FORUMS
Remove All Ads from XDA

[KERNEL] SpeedMod SGS2 ICS [K3-32 CIFS ROOT / NO-ROOT CWM5]

3,435 posts
Thanks Meter: 8,145
 
By hardcore, Senior Member on 31st March 2012, 12:10 PM
Post Reply Email Thread
6th April 2012, 09:00 AM |#271  
It_ler's Avatar
Recognized Contributor
Thanks Meter: 16,840
 
More
Results for Siyah / SpeedMod / stock kernels

Everything is specific to my phone, my "normal" usage and my home location. Most values are exact results, but some are rounded. I usually have wifi, synch (three email accounts, calendar and news), Google Talk and Skype on.
For all custom kernels, I was running on ICS XWLP3 „Frankenstein“ ROM (Modem: XXLPR).
For stock kernels, I was running on the corresponding stock ROM.
Always using the Samsung's standard / stock 1650mAh battery.
For SiyahKernel, as it has many, many configurable settings, I have changed nothing - just default settings.


I just flashed SpeedMod K3-7 and added its results.


Here are my current results:


Benchmarks
Quote:

Quadrant Standard 2.0 (best of 3 runs)

SpeedMod K3-7: 4326
SpeedMod K3-5: 4290
Siyah v3.0beta5: 4241
Stock XWLP3 kernel: 4153
Stock XXLPQ kernel: 4063


NenaMark2 (best of 3 runs)

Stock XILA2 kernel: 54,3
Stock XWLA4 kernel: 54,2
SpeedMod K3-7: 52,3
SpeedMod K3-5: 52,3
Stock XWLP3 kernel: 51,9
Stock XXLPQ kernel: 51,3
Siyah v3.0beta5: 50,3
Stock XWKK5 kernel: 47,8
Stock XWKK2 kernel: 47,6
Stock XWKL1 kernel: 38,1


Vellamo (best of 2 runs)

Siyah v3.0beta5: 1296
Stock XWLP3 kernel: 1285
SpeedMod K3-5: 1254
Stock XXLPQ kernel: 1247
Stock XILA2 kernel: 1225
Stock XWLA4 kernel: 1216
Stock XWKL1 kernel: 1173
Stock XWKK5 kernel: 1160
SpeedMod K3-7: 1145
Stock XWKK2 kernel: 1080


Linpack (best of 10 runs): Single / Multi thread

Stock XWLA4 kernel: 57.816 / 93.548
Stock XILA2 kernel: 57.302 / 91.667
Siyah v3.0beta5: 48.015 / 80.317
Stock XWLP3 kernel: 48.015 / 80.317
Stock XXLPQ kernel: 48.208 / 80.013
SpeedMod K3-7: 47.687 / 79.975
SpeedMod K3-5: 47.552 / 79.298


CF-Bench (best of 2 runs): Native / Java / Overall score

Stock XWLA4 kernel: 11396 / 3539 / 6681
SpeedMod K3-7: 11334 / 3394 / 6570
Siyah v3.0beta5: 11113 / 3386 / 6476
Stock XILA2 kernel: 11285 / 3187 / 6426
SpeedMod K3-5: 10980 / 3247 / 6340
Stock XWLP3 kernel: 10795 / 3212 / 6245
Stock XXLPQ kernel: 10862 / 3044 / 6171


AnTuTu (1st run)

Stock XILA2 kernel: 6110
Stock XWLP3 kernel: 6109
Stock XXLPQ kernel: 6075
Siyah v3.0beta5: 6009
Stock XWLA4 kernel: 6009
SpeedMod K3-7: 4670
SpeedMod K3-5: 4643


Battery stats
Quote:

Night use (wifi, synch, Talk and Skype on)

XWLP3: Battery loss of 0,89% per hour, Awake: 6,6%, Deep sleep: 93%
SpeedMod K3-5: Battery loss of 0,92% per hour, Awake: 6,8%, Deep sleep: 92%
XXLPQ: Battery loss of 0,92% per hour, Awake: 7,0%, Deep sleep: 91%
SpeedMod K3-7: Battery loss of 0,95% per hour, Awake: 7,2%, Deep sleep: 91%
Siyah v3.0beta5: Battery loss of 0,95% per hour, Awake: 7,4%, Deep sleep: 90%
XWLA4: Battery loss of 0,95% per hour, Awake: 9,2%, Deep sleep: 90%
XILA2: Battery loss of 1,09% per hour, Awake: 6,5%, Deep sleep: 91%
XWKK5: Battery loss of 1,09% per hour, Awake: 11,6%, Deep sleep: 88%
XWKK2: Battery loss of 1,44% per hour, Awake: 9,9%, Deep sleep: 89%


Mixed (day and night) usage (*)

XWLP3: Battery loss of 1,91% per hour, Wifi: 89,1%, Awake: 16,9%, Deep sleep: 83%, Screen: 10,3% => loss*(1-screen): 1,71%
SpeedMod K3-7: Battery loss of 2,12% per hour, Wifi: 100%, Awake: 18,1%, Deep sleep: 81%, Screen: 11,3% => loss*(1-screen): 1,88%
XWLA4: Battery loss of 2,19% per hour, Wifi: 100%, Awake: 19,6%, Deep sleep: 80%, Screen: 10,2% => loss*(1-screen): 1,97%
Siyah v3.0beta5: Battery loss of 2,32% per hour, Wifi: 91,2%, Awake: 16,7%, Deep sleep: 83%, Screen: 9,7% => loss*(1-screen): 2,09%
XILA2: Battery loss of 2,42% per hour, Wifi: 100%, Awake: 18,2%, Deep sleep: 81%, Screen: 9,0% => loss*(1-screen): 2,20%
SpeedMod K3-5: Battery loss of 2,56% per hour, Wifi: 100%, Awake: 17,8%, Deep sleep: 82%, Screen: 10,8% => loss*(1-screen): 2,28%
XWKK2: Battery loss of 2,37% per hour, Wifi: 81,9%, Awake: 17,4%, Deep sleep: 82%, Screen: 5,9% => loss*(1-screen): 2,23%
XWKK5: Battery loss of 2,58% per hour, Wifi: 96,1%, Awake: 19,5%, Deep sleep: 80%, Screen: 9,0% => loss*(1-screen): 2,35%
XXLPQ: Battery loss of 2,71% per hour, Wifi: 100%, Awake: 19,7%, Deep sleep: 80%, Screen: 10,7% => loss*(1-screen): 2,42%



(*) In order to get a value, which allows to compare test results with different screen on times for mixed usage battery tests, I tried to „normalized“ the battery loss per hour with the following formula:
loss*(1-screen): battery loss in % (per hours) * (100% - screen on time in %)
So, it would be the battery loss in % (per hour) for the screen off time.
That is how a kernel with the same battery loss per hour will get a better/lower value, if it has more screen on time. (And the other way round.)

SpeedMod K3-7 feels at little bit smoother than SpeedMod K3-5 and has better battery life in mixed usage on my phone.
Good work hardcore.

I will test some other kernels and maybe switch to another base ROM before coming back to SpeedMod. But I am sure to come back ;)
The Following 12 Users Say Thank You to It_ler For This Useful Post: [ View ] Gift It_ler Ad-Free
 
 
6th April 2012, 09:05 AM |#272  
Senior Member
Flag Szombathely
Thanks Meter: 137
 
Donate to Me
More
Quote:
Originally Posted by It_ler

Results for Siyah / SpeedMod / stock kernels

Everything is specific to my phone, my "normal" usage and my home location. Most values are exact results, but some are rounded. I usually have wifi, synch (three email accounts, calendar and news), Google Talk and Skype on.
For all custom kernels, I was running on ICS XWLP3 „Frankenstein“ ROM (Modem: XXLPR).
For stock kernels, I was running on the corresponding stock ROM.
Always using the Samsung's standard / stock 1650mAh battery.
For SiyahKernel, as it has many, many configurable settings, I have changed nothing - just default settings.


SpeedMod K3-7 feels at little bit smoother than SpeedMod K3-5 and has better battery life in mixed usage on my phone.
Good work hardcore.

I will test some other kernels and maybe switch to another base ROM before coming back to SpeedMod. But I am sure to come back ;)

hello Mate,
could you please test neak and abyss ICS kernel too?
The Following 2 Users Say Thank You to doongoo For This Useful Post: [ View ] Gift doongoo Ad-Free
6th April 2012, 09:08 AM |#273  
Senior Member
Thanks Meter: 23
 
More
Thanks you for the impressive work
Thank you for the impressive work on the kernel benchmarking.
These results should also be posted on the sgs2 generel forum I would recommend, so more people can see these.

Not to sound too needy, but could you also post the NEAK kernel results?

THanks

------- edit
Do you have any idea why there is such a massive difference between stock kernel and speedmod in antutu benchmark?
The Following User Says Thank You to mcmanuf For This Useful Post: [ View ] Gift mcmanuf Ad-Free
6th April 2012, 10:31 AM |#274  
It_ler's Avatar
Recognized Contributor
Thanks Meter: 16,840
 
More
Quote:
Originally Posted by doongoo

hello Mate,
could you please test neak and abyss ICS kernel too?

N.E.A.K. ICS kernel is already on my "kernels-to-test" list.
And I will have a look at abyss ICS kernel too.
But now I want to find a good "base" ROM for further testings.

---------- Post added at 11:31 AM ---------- Previous post was at 11:23 AM ----------

Quote:
Originally Posted by mcmanuf

Thank you for the impressive work on the kernel benchmarking.
These results should also be posted on the sgs2 generel forum I would recommend, so more people can see these.

Not to sound too needy, but could you also post the NEAK kernel results?

THanks

------- edit
Do you have any idea why there is such a massive difference between stock kernel and speedmod in antutu benchmark?

As I do not trust benchmark scores so much, I only post my test results in kernel-related threads. (Real life experience is more important to me.)
A kernel with best benchmark scores and battery savings on my phone with my "normal" usage can perform different on your phone, in your region with your normal usage.
But thanks for your suggestion.

Significant differences to stock kernels in some benchmarks can have many reasons.
As I do not investigate these results, I can only suggest.
Maybe the kernel developer has better answers, as he has deep knowledge of his kernel and its specialties.
The Following User Says Thank You to It_ler For This Useful Post: [ View ] Gift It_ler Ad-Free
6th April 2012, 11:03 AM |#275  
AndroidGX's Avatar
Recognized Contributor
Flag ★ Montpellier ★
Thanks Meter: 8,983
 
Donate to Me
More
So.. it seems that stock XWLP3 kernel is the winner
Gonna give a try, because I'am curious
Never understood why a stock kernel is finally much better for night/day usage... maybe overall : too much tweaks, kill the tweaks ?

Thanks It_ler
The Following 3 Users Say Thank You to AndroidGX For This Useful Post: [ View ] Gift AndroidGX Ad-Free
6th April 2012, 11:43 AM |#276  
It_ler's Avatar
Recognized Contributor
Thanks Meter: 16,840
 
More
Quote:
Originally Posted by AndroidGX

So.. it seems that stock XWLP3 kernel is the winner
Gonna give a try, because I'am curious
Never understood why a stock kernel is finally much better for night/day usage... maybe overall : too much tweaks, kill the tweaks ?

Thanks It_ler

I think that custom kernels need some time to "overtake" Google/Samsung stock kernels after a new firmware (OS) version is released.
And I am sure that some custom kernels will beat stock kernels in every benchmark and battery test soon.

And ... don't forget, I am testing custom kernels on their default settings - no UC/UV/etc.
As I am a "light" user, the difference between stock and highly configurable kernels is not as significant as for "heavy" users.
The Following User Says Thank You to It_ler For This Useful Post: [ View ] Gift It_ler Ad-Free
6th April 2012, 12:01 PM |#277  
mobs84's Avatar
Senior Member
Thanks Meter: 138
 
More
Quote:
Originally Posted by Tombias

I am using this kernel also but I was wondering if you tried siyah? And what is your opinion about that one compare to speedmod?


Tombias: Yes I have used all versions of Siyah, to answer your question its not appropriate to compare two kernels which cater to different people. I have the freedom to configure Siyah to every possible tweaks from thresholds to UV using EX tweaks, VC and lot many apps, Siyah has a better touch recovery and soon to be with Dual boot option

But speedmod is for those who wants a solid, smooth and preconfigured kernel without the hassles of personalising it by the user to achieve a good level of performance. I should say that has been the legacy of speedmod. Providing with a hassle free kernel tweaked and while keeping to close to stock without too many personalisation support..

I should say I love both and change between them based on my requirement
6th April 2012, 01:07 PM |#278  
OP Senior Member
Thanks Meter: 8,145
 
Donate to Me
More
K3-8:
- Enhancement to DVFS based CPU hotplug policy: prefer 1-core when screen off
- CWM recovery updated to Koush's 5.8.1.8
- Added ability to disabled auto-root

EDIT: Downloads pulled. Minor fix in K3-9 upcoming.
The Following 4 Users Say Thank You to hardcore For This Useful Post: [ View ] Gift hardcore Ad-Free
6th April 2012, 01:19 PM |#279  
micmaccc's Avatar
Senior Member
Thanks Meter: 39
 
More
file not found link in OP

Sent from my GT-I9100
6th April 2012, 01:30 PM |#280  
Samhain77's Avatar
Senior Member
Thanks Meter: 90
 
Donate to Me
More
Quote:
Originally Posted by hardcore

K3-8:
- Enhancement to DVFS based CPU hotplug policy: prefer 1-core when screen off
- CWM recovery updated to Koush's 5.8.1.8
- Added ability to disabled auto-root

EDIT: Downloads pulled. Minor fix in K3-9 upcoming.


just the time to see the new K3-8 changelog and the download link disappeared...... now waiting for K3-9
6th April 2012, 01:36 PM |#281  
OP Senior Member
Thanks Meter: 8,145
 
Donate to Me
More
K3-9 uploaded now.

K3-9:
- Enhancement to DVFS based CPU hotplug policy: prefer 1-core when screen off
- CWM recovery updated to Koush's 5.8.1.8
- Added ability to disable auto-root (see OP/3rd Post for procedure)
- adb root enabled (thanks to Chainfire's modified adbd)
The Following 22 Users Say Thank You to hardcore For This Useful Post: [ View ] Gift hardcore Ad-Free
Post Reply Subscribe to Thread

Tags
ics, kernel

Guest Quick Reply (no urls or BBcode)
Message:
Previous Thread Next Thread
Thread Tools Search this Thread
Search this Thread:

Advanced Search
Display Modes