a very interesting take on analyzing the iPhone 3G battery life:
Analyzing 3G iPhone Battery Life, Part 1
Posted By
JustAnotheriPhoneBlog on Wednesday, July 16, 2008
Lots of people have been complaining about battery life with the new iPhone - and not without reason. Many people, myself included, have experienced “turbo battery drain” - leaving them with seriously discharged batteries much earlier than expected based on 2G iPhone experience. I have begun a “reasonably scientific” study of the battery life and have some basic, preliminary findings and tips to share.
HOW THIS STUDY WAS CONDUCTED
Ideally, to measure current consumption, a meter would be connected in series between the iPhone’s battery and the iPhone to measure how many milliamps of current is being drawn at any given moment. However, not wanting to tear my iPhone apart, I had to settle for something not quite as good - measuring the rate of current being used to charge a full iPhone battery.
Imagine, if you will, a tank of water supplying a water pump. The pump represents the iPhone, which is 'using' the water at varying rates based on what it is doing at the time. The tank represents the iPhone battery, which, of course, will be emptied as the pump draws water from it. Now, on the other side of the tank, we have another pump connected to the city water supply, which is filling the tank as it empties, and we can measure how much water we are supplying. This should give us a reasonable estimate of how much current is being used, and is the method (of convenience) that I worked with.
I measured a fair bit of everything. I turned on individual radios (Bluetooth, Cellular, Wifi, and Location Services) to get an idea of how much current they consumed. I tried various features with all the radios turned off, and others with them on. I tried a few things on Edge, 3G, and WiFi to see what difference each network made.
THE LIMITATION
Reasonably quickly I determined that there was an immediate concern: Many of the higher current consuming activities were showing the same result - which after some quick math I determined was the result of hitting the charge limit of the charger (500ma in this case). So over a certain threshold, I could no longer determine what the consumption was only that it was over 500ma. Again, this is due to the method of measurement that I was using at the time. However, this does help us to identify what uses a lot of power, and what does not - so it’s useful data nonetheless.
LOWEST POWER ACTIVITIES
So what used the least power? As you might expect, an idle phone was the least hungry for power - but what was interesting was that comparing airplane mode (which turns off all radios) and all radios on (Bluetooth, Wifi, 3G, Location Services), the current consumption was only negligibly higher. So clearly, just having those services on isn’t a battery killer. Interestingly, having the screen on did not use a meaningful amount of power either with brightness set to 25%.
LOW POWER ACTIVITIES
Going up the chart, the next items were a bit surprising. Listening to music clocked in next, using barely more power than an idle iPhone. A call using Edge was next, but it used twice the amount of power that listening to music did (this was not a lot of power, however). Idling in Google Maps with location services on was next, using the same power as an Edge voice call. Moving up a notch, surfing on Edge and finding your location in Google Maps with location services off used the same amount of power.
HIGHER POWER ACTIVITIES
Higher still yet under the 'maximum charge' testing threshold were a few items: Using the camera came next, using the same power as an Edge voice call with the speakerphone on (max volume). YouTube on WiFi was next, coming just under the threshold.
HIGHEST POWER ACTIVITIES
All of these items maxed out the charger - so I cannot rank them by current use - but they all use more power than the activities already discussed:
* Booting Up the iPhone from the Power Off state
* Watching video (with all radios off)
* All 3G activities: Surfing and voice calls
* Web surfing on WiFi
* Google Maps location with location services on.
* YouTube under 3G
THE RAW DATA
Here is the actual findings, with the activity and the current consumption (in 12v Amps supplied to the charger) so you can get an idea of how much power is used.
ACTIVITY - CURRENT
iPhone Off 0.07
Idle, screen on, all radios off 0.14
Idle, screen off, all radios off 0.14
Idle, cell on, BT off, WiFi off 0.14
Idle, cell on, BT off, WiFi on 0.15
Idle, cell on, BT on, WiFi on 0.15
Idle, cell on, BT on, WiFi on, location services on 0.15
Music, all radios off 0.16
Call, Edge 0.18
Idle in Google Maps, location services on 0.18
Web, Edge 0.19
Google Maps location, location services off 0.19
Camera, all radios off 0.20
Call, Edge, speakerphone on 0.20
YouTube Wifi 0.21
Bootup 0.22
Video, all radios off 0.22
Web, 3G 0.22
Call, 3G 0.22
Call, 3G, speakerphone on 0.22
Web, Wifi 0.22
Google Maps location, location services on 0.22
YouTube 3G 0.22
WHAT IS APPARENTLY NOT CAUSING DRAIN
I also did a real-life test with the following:
3G on, WiFi on (and connected), BT on (and unconnected), Location Services on, 2 mail accounts (Exchange with Push On, syncing Mail, Calendar, and Contacts; IMAP personal email syncing hourly), and some reasonable uses - a few phone calls, some surfing, some app usage, and some text messaging. After 11 hours, I still had 50% battery life left.
LET US GET RID OF ONE MYTH RIGHT NOW
There have been a lot of people who suggest cycling the battery (fully discharging and then recharging) a few times to somehow improve the battery performance. Some have suggested that Apple recommend this, but in fact, here is what Apple says:
For proper maintenance of a lithium-based battery, it is important to keep the electrons in it moving occasionally. Be sure to go through at least one charge cycle per month (charging the battery to 100% and then completely running it down). (www.apple.com/batteries/iphone.html)
This is not terribly accurate. Lithium-ion batteries do NOT require complete discharge/charge cycles to improve either the initial performance OR the longevity of the battery. in fact, the opposite is true - Lithium-based batteries should be charged as often as possible. However, where some of the confusion occurs relates to a built in 'fuel gauge':
Although lithium-ion is memory-free in terms of performance deterioration, batteries with fuel gauges exhibit what engineers refer to as 'digital memory'. Here is the reason: Short discharges with subsequent recharges do not provide the periodic calibration needed to synchronize the fuel gauge with the battery state-of-charge. A deliberate full discharge and recharge every 30 charges corrects this problem. Letting the battery run down to the cut-off point in the equipment will do this. If ignored, the fuel gauge will become increasingly less accurate. (www.batteryuniversity.com/parttwo-34.htm)
So while an occasional discharge/charge cycle is beneficial, it neither prolongs the life of the battery, nor will it improve initial performance. In fact, Apple says the following on Lithium batteries:
Lithium-ion batteries pack in a higher power density than nickel-based batteries. This gives you a longer battery life in a lighter package, as lithium is the lightest metal. You can also recharge a lithium-ion battery whenever convenient, without the full charge or discharge cycle necessary to keep nickel-based batteries at peak performance. (Over time, crystals build up in nickel-based batteries and prevent you from charging them completely, necessitating an inconvenient full discharge.) (www.apple.com/batteries/)
SO WHAT DOES THIS ALL MEAN?
Not a whole lot at the moment. I am going to both try to find a better (yet non-invasive) way of measuring the higher draw items and am still conducting longer-term studies (involving email push, Exchange sync, etc) to gauge battery life. Hopefully, however, you now have a decent understanding of what activities will drain your battery faster - even if we don not have the definitive story just yet. I am still on the hunt for the real culprit - which I suspect is 3G surfing - and will post the results in a Part 2.
Forums 
