Discussion in 'Android Devices' started by RoboMonkey, Jul 22, 2010.
HTC Focused On Improving Smartphone Battery Life Velocity - Forbes.com
Sounds like the improvements are what we do manually already. They will be adding more "options" to control things.
"HTC is slated to release at least one more 4G handset this year, for Washington-based wireless broadband provider Clearwire"
Yeah I think the main issue is the cavalier approach to data- and battery-intensive processes. Practically everything turns on and stays on at the beginning. If everything was off by default like iPhone, it would get great battery life. But it would also be far less functional and integrated which is one of the things that puts this phone head and shoulders above iPhones. Catch 22 I suppose.
It will be nice if future updates include features you need multiple apps for now, such as system monitoring that is app/hardware specific.
it's nice to know that "android system" is using all my battery... but to know WHAT within the system is using it would be nice
without having to fish through so many 3rd party apps and screen them ourselves. that should be done by HTC/SPRINT for us. just like windows monitoring and controlling what applications use system resources. and granular control over these things is a must.
True. It would be cool if they had one central area for battery control. So that you can put a widget or shortcut onto your homescreen which when press links you to a page that has controls for WiFi, Brightness, Mobile Data, 4G, system processes, etc.
I think it's the Sprint apps that are the culprit. Once I rooted my phone and flashed a new ROM my battery life really went up. To me that points not to HTC or Google but to Sprint. But I haven't done any actual benchmarking so maybe I'm really off base.
Nothing will matter until the software code that manages recharging and power use is addressed by HTC. That's also the reason the phone stops charging when full and then goes directly to battery use without ever resuming charging when plugged in (meaning the software is preventing trickle charging and may not even be allowing a true 100% charge).
I've charged the battery outside the phone once to test it versus charging inside the phone. The externally charged battery lasted about 20% longer, including keeping it's original full charge much longer. That's part of the reason I'm convinced the issue really is software-related and not the batteries themselves.
What would make things even more conclusive is if someone from this forum who has another brand of phone that uses the same battery could test the battery inside that other phone. I believe that would show the battery's not a problem at all.
They're not in the battery manufacturing/R&D chain, so not sure what they plan on doing on their end. The Broadcom BCM4329 is pretty damn efficient as it is.
Are you sure it never charges again? I was under the impression that testing revealed that it does start charging again (just at some currently unknown point)
Simple proof - leave phone on, charging all night, wifi on, let phone merrily sync and so forth while you sleep. Next day, use phone while on charger - all day. End of day, take phone off charger, you find the typical rapid drop to +/- 90% or so - but not significantly lower, like 60% or or 20% anything you'd expect from whatever your day of use is.
Empirically, I believe the re-charge trigger point is anywhere between 82~92% - but that's not scientific.
If anyone feels like searching, back in June a couple of guys here used power monitors and tracked the charging activity from the wall - very cool.
PS - I would believe that the power re-cycle point on the Evo is set differently than other phones because they have some sort of power profile they use to configure this sort of thing - with the larger screen and extra radios changing the potential power draw, I'd imagine that would change that set point. We could ask HD2 users about their experiences - that would seem like the closest analog to our phones.
One day if I'm bored enough, I'm going to figure out the best way to consume as much power as possible, roughly figure out what the consumption rate is, time how long it takes to consume say 20% battery, and then run that while charging (and hooked up to my Kill-A-Watt).
We'll be able to see how long it takes for the charge to start drawing power again, and can roughly determine at what point charging is once again triggered.
do you think playing snes games and nes games, using those emulators is a big drain on battery? as opposed to just talking on the phone?
Fire up Pandora or the Sirius Radio app in a questionable coverage area! You can have fun watching the battery meter go down.....
While I can't say "never", it did not resume charging once in the 10 hours I observed. That's enough for me to think it's probably not coming back to charge. I guess the next step for me would be to see if it runs out of battery completely despite being plugged in. At a minimum, I'd hope the software would just start the recharging process when it give the "low battery" warnings.
How do you know it didn't resume charging? When you unplugged it, did it drop all the way down to say, 40 or 50%
As far as I've seen, there is no transparent method on the Evo itself -- you need to resort to an external indicator of charging, like measuring the power draw from the AC adapter.
Furthermore time (alone) isn't necessary a good measure for drawing that conclusion. Hypothetically, let's say that the charger resumes float charging at 90% depletion. If in that 10 hours battery consumption is minimal and the battery is still about 90%, then that would sufficiently explain why no charging occurred.
I maintain a home office - and use my phone a LOT during the day - with a wired headset.
When I'm in videoconference with a laboratory via my laptop, I'll use my phone to cross-check emails on a given matter - or pop over to our internal website to pull up references - or take long protracted calls (on the order of hours).
Point is - in my senior moments, I'll often simply forget to unplug my phone when I'm in that mode - and then end up unplugging it only after a hard day.
And I never experience low charge levels.
That's entirely anecdotal, so take it for what it's worth - but it seems to come down on the side that for a normally-functioning Evo, the re-initiate-charge trigger (like that? I made it up) is a reasonably high percentage value.
my phone never goes below 90% when fresh off the charger. I've had the phone scan continually for 3g, 4g, bluetooth and left navigation on while playing homerun battle (drains about 25% of the battery in 15-20 mins in this usage pattern when unplugged, as I didn't have 3g or 4g coverage in my area... we now have 3g, lol) to test this. And while I don't have a definite percentage for you guys, it does resume charging at some point.
People like to blame "Bloatware" for "slowing down" their systems, but save a few msec on boot (leading to cries of "it's loading stuff that I don't want!") those apps are seriously just non-factors. Whether you use "ps" in console mode, OSMonitor or "System Monitor", those apps might be loaded (and swapped out when not needed- that's one of linux' strengths, esp for applications that are loaded into ROM in the first place) and not using a cycle of CPU once they've inited.
It's just more Android misconceptions being lauded as facts. What I suspect people are seeing when they "swap ROMs and get better battery life" is the effect of a freshly-booted system.
Wow your phone came with this stock. If you want it to look cleaner just get switchpro from the market.
So far I am happy with my battery!!
Only things I hope they do focus on UN necessary applications such as Amazon MP3, 4 different navigators, and others...
And yes, it dropped down to 60% within 2-3 mins after unplugging when left plugged in overnight when I knew the charge to fill would be completed at approximately 9pm. The following day when I unplugged at 7:30am, it dropped pretty quickly. In that trial round, I left the phone on while charging and overnight specifically to determine whether the phone would resume charging. It didn't, and that's why I don't believe there's trickle charging - at least not until some level below 20% which is as far down as my trials have gone. It takes a few days to do each round.
I've done each trial round using time (X) and same settings (Y) as known variables because I can control and monitor them.
The phone will show charging by the red/orange indicator light and by the battery in the notifications top bar when charging. My checks of those have never shown additional charging once charge has been complete.
The only time there's been virtually no drop in battery level has been when the phone has been shut completely off.
I'm sure this would be easier for me if I had proper lab equipment, but I don't have that type of electronic testing environment in my house.
Wow, Rig - if my phone is any basis whatsoever - I'd say you've got something, somehow, defective there.
My light day on a charge, unplugged mind you - is 12 hours or much better. On a heavy day, I can drain a battery in 4 or 6 hours.
So, if I'm getting never below the mid 80s when unplugging from a hard day's use - and never have such a bad result with ANY sort of overnight charge - then I'd conclude that mine's golden, there's insane variability, or you've got a lemon.
With the proviso that the bundleware doesn't seem to always work deterministically for all of us.
Take my old favorite, the one you and I have spoken about before, if I'm not mistaken - the Amazon MP3 store.
While I never once used or launched and tried to wipe its data - on mine, that beast kept cropping up and babbling on the net - and that's not energy-free. I got that from my syslogs and those were after rooting.
Ditto for Facebook.
I removed them and I did see a definite non-placebo effect in battery life - and a correspondent drop in CPU frequency and Network IO (courtesy of System Panel).
I dunno - maybe I just had something so frakked up from when I was new to my phone and I somehow self-inflicted those wounds.
I totally trust you and you wouldn't lie - others claim the bundleware isn't any load to speak of - but some of us have varying mileage.
So - even if the problem is PEBKAC - removal worked for me.
Besides - the Linux server I'm running two feet away has no processes running other than those I've decided should run in the first place.
Good sysadmin is just good sense, imo.
(PS - Love the term bundleware - copped it from /. yesterday. )
I don't have a lemon. I'm controlling the variables of time and what's working on the phone. That's allowing me to somewhat inch my way towards if there's a point at which charging resumes. I haven't found it yet despite going to a low of 20%.
With my normal use, I get a full day of moderate-to heavy use and end up with about 50% charge at night. That's with a day that runs 6:45a-11:30p.
Funny thing is I've never complained about Evo or thought the battery wasn't adequate. It's what I expect for a phone with these features that I use.
I'm just doing the tests because I want to confirm that there point at which the phone will resume charging. That's not the same as trickle charging, and even HTC acknowledges that the power port will not trickle charge.
The only trickle charging of this phone takes place via the usb port. USB charging takes a long, long time but I think I can cut some steps out when I do USB charging trials next.
HTC's support indicates that the only way to get trickle charging and keep the Evo fully charged is via usb. I believe that. I was just trying to find out if regular charging as a trip-point to resume to full charge. It just doesn't look like it does.
My final test will be to leave the phone on and streaming to see if it runs completely out of power while still plugged into the charger. That will leave no room for doubt!