Root battery life and 'display'

If there display is about 6% they are hardly using the phone at all lol.

60% is about normal for a good days use. Mine would sit about there, I do get more than 12 hours though. Try a different ROM like redux. That will get you really good battery life.

Quick note... My display is currently at 70% (2 hours 39 mins on time) using setcpu and RCMixS rom. Battery at 34% after 14 hours.

Redux is gingerbread also. All sense ROMs will use more battery than AOSP gingerbread. Try ReflexTSense HD it had good battery life for me or iNsertCoin 1.9.5 otherwise yea... Redux probably the best battery you will get. Use it with ManU 2.0 kernel

Oh and download "battery calibration" app from the market and calibrate your battery. Should help

Redux, Oxygen, CM7 - all are ASOP (non-Sense) ROMs with a better battery usage.
A lot depends however on how you use the phone. For example, if you have many widgets running, then battery will bleed no matter what ROM you use.

davo26 said:

grrrr im so jealous of people with good batteries spose its the luck of the draw and i lost out

i might try a different rom then, is that one like leedroid? i quite like this one. i tried CM7 and didnt like its lack of 'looks', i spose it seemed more for techies who knew what they were doing

btw, my our batteries all said 1400mAh, is that still where we're at, or can you buy better now?

thanks

Click to expand...

Batteries only need to run at something like 60% capacity before they are legally declared fot for purpose and it seemed you lost out. However I the HTC Desire battery is known to be better than 3rd party batteries rated at 1600mAh so don't go down that road.

It may be rather obvious but try turning things off, Bluetooth, vibrate and especially GPS when not in use. Also consider that you may have an errant app drawing on your battery and you just don't know it.

Bestest way to check for a bad app is a fresh install. Then by restoring only 4 or 5 apps at a time giving it a good 4-5 hours to look at power outage.

davo26 said:

can you think of any reason why this would be? is it just because i use it more, or is there too big a difference for this?

Click to expand...

After I rooted and switched ROMs I became convinced that the HTC Sense stock ROM reports battery usage differently than other ROMs. I'm not sure which is more accurate, but I always had very low (single digit) percentage showing up for display even with auto brightness on using stock ROM. Now, I get 60% or higher every day. A huge difference, but no such huge difference in battery life or usage habits.

amlothi said:

After I rooted and switched ROMs I became convinced that the HTC Sense stock ROM reports battery usage differently than other ROMs. I'm not sure which is more accurate, but I always had very low (single digit) percentage showing up for display even with auto brightness on using stock ROM. Now, I get 60% or higher every day. A huge difference, but no such huge difference in battery life or usage habits.

Click to expand...

Another thing worth mentioning is that if they're not rooted and running Froyo their phone is going to report less display usage in terms of battery drain. Because in my experience it would usually be Cell Standby or Phone Idle that used the most battery in Froyo while in GB it is always display and I find GB is much more accurate in reporting battery use. I'm a Sense guy (I know, shameful, sorry guys), and went from LeeDroid 2.4.1 to 3.0.7 and saw that the jump from Froyo to GB completely changed how battery is reported. My InsertCoin Nightly battery use is similar to LeeDroid 3.x.x as well in terms of reporting.

On topic: Sorry man, looks like you need a new battery, but hey that's good news in a way. You know the problem. And knowing is half the battle.

davo26 said:

i agree, but there again, thats another gamble isnt it :-/ after all, id have thought a new battery in a new desire would be good. thers no saying anything else i bought would be any better.

cheers

Click to expand...

If it's new, wouldn't HTC replace it under warranty? Or is it not that new?

Also just to throw a bit of factual information into the mix.

Remember battery usage is always 100%. If the screen is 6%, everything else must add up to 94%. That 6% is out of the battery used so far, 6% of it was screen. 94% of it was phone signal, standard operation and applications using CPU. Its always 100%.

Now riddle me this....

A screen is very power hungry. Very. However so is a CPU.

Custom ROM? Undervolted kernel? YES!. Wait, what does undervolted mean? Does it mean that the CPU uses less power for the same frequency as a CPU with a standard kernel? YES!.

So everything that uses CPU, on an undervolted kernel is using much less power. Where previously the CPU and radio power would account for 94% of the battery, its much more efficient now. You cannot lower the amount of power the screen uses like this. The screen uses what the screen uses. But everything else uses MUCH less than before, therefore the screen % increases as it always adds up to 100%.

suroot said:

also just to throw a bit of factual information into the mix.

Click to expand...

:d 100%

SUroot said:

So everything that uses CPU, on an undervolted kernel is using much less power. Where previously the CPU and radio power would account for 94% of the battery, its much more efficient now. You cannot lower the amount of power the screen uses like this. The screen uses what the screen uses. But everything else uses MUCH less than before, therefore the screen % increases as it always adds up to 100%.

Click to expand...

Everything you said makes sense, except that it just doesn't seem to correlate with the actual amount of "up time" I get on my phone with a single charge.

Assuming my usage is pretty constant (which is mostly is), and given that I'm at 30-40% at the end of the day on most days (this is post root using GV with undervolted ManU kernel) and that I was usually at 20-30% battery useage at days end (using unrooted Stock ROM)... the math doesn't really work for me.


For simplicity, let's say pre-root my display was using 6% of the battery over a single day. Now, it's being reported as 60% over a single day. We assume that the power used by the screen is constant (I'm using auto brightness) so the amount of charge actually used is the same.


Pre-root: 80% battery used * 6% used for display = display using 4.8% of TOTAL battery capacity.

Post-root: 70% battery used * 60% display = display using 42% of TOTAL battery capacity.

The math just doesn't work. The screen is either using ~10x more battery post-root, or my battery should be lasting 10x longer.

Yeah, there's a difference with rooting and an undervolted kernel, but I don't notice a 10 fold difference. I don't even notice a 2 fold difference.


The only reasonable conclusion I can come up with to explain my observations is that the battery usage metering is completely off in one (or both) ROMs.


Look at cdtDroid's data:

My display is currently at 70% (2 hours 39 mins on time) using setcpu and RCMixS rom. Battery at 34% after 14 hours.

Click to expand...

Used 66% of total battery capacity. 70% is display. That's 46% of TOTAL battery capacity used by the display in less than 3 hours.




I rarely, if ever, saw my display usage listed at over 12% using the stock ROM, even on days when I know my phone was on for 2-3 hours.

Either something in HTC Sense causes the display usage to be grossly underestimated, or something in GV (and other ROMs people are using) is grossly overestimating it.

You can't use the argument that 60% is 10x 6% therefore the phone will last 10 times longer. That's faulty maths.

The screen uses the same amount of power regardless (of course autobrightnes is a huge variable depending on the weather.)

You have to look at the other components. They're the ones using less power. The only way to measure how much longer your phone lasts is to not use it. It's in standby that all these savings are made. The more time in standby, the more you will see a difference.

If you really wanted to try and work it out mathematically, take the 94% everything else, find the multiplier for 40% everything else. 2.35.

If you don't use your screen, your phone should last 2.35 times longer

2h39 minutes screen time, 46% total battery. Yeah that's about right. If you have your screen on constantly for 5 hours on any Rom, your battery is dead.

SUroot said:

If you really wanted to try and work it out mathematically, take the 94% everything else, find the multiplier for 40% everything else. 2.35.

Click to expand...

Yeah, I never was good at math!

So I should be getting more than 2x the battery life, but I'm not. Why not?

Does Gingerbread use that much more power than Froyo that it would negate a 2 fold advantage?

Another way to ask this question: If I used a non-Sense Froyo ROM with an undervolted kernel, would I double my battery life?

2x the battery life soley on CPU utilising components (Android system, applications etc). You have to take screen and radio out of this so it wont really equate to a full 2x battery life, unless you have a 2.2 HTC sense stock, in aeroplane mode and then whatever aosp in aeroplane mode, both not using the screen, yeah I bet you get twice the standby out of the AOSP.

Theres so many factors involved though that it is hard to get a true comparisson.

Both would need to be aeroplane mode, screen off and factory reset.

HTC sense is a user interface firstly (but much more than that, its a framework) htc widgest like "friendstream" and "people" will not work on AOSP roms (roms without sense).

Alternative launcher's such as Launcherpro have replacement apps for this.

Err, yes you can link contacts with facebook and gmail and pictures, sure.

davo26 said:

am i right in thinking that the facebook widget is sense, as thats not there any more. so theres no way of having my contacts linked with facebook and gmail, and having pics for each contact?

Click to expand...

You can link Facebook and contacts but you need to download 'Facebook for Android' from the market.

go to contacts, edit contact, press menu then select 'join' and choose the contacts you want to link up