• After 15+ years, we've made a big change: Android Forums is now Early Bird Club. Learn more here.

Root [ROM][Beta][FINAL] CM7 Android For Triumph [Tickerguy Edition, Kernel: mine] 11/19

portezbie said:
I thought juice defender was working for me again, but it doesn't seem to be after all. If it is on 3g it seems to work properly, but if the wifi is on it doesn't go off when the screen is off as it should.
Click to expand...
I do not know what juice defender does (or how) but if you have the Wifi policy set to "never when plugged in" or "when screen is off" once the phone is sleeping, off charge, and a few minutes of no data transport go by it will drop the Wifi. By the same token if you say "never sleep" then it will never sleep the Wifi.

Check your policy settings - it definitely works, I've tested this extensively.

As I said in a good signal area 3g is preferable in terms of power budget. In a marginal (or crappy) signal area wifi is preferable. The key is to avoid searching - if either of the radios do that power consumption will go through the roof.
 
bobloadmire said:
that seems like normal drain to me.
Click to expand...


My WiFi is off... (I never use it) and I've always had things being pushed (Gmail), so that isn't the issue. This is not normal considering on all the other builds I would have to use it heavily before it would be anywhere near 50%. Considering how I use my device about equally every day, this is not normal drain.

Here's my bottom line: I'm not "complaining" I'm simply stating that my battery life is not what it was on previous builds. Maybe this will clear up on the next build, who knows. I love tickerguy's work, and I appreciate all that he has done for us, so this is in no way drying to "diss" his work.
 
Then something is hosed with what's loaded and in use on your phone, because my battery life has not changed in 3g mode at all, and is better when in a strong Wifi area where I can remain connected.
 
tickerguy said:
Then something is hosed with what's loaded and in use on your phone, because my battery life has not changed in 3g mode at all, and is better when in a strong Wifi area where I can remain connected.
Click to expand...

That is what I needed to know. Because I know for a fact your builds never have battery issues (at least for me). I will wait until your next build to do a all-out wipe to see if the issue clears up.

Thank you!
 
lifeblows10 said:
Here's my bottom line: I'm not "complaining" I'm simply stating that my battery life is not what it was on previous builds. Maybe this will clear up on the next build, who knows. I love tickerguy's work, and I appreciate all that he has done for us, so this is in no way drying to "diss" his work.
Click to expand...
Throwing my $0.02 in, I am having a similar battery drain issue with this 10/26 build whereas I never had one before. Today, I unplugged my phone at 7:30am when I left for work, and with data and WiFi off, by 1pm when I went on my lunch break, my battery was down to 64%. This is with essentially no usage aside from a few text messages. My biggest reported battery drain was Android System at 40 some percent.

This was never an issue prior (and shouldn't be an issue now with both data and WiFi off), as I would typically have percentages in the lower 90s with that kind of usage. I believe this must be something new from the latest build.

That being said, this ROM is marvelous. Tickerguy, thank you so much for your time and effort that went into this! Your work is severely under-appreciated.
 
ryanstein said:
Throwing my $0.02 in, I am having a similar battery drain issue with this 10/26 build whereas I never had one before. Today, I unplugged my phone at 7:30am when I left for work, and with data and WiFi off, by 1pm when I went on my lunch break, my battery was down to 64%. This is with essentially no usage aside from a few text messages. My biggest reported battery drain was Android System at 40 some percent.

This was never an issue prior (and shouldn't be an issue now with both data and WiFi off), as I would typically have percentages in the lower 90s with that kind of usage. I believe this must be something new from the latest build.

That being said, this ROM is marvelous. Tickerguy, thank you so much for your time and effort that went into this! Your work is severely under-appreciated.
Click to expand...


This used to be an issue on CM7 on the Evo when I had that. The trick was to go into the recovery, wipe battery stats, turn off the phone and charge till it was full(preferably into a wall plug, not your computer). Then, once it is full, unplug the cord and plug it back and make sure it is charged fully. Then turn your phone back on and continue your day to day use and see if that helps.

If not then I am at a loss for your situation man, 10/26 seems to be working perfectly well for me.
 
wanfaltm said:
This used to be an issue on CM7 on the Evo when I had that. The trick was to go into the recovery, wipe battery stats, turn off the phone and charge till it was full(preferably into a wall plug, not your computer). Then, once it is full, unplug the cord and plug it back and make sure it is charged fully. Then turn your phone back on and continue your day to day use and see if that helps.

If not then I am at a loss for your situation man, 10/26 seems to be working perfectly well for me.
Click to expand...

I'm having almost the exact same problem. I lose about 10 percent an hour and it is over 70 percent android system. Either we're all doing the same thing wrong or there's some kind of bug in the system that not everybody is experiencing.
 
just wanted to check back and say it seems like juice defender is working for me now. Not sure what the problem was, shrug, who knows.

Thanks for all the help though.

Looking forward to your next awesome realese ticker, thanks again for all your hard work.
 
For me, wifi will sleep sometimes while the screen is off, and sometimes it won't. I'd rather not use JD if I don't have to. You said it's working perfectly for you TG? It seems to be sporadic .
 
wanfaltm said:
This used to be an issue on CM7 on the Evo when I had that. The trick was to go into the recovery, wipe battery stats, turn off the phone and charge till it was full(preferably into a wall plug, not your computer). Then, once it is full, unplug the cord and plug it back and make sure it is charged fully. Then turn your phone back on and continue your day to day use and see if that helps.

If not then I am at a loss for your situation man, 10/26 seems to be working perfectly well for me.
Click to expand...


What percent of the battery does it need to be on when you wipe the stats? Near dead?
 
For some reason automatic brightness isn't working for me. Not a big deal though.
Everything else such as wifi on sleep working flawlessly!
 
Last update seems to have broken my phone, battery life is really bad

it's showing dialer used 79 % of the battery in the past 12 hours, I found that odd as I dont make any calls with the phone
 
Tritri said:
For me, wifi will sleep sometimes while the screen is off, and sometimes it won't. I'd rather not use JD if I don't have to. You said it's working perfectly for you TG? It seems to be sporadic .
Click to expand...
It will not sleep an active connection that is moving data; this is CM7 behavior and not under my control. If I have it set to sleep "when screen off" or "never when plugged in" when on battery the Wifi typically goes to sleep a couple of minutes after the phone does. But - if you're running something that is moving data it will leave the Wifi connection up.

If you're running into this something in the background is moving data on your phone and keeping the connection active.
 
Thanks Tickerguy.... My phone battery dies under load pretty quickly since new, so I have a bad battery with this phone. So until my new one from China comes (most likely next week) I'll leave that topic alone.
Except for all the known and reported problems which at this point for me are all pretty minor; the .4 beta Rom rocks. It is far more responsive then stock ever was and my phone has no wifi or 3G issues. I also have not gotten any sudden reboots since the mid alphas of this rom. I have not had to reinstall gapps since the betas started, but I have to manually restore the apps Youtube and google maps for some reason(s) My G mail and android market work on first boot....For the first beta I did a complete and clean wipe of all data and since have only wiped cache and dalvik cache. Thanks again Tickerguy and Isaac for making a good phone so much more responsive and stable. I no longer fell the need to overclock, since this thing is flying pretty well for me right now. I still would recommend at least a couple of reboots after installing or upgrading this Rom to stablize everything, at least I have found this to be true and it can't hurt (except maybe some battery life if it's not plugged in when rebooting).
 
Well the one place that I will continue to try to improve power consumption is that the fix for the wifi lockups involved prohibiting the deepest sleep modes. The thing is that this is mostly a non-starter in terms of impact on battery life if you've actually got the cell radio up, as it's getting polled regularly and has to respond, so the phone has to wake up to do that.

I always used to chuckle at my HD2 which would claim "2ma!" all day long while the battery would lose 5% an hour on standby. Do the math - it's a 1300ma battery, so 2ma is how many hours? Yeaaaaahhhh. How do you spell bull and, well, you know. Now it would draw about that in airplane mode with the radios off; "deep sleep" really is, and there's very little power being drawn when the CPU is in that state. But in order to actually do anything it has to wake and then turn all the GPIO power back on, wait for it to settle, then do its thing.

It will make an impact though on the Triumph as things sit now if you're in airplane mode as right now the kernel doesn't know that the wifi driver is not actually loaded and thus it can drop the lockup protection. Froyo really does go into deep sleep when it can. I need to dig around more in the kernel on this; the libra interface is tiny and the libra code itself (which is the right place for this) we don't have source for, so I had to stick it in the sdio code which is technically the wrong place for this protection but it's all we got and it's the closest to the suspend code in question that I could actually reach.

I'm not giving up on fixing this but the trade-off seems reasonable for now, since the battery life difference is either anywhere from tiny to nil in normal use or actually of net benefit if you're on Wifi instead of cell in a crap service area. My goal is to be able to have the kernel know if the driver is loaded or not and only hold the power state at the higher level if it is. This way if you have Wifi on you pay the ~35ma power penalty (which is unavoidable in that instance), but if not it can drop all the way down to maybe 10 milliamps provided the cell polling doesn't get you (and to a degree it will.) In airplane mode single-digit milliamps should be doable once I get that figured out.

The backlight is murderously bad in terms of power drain and the Snapdragon chip can pull the amps too - between them it's entirely possible to pull over 700ma which will draw down the battery in a couple of hours from full.

One thing to be aware of is that when you flash a new ROM the battery status gets whacked and as such you have to be careful until you've let it charge fullly to 100% and then let it sit at 100% until the charge current goes to zero. That can take a couple of hours or more. If you don't do that then the claimed charge state can be very wrong.
 
Looks like I can get power down to 17ma or thereabouts if Wfi is not on when the phone sleeps. That's a 50% reduction, translates into about 80 hours of standby, and doesn't suck!

I'm after 10ma, but if 17 is the best I can do, well it is what it is.

More in a few days once I've had a bit to play with this load - this has the potential to go really bad if it does (lockup city) so I'm not going to loose it on people until I've run it for a while.
 
I had what appeared to be really bad battery life yesterday but I let it charge overnight and used a Root App called BatteryCalibration this morning which is a free app that will delete the batterystats.bin so a new one is created. I lost about 10% every hour and a half which is what I was used to with stock with about the same usage. This is compared to the 10% every half hour I had going on yesterday. Anyhow just thought I would throw that out there.
 
That's about right if you have some background data running.

Currently with no data moving and the cell radio up you should have ~35ma of drain with the screen off. A bit more with Wifi since the cell radio AND Wifi are on. This assumes a good signal. At present the kernel will not go into "deep sleep" irrespective of the wifi load state (that was how I solved the lockup); this is something that I believe I've fixed in that I can hold the CPU state only if the wifi driver is loaded - if it's off then I should be able to allow the CPU to go all the way down leaving only the ISRs enabled for timed and event wakes.

It's supposed to be able to do this with the Wifi driver loaded but I have no source and the lockup appears to be coming from inside that code somewhere. Best guess is that the ISR (interrupt) is being called without power being restored to the chip first and the completion code hangs as the chip is unpowered. That in turn implies that the kernel we were given may not be the actual kernel in the Froyo release, but heh, how do you ever know, right? This is probably the best I can do unless someone can pry it out of Qualcomm but it's dramatically better than we could do otherwise.

Wifi enabled when sleeping and not connected is a power nightmare anyway since the active search for SSIDs is a monstrous power pig. For that reason you don't want to leave Wifi up if you're in a place with no wifi signal anyway, code patch or no.
 
Hey tickerguy,

Now that most of the major annoyances type of bugs are fixed I was wondering if you could maybe try to solve the AGPS does not work bug. Or is this something that really can't ever be fixed on out phone?

And thanks for all your work. Appreciate it! A lot!
 
You must log in or register to reply here.
Back
Top Bottom