What govenour do you guys recomend on r62 for battery? 
Really nice build, would never have thought it is only 2 cores.
Really nice build, would never have thought it is only 2 cores.
[KERNEL][GPL][N4] franco.Kernel - r217
- Thread starter franciscofranco
- Start date
Re: [KERNEL][GPL][N4] franco.Kernel - r53
So it looks like r62 is not for me. I'll hold for the next one, probably a week later.
Wow, 1000 pages!
Sent from my Nexus 4 using xda premium
So it looks like r62 is not for me. I'll hold for the next one, probably a week later.
Wow, 1000 pages!
Sent from my Nexus 4 using xda premium
AW: [KERNEL][GPL][N4] franco.Kernel - r53
R62 is good but not as good as 61. 61 was better and also gives me around 5 hours screen in time.
Maybe you can put now the Option for a 2-4 core Switch ? So everyone can decide
And as i use a screen protector, i would love To see a way to increase sensitivity of the screen with your kernel
Gesendet von meinem Nexus 4
R62 is good but not as good as 61. 61 was better and also gives me around 5 hours screen in time.
Maybe you can put now the Option for a 2-4 core Switch ? So everyone can decide
And as i use a screen protector, i would love To see a way to increase sensitivity of the screen with your kernel
Gesendet von meinem Nexus 4
Yes I know they're for testing and I KNOW 62 was designed to only use 2 cores. I'm not complaining about that, but like I said in the original post, there doesn't seem to be a Kernel with all 3 features I specified. Maybe someone can prove me wrong?
1 - Only r61-test or newer
2 - Thats not the point in this testing...
3 - Also not the point in this testing (r61-test has all cores online)
That doesn't help...why it isn't for you?
Get rid of that screen protector. Come on guys, manufacters create those beautifull screens, invent scratch resistant glasses and you put a cheap china plastic on it and loose the sensitivity and colors etc. Just be more careful with the device
Re: [KERNEL][GPL][N4] franco.Kernel - r53
Turn on and off cores is expensive. So Franco decided to TEST turning on all the cores no matter what when screen is on.
R61 and r62 are two TEST builds, or say experimental builds with same logic with different number of cores online to compare battery life.
This has been repeated soooooo many times on the last 3 pages. Please do read before flashing anything, or wait for the first milestone. OP said very clearly that if you don't know what's going on, do not flash test builds.
Sent from my Nexus 4 using xda app-developers app
Read more and you will understand. There is a flow in the discussion. If not possible just search for Franco's replies in this thread.
Turn on and off cores is expensive. So Franco decided to TEST turning on all the cores no matter what when screen is on.
R61 and r62 are two TEST builds, or say experimental builds with same logic with different number of cores online to compare battery life.
This has been repeated soooooo many times on the last 3 pages. Please do read before flashing anything, or wait for the first milestone. OP said very clearly that if you don't know what's going on, do not flash test builds.
Sent from my Nexus 4 using xda app-developers app
Franco has also said all builds are test builds until the milestone build so what you say makes zero sense, just as having test builds makes zero sense.
I give up for tonight and for a few days. Keep the thread clean, be nice to each other, don't bull**** just by bull****ing, this is a development forum, not a kindergarden. And please, read some, that way everyone will be helpful.
Good night gentleman.
Good night gentleman.
If it's the first time you've heard this then you've not been here very long.
Re: [KERNEL][GPL][N4] franco.Kernel - r53
Thx franco go relax or rest get ready to watch sun football go ravens:thumbup:
Thx franco go relax or rest get ready to watch sun football go ravens
:thumbup:
Just flash r62 and all core were on-line, wiped data and still all 4 core on-line, but after a reboot all ok, just 2 core now. If any one experience this problem just reboot after you flash r62. Just did a antutu with r61-15k and r62-11k not sure how this is translate in real world application. First time with r61 manage 17k...
But all our phones are NOT the same, I regularly got 5hrs on stock, I've not read that many posts of people getting 7hrs, have you?
Last edited:
Re: [KERNEL][GPL][N4] franco.Kernel - r53
Sent from my Rock & Roll Nexus 4
ROM: MINCO V6
Kernel: Franco r62
T-Mobile
Fire up those cores dude! Good luck
Sent from my Rock & Roll Nexus 4
ROM: MINCO V6
Kernel: Franco r62
T-Mobile
Re: [KERNEL][GPL][N4] franco.Kernel - r53
Franco I think its time you get a group of ten or so advanced users to test these types of kernels in the future.
The general population in the thread is not meant for such a test / discussion. Or just do it blind and don't tell the people what's in it I suppose.
Franco I think its time you get a group of ten or so advanced users to test these types of kernels in the future.
The general population in the thread is not meant for such a test / discussion. Or just do it blind and don't tell the people what's in it I suppose.
Re: [KERNEL][GPL][N4] franco.Kernel - r53
Is anyone getting large amounts of msm_hsic_host wakelocks again? It went away but is back after flashing r61 & r62?
Only been off charge for 30mins and it has kept device awake for 5mins.
Sent from my Nexus 4 using xda premium
Is anyone getting large amounts of msm_hsic_host wakelocks again? It went away but is back after flashing r61 & r62?
Only been off charge for 30mins and it has kept device awake for 5mins.
Sent from my Nexus 4 using xda premium
It went never away after Franco returned every HSIC patch to stock again, somewhere around r30-40, it was always there and will be always there until Google fixes it.
You won't find 10 advanced users that give reliable feedback in this thread.
My understanding of what test builds are for
For those who missed the whole point of this test build flashing exercise, I hope I can enlighten you guys. From my understanding:
1. r53 uses hotpluging improvement that turns on/off cores based on load. This is the original Franco solution for minimizing battery usage without sacrificing performance. This was the goal of the many recent nightly builds, apart from fixing data drops and wakelocks. If you desperately want UV, then this build is not for you. There is a good reason why UV, being a new thing Franco has been developing, still has to "answer" some questions in his mind, and resolve "unfinished" issues with this new feature before bringing UV to the nightly. So in due time, it is ready when it is ready.
2. r61-test primarily has all 4 cores online, but uses powerstates with is more battery friendly than hotplugging. This is a relatively new idea, that is why this is in the test build. So flashing-trigger-happy dudes can try out if it really does have greater performance and response than hotplugging. Since it will really take time to turn cores on or off, having all cores online all the time is really an advantage in performance. So Franco is waiting for feedback if this performance increase is really a battery friendly solution. Like other test builds, Franco releases these things so he can get "answers" which he will consolidate in all to make another impressive improvement in the nightlies. It just so happens that Franco also used this opportunity window to bring out UV and get answers in that aspect as well. So flash away and give feedback, that's the whole point.
3. r62-test is basically a r61-test replica in terms of configurations in the governor except, that only two cores will be online all the time except during boot. The whole point is to check whether in the powerstate implementation, will using 2 less cores do have a "significant" difference in the battery. I think, Franco is thinking that if the difference is so negligible, we can expect four-cores-always-online type of thing in the next nightly. Otherwise, it could be a mixture of powerstates and dynamic ramping of two extra cores when needed, I don't know, I'm not a kernel developer to know how these are implemented, but I do get the essence of this whole exercise. So r62-test being 2 cores only, doesn't mean 2 cores is always better and "The Right" solution than 4. Neither is 62-test a definite improvement over 4 cores, or that Franco Kernel will always have 2 cores forever from now on, turning our N4 into a "lesser" beast whose purpose of existence is defeated, of course not!
Whether there are advantages or disadvantages between r61-test and r62-test, the point of the test build is to be tested, the point is for us to know, enumerate and qualify the differences in this forum. So Franco can get the answers he needs from these two test builds and come up with best configuration/setup he sees fit for the next nightly. I hope I'm correct in this.
Happy flashing everyone!
For those who missed the whole point of this test build flashing exercise, I hope I can enlighten you guys. From my understanding:
1. r53 uses hotpluging improvement that turns on/off cores based on load. This is the original Franco solution for minimizing battery usage without sacrificing performance. This was the goal of the many recent nightly builds, apart from fixing data drops and wakelocks. If you desperately want UV, then this build is not for you. There is a good reason why UV, being a new thing Franco has been developing, still has to "answer" some questions in his mind, and resolve "unfinished" issues with this new feature before bringing UV to the nightly. So in due time, it is ready when it is ready.
2. r61-test primarily has all 4 cores online, but uses powerstates with is more battery friendly than hotplugging. This is a relatively new idea, that is why this is in the test build. So flashing-trigger-happy dudes can try out if it really does have greater performance and response than hotplugging. Since it will really take time to turn cores on or off, having all cores online all the time is really an advantage in performance. So Franco is waiting for feedback if this performance increase is really a battery friendly solution. Like other test builds, Franco releases these things so he can get "answers" which he will consolidate in all to make another impressive improvement in the nightlies. It just so happens that Franco also used this opportunity window to bring out UV and get answers in that aspect as well. So flash away and give feedback, that's the whole point.
3. r62-test is basically a r61-test replica in terms of configurations in the governor except, that only two cores will be online all the time except during boot. The whole point is to check whether in the powerstate implementation, will using 2 less cores do have a "significant" difference in the battery. I think, Franco is thinking that if the difference is so negligible, we can expect four-cores-always-online type of thing in the next nightly. Otherwise, it could be a mixture of powerstates and dynamic ramping of two extra cores when needed, I don't know, I'm not a kernel developer to know how these are implemented, but I do get the essence of this whole exercise. So r62-test being 2 cores only, doesn't mean 2 cores is always better and "The Right" solution than 4. Neither is 62-test a definite improvement over 4 cores, or that Franco Kernel will always have 2 cores forever from now on, turning our N4 into a "lesser" beast whose purpose of existence is defeated, of course not!
Whether there are advantages or disadvantages between r61-test and r62-test, the point of the test build is to be tested, the point is for us to know, enumerate and qualify the differences in this forum. So Franco can get the answers he needs from these two test builds and come up with best configuration/setup he sees fit for the next nightly. I hope I'm correct in this.
Happy flashing everyone!
Our Apps
Get our official app!
The best way to access XDA on your phone
Nav Gestures
Add swipe gestures to any Android
One Handed Mode
Eases uses one hand with your phone