANCE: observation for Anryl 10.07
[/LIST]
23 March 2012 v1.22
[LIST]
[*]Added chad.goodman's [url]http://forums.infectedrom.com/showthread.php/1162-KERNEL-Private-Kernel-Download-Thread[/url]
[/LIST]
18 March 2012 v1.21
[LIST]
[*]Added battery observations based on mv, and compare to %
[*]Battery results for -50UV on ANRYL 10.07 added
[*]Added settings to enhance battery longevity guide
[/LIST]
14 March 2012 v1.20
[LIST]
[*]Added recommended configurations for Anryl 10.07
[*]Will add battery test results after UV applied
[/LIST]
6 March 2012 v1.19
[LIST]
[*]Added Leedroid 5.3.5 battery observations
[*]Added Leedroid 5.3.5 battery drain
[*]Changed title name from [GSM][KERNELS][Date][Version]Difference between each kernel via Antutu, and more...
*This thread is no longer just about Antutu or benchmarks, its about the different kernels and observations. Its all on the same user setup, same ROM, same apps and same sync configurations to provide a consistent result.
*Thread name changed also to reflect that its the SAME ROM, as I get PM's regarding better results in different ROM, better battery on different ROM's etc
[/LIST]
4 March 2012 v1.18
[LIST]
[*]Added Leedroid 5.3.5 kernel
[*]Added Leedroid 5.3.5 results for all tests except Battery drain %
[/LIST]
19 February 2012 v1.17
[LIST]
[*]Battery test for Anryl 10.07 returned more positive results
[*]Battery doesnt drop erratically like other kernels for Anryl 10.07
[*]Estimated 24 hours of light/medium use(averaging 5% per hour)
[*]Discovered that graphic glitches can be caused by Go Locker (Thanks Monx)
[/LIST]
16 February 2012 v1.16
[LIST]
[*]To test Anryl kernel 11.03
[*]Added anryl 10.07 battery results (Tested across 5 hours of sleep)
[*]To move Leedroid 3.70 results to The Vault
[*]Added Leedroid 4.2.1 results for Antutu, Nenamark, Smartbench
[/LIST]
15 February 2012 v1.15
[LIST]
[*]Added anryl kernel 10.07
[*]Moved anryl 10.05 to The Vault
[*]Added Nenamark 2 results for anryl 10.07
[*]Added Smartbench 2012 results for anryl 10.07
[*]Added Antutu results for anryl 10.07
[*]Added XDA mirror for Unity3b kernel, since multiupload is down
[/LIST]
13 February 2012 v1.14
[LIST]
[*]Added Nenamark 2 results for anryl 10.05 on 1.5GHZ
[*]Added Smartbench 2012 results for anryl 10.05 on 1.5GHZ
[*]Added Antutu results for anryl 10.05 on 1.5GHZ
[/LIST]
12 February 2012 v1.13
[LIST]
[*]Updated Antutu to 2.6
[*]Updated ROMEOS ROM to 1.30a (with PATCH)
[*]Rerunning tests to check consistency (seems similiar, or consistent hence ROM didnt affect output)
[*]Rerunning tests on Anryl 10.05 kernel
[*]Moved Anryl 10.01 kernel to The Vault
[*]Updated Antutu, Nenamark 2, Smartbench 2012 for Anryl on 1.2GHZ
[*]Noticed lower voltages set for Anryl 10.05 as compared to Anryl 10.01
[/LIST]
5 February 2012 v1.12
[LIST]
[*]Battery tests for HTC Stock Kernel
[*]Estimated Battery Drain for HTC Stock Kernel
[/LIST]
2 February 2012 v1.11
[LIST]
[*]Battery tests for Anryl 10.01
[*]Estimated ma drain for Anryl 10.01
[*]Gaming observations for Anryl 10.01
[*]To do tests on HTC STOCK kernel next
[/LIST]
1 February 2012 v1.10
[LIST]
[*]Smartbench 2012 results added for Anryl 10.01
[/LIST]
1 February 2012 v1.09
[LIST]
[*]Battery Life during Idle and Sleep for Anthrax RLS001a
[*]Estimated -ma drain by Battery Monitor Widget for Anthrax RLS001a
[*]Update formatting for NEW KERNEL by Anryl, based on Kamma v1.0, code name Akernel 10.01(More information on this kernel later)
[*]Created THE VAULT in Post 3 in this thread
[*]Moved Leedroid 3.60 kernel and Kamma v1.00 to the vault
[*]Added Notes #10 to opening post
[*]Updated formatting
[*]Added Excel Comparison Section
[*]Attached Excel Spreadsheet but only for v1.08. Latest kernels included/removed later
[*]Added results for Antutu for Anryl 10.01
[*]Added results for Nenamark 2 for Anryl 10.01
[/LIST]
30 January 2012 v1.08
[LIST]
[*]New discovery ; Graphic glitches caused by unclean kernel installation, cache and dalvik clearing.
[*]Resolved via reboot
[*]Requires more testing
[*]Modified and included this discovery in Notes #6
[/LIST]
30 January 2012 v1.07
[LIST]
[*]Battery Life during Idle and Sleep for Kamma 1.00 (Needs 1 more test)
[*]Estimated -ma drain by Battery Monitor Widget for Kamma 1.00
[*]Added Nenamark 2 results for Anthrax RSL001a
[*]Added Smartbench 2012 resutls for Anthrax RSL001a
[*]Added Notes #9, which was confirmed by me after testing all kernels
[/LIST]
29 January 2012 v1.06
[LIST]
[*]Battery Life during Idle and Sleep for Unity3b
[*]Estimated -ma drain by Battery Monitor Widget for Unity3b
[*]Added Nenamark 2 results for Kamma v1.00
[*]Updated Smartbench 2012 to v1.0.0
[*]Added Smartbench 2012 results for Kamma v1.00
[*]Added battery observations Kamma v1.00
[*]Updated gaming observations on Kamma v1.00
[/LIST]
28 January 2012 v1.05
[LIST]
[*]Battery Life during Idle and Sleep for Leedroid 3.70
[*]Estimated -ma drain by Battery Monitor Widget
[*]Nenamark 2 results added for Unity3b
[*]Smartbench 2012 results added for Unity3b
[/LIST]
28 January 2012 v1.04
[LIST]
[*]Nenamark 2 results added for Leedroid 3.70
[*]Smartbench 2012 results added for Leedroid 3.70
[/LIST]
27 January 2012 v1.03
[LIST]
[*]Artifacts in Unity3b in game : Frontline Commando discovered
[*]Included neol23 -75mv voltage configuration for Unity3b, where neol23 deem it stable.
[*]Included neol23 configuration from Leedroid ROM, with Unity3b. Important : Voltage configurations
[*]Included TODO list of Nenamark2 and Smartbench2012
[/LIST]
27 January 2012 v1.02
[LIST]
[*]Kernel names can be clicked to be redirected to the ROM/kernel main page
[*]Developer names included, with link to profile
[*]Direct download links included, for easier access (I hope I dont violate anything. If I do, I will take them down immediately)
[*]Slight history of the kernels, and what/why they are there
[/LIST]
27 January 2012 v1.01
[LIST]
[*]Added notes and points form
[*]Changed formatting
[*]linked ROM developer
[*]defined gaming glitches
[*]added more TODOLIST
[*]changed Anthraxv2 to Anthrax SHOOTERU-GSM-SENSE-AnthraX-RLS001a
[*]more formatting (I know its not the best planned thread, I just pasted mostly from my notepad)
[/LIST]
27 January 2012 v1.00
[LIST]
[*]Basic frame formatting
[*]Antutu scores
[*]Gaming perspective
[*]more to come
[/LIST]
The Vault
Whats this for ?
Archived information, which is no longer needed in Opening Post
Information here would no longer be updated as they are not the latest kernels. We all like latest, but my point is that sometimes, latest doesnt mean best.
Though not latest, they might prove to be stable and better than others but would add much more work to compare each time.
Code:
[QUOTE]
[SIZE="5"][B][U][URL="http://forum.xda-developers.com/showthread.php?t=1245424"]Leedroid 4.2.1[/URL][/U][/B][/SIZE]
by Leedroid
Download: [url]http://leedroid.ftp.rshost.eu/Shooter-Kernels/LeeDrOiD_Evo-3D-GSM_V4.2.1-1.94GHz-KERNEL.zip[/url] provided by Leedroid
CPU SPEED: [U]1.5GHZ[/U] VOLTAGE: [U]1250mv[/U]
CPU SPEED: [U]1.2GHZ[/U] VOLTAGE: [U]1150mv[/U]
Battery: NA
Battery Test : NA
Estimated ma: NA
Gaming : Better than 3.70.
Issues : NA
[U][B]Antutu Benchmark at 1.2GHZ without SD[/B][/U]
RAM 690
CPU INTEGER 1603
CPU FLOAT POINT 693
2D 255
3D 1149
DATABASE I/O 250
[B]Total (4636)[/B]dropped from leedroid 3.70
Nenamark 2 : [B]29.2 FPS ~ 29.3 FPS[/B]
Smartbench 2012 : [B]1238[/B]
[B][U]Antutu Benchmark at 1.5GHZ without SD[/U][/B]
RAM 790
CPU INTEGER 1804
CPU FLOAT POINT 800
2D 284
3D 1171
DATABASE I/O 320
[B]Total (5169)[/B]
Nenamark 2 : [B]31.5 FPS[/B] Finally a kernel above 31 FPS
Smartbench 2012 : [B]1482[/B]
[/QUOTE]
[QUOTE]
[SIZE="5"][B][U][URL="http://forum.xda-developers.com/showpost.php?p=21835672&postcount=4381"]Leedroid 3.70[/URL][/U][/B][/SIZE]
by Leedroid
Download: [url]http://db.tt/SHTLBjsX[/url] provided by Leedroid
CPU SPEED: [U]1.5GHZ[/U] VOLTAGE: [U]1200mv[/U]
CPU SPEED: [U]1.2GHZ[/U] VOLTAGE: [U]1087.5mv[/U]
Battery: Great. Occasional % jumps but still correct readings after jumps in variance of 2-3%
Battery Test : 1% every 60 minutes during idle.
Estimated ma: -9 to -12 ma drain
Gaming : Better than 3.60. OK so far.
Issues : Noticing data drops. Not all the time, but there are..
Random reboots currently persists in Leedroid ROM. Not on mine, yet :D
[U][B]Antutu Benchmark at 1.2GHZ without SD[/B][/U]
RAM 834
CPU INTEGER 1801
CPU FLOAT POINT 794
2D 295
3D 1149
DATABASE I/O 280
[B]Total (5153)[/B]
Nenamark 2 : 26.5 FPS ~ 26.6 FPS
Smartbench 2012 : 1581
[B][U]Antutu Benchmark at 1.5GHZ without SD[/U][/B]
RAM 931
CPU INTEGER 2025
CPU FLOAT POINT 886
2D 296
3D 1195
DATABASE I/O 370
[B]Total (5703)[/B]
Nenamark 2 : [B]27.5 FPS ~ 27.5 FPS[/B]
Smartbench 2012 : [B]1852[/B]
[/QUOTE]
[QUOTE]
[SIZE="5"][U][B]Leedroid 3.60[/B][/U][/SIZE]
by Leedroid
Download : [url]http://leedroid.ftp.rshost.eu/Shooter-Kernels/LeeDrOiD_Evo-3D-GSM_V3.6.0-1.94GHz-KERNEL.zip[/url]
CPU SPEED: [U]1.5GHZ[/U] VOLTAGE: [U]1200mv[/U]
CPU SPEED: [U]1.2GHZ[/U] VOLTAGE: [U]1087.5mv[/U]
Battery: Great. Occasional % jumps but still correct readings after jumps
Gaming : Slight glitches here and there.
[B][U]Antutu Benchmark at 1.2GHZ without SD[/U][/B]
RAM 800
CPU INTEGER 1792
CPU FLOAT POINT 797
2D 294
3D 1143
DATABASE I/O 310
[B]Total (5136)[/B]
[B][U]Antutu Benchmark at 1.5GHZ without SD(5606)[/U][/B]
RAM 919
CPU INTEGER 2020
CPU FLOAT POINT 894
2D 289
3D 1169
DATABASE I/O 315
[B]Total (5606)[/B]
[/QUOTE]
[QUOTE]
[SIZE="5"][B][U][URL="http://forum.xda-developers.com/showthread.php?t=1311812"]Kamma v1.0[/URL][/U][/B][/SIZE]
by [URL="http://forum.xda-developers.com/member.php?u=319765"]kamma[/URL]
Download : [url]http://www.multiupload.com/X349UFSSFJ[/url]
His work has been discontinued because he bought a Galaxy Nexus
His work [B][U]might[/U][/B] be followed up and rebuild by anryl, [url]http://forum.xda-developers.com/showpost.php?p=21797789&postcount=348[/url]
CPU SPEED: [U]1.5GHZ[/U] VOLTAGE: [U]1250mv[/U]
CPU SPEED: [U]1.2GHZ[/U] VOLTAGE: [U]1150mv[/U]
Battery: In testing, seems to be amazing and great.
Battery Test : 1% every ~70 minutes during idle. Needs 1 more test though.
Estimated ma : -6ma to -9ma, occasionally -12ma
Performance : Similiar to anryl kernel for performance.
Movie Test : 100% to 64%
Gaming : No glitches. All games smooth, no glitches, no artifacts, like STOCK.
[B][U]Antutu Benchmark at 1.2GHZ without SD[/U][/B]
RAM 722
CPU INTEGER 1586
CPU FLOAT POINT 690
2D 295
3D 1163
DATABASE I/O 290
[B]Total (4746)[/B]
Nenamark 2 : [B]27.2 FPS ~ 27.3 FPS[/B]
Smartbench 2012 : [B]1296[/B]
[B][U]Antutu Benchmark at 1.5GHZ without SD[/U][/B]
RAM 826
CPU INTEGER 1799
CPU FLOAT POINT 768
2D 296
3D 1197
DATABASE I/O 375
[B]Total (5261)[/B]
Nenamark 2 : [B]28.6 FPS ~ 28.8 FPS[/B]
Smartbench 2012 : [B]1526[/B]
[/QUOTE]
[QUOTE]
[SIZE="5"][B][U][URL="http://forum.xda-developers.com/showpost.php?p=21914565&postcount=2"]Anryl 10.01[/URL][/U][/B][/SIZE]
by [URL="http://forum.xda-developers.com/member.php?u=410957"]Anryl[/URL]
Download : [url]http://www.4shared.com/zip/LQAQH8m4/akernel_v1001.html[/url]
CPU SPEED: [U]1.5GHZ[/U] VOLTAGE: [U]1250 mv[/U]
CPU SPEED: [U]1.2GHZ[/U] VOLTAGE: [U]1150 mv[/U]
Battery: Good, but not great during use. Seems to drop rapidly (%)
Battery Test : 1% every 66 minutes during idle.
Estimated ma : -4 ma to -27ma
Gaming : Normal. As usual, glitches after first reboot, all seems OK.
[B][U]Antutu Benchmark at 1.2GHZ without SD[/U][/B]
RAM 709
CPU INTEGER 1547
CPU FLOAT POINT 678
2D 295
3D 1173
DATABASE I/O 270
[B]Total (4672)[/B]
Nenamark 2 : [B] 27.7 FPS ~ 27.8 FPS[/B]
Smartbench 2012 : [B]978[/B]
[B][U]Antutu Benchmark at 1.5GHZ without SD[/U][/B]
RAM 810
CPU INTEGER 1770
CPU FLOAT POINT 777
2D 296
3D 1185
DATABASE I/O 345
[B]Total (5187)[/B]
Nenamark 2 : [B] 28.2 FPS ~ 28.2 FPS[/B]
Smartbench 2012 : [B]1406[/B]
[/QUOTE]
[QUOTE]
[SIZE="5"][B][U][URL="http://forum.xda-developers.com/showpost.php?p=21914565&postcount=2"]Anryl 10.05[/URL][/U][/B][/SIZE]
by [URL="http://forum.xda-developers.com/member.php?u=410957"]Anryl[/URL]
Download : [url]http://www.4shared.com/zip/qp60FXjh/akernel_v1005.html[/url]
He is continuing the work of Kamma :)
Recompile and uses latest sources.
For Kamma's kernel results, look at Post 3 (The Vault)
For Anryl kernel old results, look at Post 3 (The Vault)
CPU SPEED: [U]1.5GHZ[/U] VOLTAGE: [U]1125 mv[/U]
CPU SPEED: [U]1.2GHZ[/U] VOLTAGE: [U]1025 mv[/U]
Battery: N/A (%)
Battery Test : N/A
Estimated ma : N/A
Gaming : OK after Reboot
[B][U]Antutu Benchmark at 1.2GHZ without SD[/U][/B]
RAM 650
CPU INTEGER 1564
CPU FLOAT POINT 689
2D 296
3D 1167
DATABASE I/O 280
[B]Total (4646)[/B]
Nenamark 2 : [B] 26.8 FPS[/B]
Smartbench 2012 : [B]1199 (Up from 978 on anryl 10.01)[/B]
[B][U]Antutu Benchmark at 1.5GHZ without SD[/U][/B]
RAM 877
CPU INTEGER 2042
CPU FLOAT POINT 888
2D 296
3D 1192
DATABASE I/O 335
[B]Total (5630)[/B]
Nenamark 2 : [B] 28.2 FPS ~ 28.4 FPS[/B]
Smartbench 2012 : [B]1087[/B](I ran this a few times, and yet got this low score) Puzzling... AGAIN similiar like UNITY3B
[/QUOTE]
[QUOTE]
[SIZE="5"][U][B][URL="http://forum.xda-developers.com/showthread.php?t=1428588"]Anthrax SHOOTERU-GSM-SENSE-AnthraX-RLS001a...already kanged[/URL][/B][/U][/SIZE]
by [URL="http://forum.xda-developers.com/member.php?u=3862386"]chad.goodman[/URL]
Download: [url]http://www.androidfilehost.com/main/EVO_3D_Developers/Despicable.Me.Roms/KERNELS/[/url]
(show some support and click some ad's on link)
Register and look for chad.goodman's latest kernel for our device :)
[CODE]
http://forums.infectedrom.com/showthread.php/1162-KERNEL-Private-Kernel-Download-Thread
For all general awareness,
Definition of "..already kanged.." from Urban Dictionary
http://www.urbandictionary.com/define.php?term=Kanged
1. To have one's developer code and work taken, manipulated, rebranded, and re-released by someone else with no credit.
Ultimate Droid kanged CyanogenMod by taking CM's code, theming the ROM, and then re-releasing it without taking out CM-specific easter eggs, and stripping out the credit and changelog for the CyanogenMod developers.
2. The act of having your meeting hijacked, in favor of a convoluted and meandering presentation of impractical concepts that can never be implemented. Also a term for a discussion that impacts you, but does not include you.
"I got Kanged at the new business presentation yesterday."
Click to expand...
Click to collapse
CPU SPEED: 1.5GHZ VOLTAGE: 1250mv
CPU SPEED: 1.2GHZ VOLTAGE: 1187.5mv
Battery: Erratic sudden battery drops in %.
Battery Test : 1% every ~72 minutes during idle.
Estimated ma: -9ma, occasionally -14ma
Gaming :Slight bugs in games. Shadowgun artifacts. After reboot, seems OK.
Antutu Benchmark at 1.2GHZ without SD
RAM 710
CPU INTEGER 1527
CPU FLOAT POINT 684
2D 280
3D 1113
DATABASE I/O 265
Total (4579)
Nenamark 2 : 29.0 FPS ~ 29.1 FPS
Smartbench 2012 : 1368
Antutu Benchmark at 1.5GHZ without SD
RAM 797
CPU INTEGER 1753
CPU FLOAT POINT 768
2D 295
3D 1195
DATABASE I/O 395
Total (5203)
Nenamark 2 : 30.0 FPS ~ 30.2 FPS
Smartbench 2012 : 1402
[/QUOTE]
GSM-SENSE-AnthraX-2.6.38.6-032212_1830
by chad.goodman
Wrongly accused of stealing peoples code for other HTC devices. Has decided to move to another forum, and only allow grown ups to use his kernel.
chad.goodman's disclaimer
Register and look for chad.goodman's latest kernel download for our device Problems or bugs, let him know at his thread. Not here.
Code:
This thread is intended to be viewable by members of infectedrom.com only. If there is a security flaw that allows you to read this, and you are not a member of infectedrom.com, you are not authorized to use the "software package" and should leave this page immediately. Thank You.
[B]http://forums.infectedrom.com/showthread.php/1162-KERNEL-Private-Kernel-Download-Thread[/B]
CPU SPEED: 1.5GHZ VOLTAGE: 1250mv
CPU SPEED: 1.2GHZ VOLTAGE: 1187.5mv
Battery: No more erratic battery drops. Requires more observations.
Battery Test : N/A
Estimated ma: -xxx ma to - xxx ma
Movie Test : 100% drop to 66%
Performance : Feels slower during use in normal conditions compared to Leedroid & Anryl kernels
Gaming : Multitouch seems to fault every few touches.
Issues : 3D camera lags with amaze cam 0.11
Antutu Benchmark at 1.2GHZ without SD
RAM 790
CPU INTEGER 1821
CPU FLOAT POINT 813
2D 294
3D 1189
DATABASE I/O 270
Total (5177)
Nenamark 2 : 30.3 FPS ~ 30.4 FPS
Smartbench 2012 : 910
Antutu Benchmark at 1.5GHZ without SD
RAM 885
CPU INTEGER 2061
CPU FLOAT POINT 917
2D 296
3D 1230
DATABASE I/O 305
Total (5694)
Nenamark 2 : 30.3 FPS ~ 30.4FPS
Smartbench 2012 : 766
GSM-SENSE-AnthraX-RLS04
by chad.goodman
Wrongly accused of stealing peoples code for other HTC devices. Has decided to move to another forum, and only allow grown ups to use his kernel.
chad.goodman's disclaimer
Register and look for chad.goodman's latest kernel download for our device Problems or bugs, let him know at his thread. Not here.
Code:
This thread is intended to be viewable by members of infectedrom.com only. If there is a security flaw that allows you to read this, and you are not a member of infectedrom.com, you are not authorized to use the "software package" and should leave this page immediately. Thank You.
[B]http://forums.infectedrom.com/showthread.php/1162-KERNEL-Private-Kernel-Download-Thread[/B]
CPU SPEED: 1.5GHZ VOLTAGE: 1250mv
CPU SPEED: 1.2GHZ VOLTAGE: 1187.5mv
Battery: Seems to be amazing during use, but in testing the Idle drain was HIGH.
Battery Test : 1% every 30 minutes during idle. (Updated and second test done)
Estimated ma : -25ma to -39ma (this is almost double of any other kernel)
Movie Test : 100% to 57%
Performance : Feels slower during use in normal conditions compared to Leedroid & Anryl kernels
Gaming : Gaming seems fine on the 4 games I used to test.
Antutu Benchmark at 1.2GHZ without SD
RAM 776
CPU INTEGER 1846
CPU FLOAT POINT 815
2D 296
3D 1157
DATABASE I/O 290
Total (5180)
Nenamark 2 : 27.1 FPS ~ 27.2 FPS
Nenamark 2 w/Samsung-EGL drivers : 35.9 FPS
Smartbench 2012 : 959
Antutu Benchmark at 1.5GHZ without SD
RAM 872
CPU INTEGER 1988
CPU FLOAT POINT 876
2D 296
3D 1189
DATABASE I/O 325
Total (5546)
Nenamark 2 : 28.1 FPS ~ 28.1 FPS
Nenamark 2 w/Samsung-EGL drivers : 37.8 FPS
Smartbench 2012 : 1423
Click to expand...
Click to collapse
Click to expand...
Click to collapse
[/CODE]
Excel Comparison(Very old, revision 1.00)
Total size of 17KB
Excel Version updated to match 1.08 of the thread
Highlighted in GREEN (Highest for that particular task)
Highlighted in RED (Lowest for that particular task)
Yeah, very cool Idea.
That way it'll get a little clearer for the people which kernel to choose --> performance wise
I did something similar about a year ago for my DesireHD --> << press hard on this one >> (Linpack only).
Just one thing:
Maybe Antutu "only" isn't representative enough?
It is a synthetic benchmark isn't it.?
Does somebody have an idea how to bench a little closer to real life circumstances?
I see what you are getting at, but how a kernel works for one person does not say that much. When running antutu with the Anthrax V2 kernel @ 1.5Ghz I score 5829 (without SD) which is considerably higher than what you scored. In the end every device is different and will therefore score different and work better and worse with certain settings..
nice idea dude, but benchmarks are nothing you should believe in
Too bad you can't include chad's Qualcomm series in your comparaison. Oh well, Xda's loss.
Very good idea. Which Anthrax v2 kernel did you use: SHOOTERU-GSM-SENSE-AnthraX-RLS001a.zip or GSM-SENSE-AnthraX-2.6.38.6-REV_005.zip? I believe GSM-SENSE-AnthraX-2.6.38.6-REV_005.zip is much more stable than SHOOTERU-GSM-SENSE-AnthraX-RLS001a.zip.
Thanks for putting in so much work. I would be more interested in a battery drain benchmark. But good work.
Nice thread. I also am interested in seeing benchmarks for battery performance.
Thanks for time and effort put in, it helps give an idea of what each kernel is capable of.
locked & loaded
jcsy said:
Reserved For changelogs / updates
27 January 2012 v1.00
Basic frame formatting
Antutu scores
Gaming perspective
-more to come
Click to expand...
Click to collapse
Mate,
I've. Always found leedroid. Kernel v1 is the fastest on any rom. I've tried
---------- Post added at 02:51 AM ---------- Previous post was at 02:49 AM ----------
metamasterplay said:
Too bad you can't include chad's Qualcomm series in your comparaison. Oh well, Xda's loss.
Click to expand...
Click to collapse
It's not xda's fault, there keeping legal and covering there own arses
Cant blame em, if your one of the few who kept them, u can use them
McKebapp said:
Yeah, very cool Idea.
That way it'll get a little clearer for the people which kernel to choose --> performance wise
I did something similar about a year ago for my DesireHD --> << press hard on this one >> (Linpack only).
Just one thing:
Maybe Antutu "only" isn't representative enough?
It is a synthetic benchmark isn't it.?
Does somebody have an idea how to bench a little closer to real life circumstances?
Click to expand...
Click to collapse
DNAxDREAMS said:
I see what you are getting at, but how a kernel works for one person does not say that much. When running antutu with the Anthrax V2 kernel @ 1.5Ghz I score 5829 (without SD) which is considerably higher than what you scored. In the end every device is different and will therefore score different and work better and worse with certain settings..
Click to expand...
Click to collapse
Yeah every ROM is different. Every device performs differently with different ROM's ,
Best ROM here would beat my ROM out flat, but I used the same platform to test to give the difference between all of them Thats all
Flashmaniac said:
nice idea dude, but benchmarks are nothing you should believe in
Click to expand...
Click to collapse
i'm aware that benchmarks ARE NOTHING, honestly. Made out of curiosity and to cure yours as well. They are a good representation only. Actual life performance varies big time, ask Mr. Iphone He had bad benchmarks everywhere and yet performed really well.
tmihai20 said:
Very good idea. Which Anthrax v2 kernel did you use: SHOOTERU-GSM-SENSE-AnthraX-RLS001a.zip or GSM-SENSE-AnthraX-2.6.38.6-REV_005.zip? I believe GSM-SENSE-AnthraX-2.6.38.6-REV_005.zip is much more stable than SHOOTERU-GSM-SENSE-AnthraX-RLS001a.zip.
Click to expand...
Click to collapse
i used "SHOOTERU-GSM-SENSE-AnthraX-RLS001a.zip"
i dont have "GSM-SENSE-AnthraX-2.6.38.6-REV_005.zip"
In his thread, chad.goodman mentioned
#2: If you happen to have my older kernels, please do not post links to them, or share them in public forum. I was legally ordered to pull them all, so I did. If you downloaded them before 8PM Pacific Standard Time on 1-3-2012, you are fine to use them as you see fit.
Click to expand...
Click to collapse
I shall not violate his order, as I joined EVO3D section after this time
gav-collins1983 said:
Mate,
I've. Always found leedroid. Kernel v1 is the fastest on any rom. I've tried
---------- Post added at 02:51 AM ---------- Previous post was at 02:49 AM ----------
It's not xda's fault, there keeping legal and covering there own arses
Cant blame em, if your one of the few who kept them, u can use them
Click to expand...
Click to collapse
i dont have v1. i joined EVO3D forum very much AFTER this was hidden
my opinion is that I use the same test bed on all tests. Meaning, if you use another ROM and test , it should be similiar for u as well for all different kernels, maybe +200 +300 more since its generally faster
perhaps my ROM has bugs in gaming, but AFAIK its quite toned down in terms of modifications and enhancements
I'm heading out, will modify this later
xb360 said:
Nice thread. I also am interested in seeing benchmarks for battery performance.
Click to expand...
Click to collapse
I found that its hard to measure battery performance but so far i spotted an app called Drain guard on XDA which measures drain only during idle for u.
However % drains of each kernel varies greatly, some jumping a few % and staying at one % for ridiculous amounts of time has made me drop that idea, for now.
Need to find something more consistent, such as running a game for 10 minutes and measure drain, but even that won't be accurate since % is reported wrongly and voltage is jumps up and down. I know because i monitor this day and night. Any ideas?
---------------------Tapatalk ----------------------
Sent from my HD 2 running Hyperdroid CM 7 or just, my new HTC Evolution 3D running ROMeOS Unsensed
I found that kernel in the non-gpl folder, haven't tried it yet. Today I switch to Evo 3D as my primary phone, I am still using the stock kernel.
Hey guys am a new Evo User, come from Desire !
Is there an chance to get more information of the old kernels banned because of qualcomm?
What are the significant differences to the new ones.
Would be very glad for a private message with infos, or maybe somebody who has is it already is able to send it (via Mail)
jcsy said:
Reserved For changelogs / updates
...Included neol23 -75mv voltage configuration for Unity3b, where neol23 deem it stable.
Click to expand...
Click to collapse
Please update with those custom kernel settings:
http://forum.xda-developers.com/showpost.php?p=21880484&postcount=4514
Stable 100%, no reboots, autobacklight works on calls, no lags at all (games & tiping - set the MAX CPU Load to 80).
Tested to 1,51Ghz OC - higher could be unstable - not tested the UV settings, though .
Everything ON, except BT and GPS - u can see the battery life ...
Thanks,
updated with 1.03, and added you into the Unity3b section for user input.
what a great post there man amazing detailed information
i cant seem to find any other kernels other than the one on chad.goodman's site so i assume those were taken down already
missing 1.18Ghz screenshot though, but I would assume 1000mv, or 1025mv?
I doubt Antutu Benchmark will tell kernel's graphic performance, since its 2D and 3D benchmarks are too weak for modern devices.(most of phones have ~60fps cap.)
I recommend use Nenamark2 and Smartbench 2012 with "Performance" governor for more accurate results.
Added to my next TODOLIST for you
Thanks