[TESTING] - NAND OMGB (1.2.3)

[imohtep]

Hi, i think the battery drain with data service is critical in this rom, i have 3 or 4 hour battery life with only whatsapp and push mail, and in winmo i have 8 hour with push email and messenger online; i need data all time for my work

I like this rom (without SOD) but battery drain is critical, for this reason lamentably i roll back to winmo i hope this and SOD have any solution in the future release...!!

Thanks to all developers with this great effort and hard work have managed to extend the useful life of our phone..

[htrakr]

I never really had any issues with battery life and usually lasts the whole day (but then again I am not a massive data user (twitter, facebook and gmail and some app updates) and I have wifi on pretty much the whole day and only a few calls/texts and less than 30 minutes of GPS)
However I noticed something interesting after installing the battery icon mod from UOT kitchen (By itself it is great instead of the useless icons that come stock).
Normally the battery meter is useless and would jump 20-30% randomly and vary from 20% to 90% irrespective of what the actual battery was.
But after the battery icon mod now the battery is very stable and has gone down from 95% to 40% in a very linear way without randomly jumping around. I am sure it has nothing to do with the mod as it only mods the framework-res and systemUI apks and doesnt touch the drivers.
Is there some type of log I can pull to see how the battery is affected

[milda25]

Originally Posted by htrakr

I never really had any issues with battery life and usually lasts the whole day (but then again I am not a massive data user (twitter, facebook and gmail and some app updates) and I have wifi on pretty much the whole day and only a few calls/texts and less than 30 minutes of GPS)
However I noticed something interesting after installing the battery icon mod from UOT kitchen (By itself it is great instead of the useless icons that come stock).
Normally the battery meter is useless and would jump 20-30% randomly and vary from 20% to 90% irrespective of what the actual battery was.
But after the battery icon mod now the battery is very stable and has gone down from 95% to 40% in a very linear way without randomly jumping around. I am sure it has nothing to do with the mod as it only mods the framework-res and systemUI apks and doesnt touch the drivers.
Is there some type of log I can pull to see how the battery is affected

I would like to try this app, but after running it and press button "Select Mod" I got empty list. Have you some info about installing/using this app, please?

[htrakr]

I just followed the instructions in the XDA thread

[KITCHEN][01/22] UOT Kitchen (Ultimate Online Theme Kitchen) version 4.0 - xda-developers

[resinis]

im skeptical because sometimes my battery meter works linearly and sometimes it jumps around. but usually it seems to report about +/- 30% of what it actually is. its pretty bad, but thats a minor problem. the fact that the phone uses up the battery 3-5x faster when idle then in winmo6 is unbearable.

[arrrghhh]

Originally Posted by resinis

im skeptical because sometimes my battery meter works linearly and sometimes it jumps around. but usually it seems to report about +/- 30% of what it actually is. its pretty bad, but thats a minor problem. the fact that the phone uses up the battery 3-5x faster when idle then in winmo6 is unbearable.

Then use WinMo, no one is forcing you to use Android.

I doubt Android will EVER match WinMo for battery life. Even if we could get it so the drain was identical in airplane mode with NO apps installed/syncing... I seriously doubt if you turned on sync and all that other crap Android would have a chance of lasting as long as WinMo does.

The simple fact is Android does a lot more - in the background or otherwise, so it's naturally going to drain more.

[mbellot]

Originally Posted by arrrghhh

The simple fact is Android does a lot more - in the background or otherwise, so it's naturally going to drain more.

Maybe. But I also honestly think there is something fundamentally messed up with the charge/discharge logic.

This still makes me wonder what was so special about that particular day and a half...

I haven't ever replicated a run that long, but I have gone 24+ hours at times. Typical seems to be 14-16 hours, but lately it's closer to 10-12.

Forcing regular "calibrations" appears to help temporarily, but things always degrade within 3-5 days, and it "seems" to be tied to how often I connect a charger (more connections = faster decline of apparent battery life).

[arrrghhh]

Originally Posted by mbellot

Maybe. But I also honestly think there is something fundamentally messed up with the charge/discharge logic.

If you only understood how difficult batt algo's actually are... I'm sure it's messed up.

Originally Posted by milda25

I do not understand why you keep defending the claim that OMGB, LK or the kernel has a problem with power management. SD series has much better battery life (at least 24 hours) while running Android from the SD card in the background doing the same things as NAND series. I do not know what the problem is, but I am sure that something certainly is.
Claiming that nobody forces me to use the Android is not very constructive. Android is a great OS, I want to use it and I am very glad that there are people who provide porting for our old devices. WinMo is a dead platform for me.
I understand that there is still plenty of things to deal with, but over time the power management to have someone take a look. Any problem there certainly is.

Then fix it! I never said there's not a bug, just that you should never expect, even if all else is equal, for WinMo and Android to get the exact same battery run times! Yeesh. You guys need to put up or shut up is all I'm saying.

[[ACL]]

Originally Posted by mbellot

Maybe. But I also honestly think there is something fundamentally messed up with the charge/discharge logic.

This still makes me wonder what was so special about that particular day and a half...

I haven't ever replicated a run that long, but I have gone 24+ hours at times. Typical seems to be 14-16 hours, but lately it's closer to 10-12.

Forcing regular "calibrations" appears to help temporarily, but things always degrade within 3-5 days, and it "seems" to be tied to how often I connect a charger (more connections = faster decline of apparent battery life).

Actually no, its not just the charge discharge logic. The whole batt code is broken and needs love. Do a text search on the batt code and you will find the word "hack often. I've been so busy at work i havent had time to do anything lately. If anyone wants to dive in, just study the htc_battery_smem.c file and you will see why its so broken.

[mbellot]

Originally Posted by [ACL]

Actually no, its not just the charge discharge logic. The whole batt code is broken and needs love. Do a text search on the batt code and you will find the word "hack often. I've been so busy at work i havent had time to do anything lately. If anyone wants to dive in, just study the htc_battery_smem.c file and you will see why its so broken.

Against my better judgment... I pulled the file off of git.

Assuming I got the right one, it seems like none of the kernel messages are being printed other than the "Initialized" one, so it's hard to see what (if anything) might be getting determined/set up.

Even more against my better judgment... I'm going to ask if there is a "dummy's" guide to getting up and running with a dev system. I'm on Windows (XP), but should be able to set up a dev environment in a virtual PC if it "must" be done under linux.

I guess four hours of sleep a day is simply too much.