In a recent moblogging test I was rather shocked to hear my smartphone sound the low-battery warning sound after only 45 minutes on the go! I was playing music, doing a live GPS track and running an IM program at the same. No calls, no gaming, no web surfing! While I was out, I blogged about the problem (from my UMPC.) Yes I was pretty annoyed and you’ll see that from the text.
I should have known better because I’ve seen this problem before. Only a few weeks earlier I had killed the same device in 4 hours while playing with it on a train. I had also heard about Robert Scoble, the guy that live video blogs using an N95, getting only 40 minutes from his device and I had even been told by engineers that 3G uploads simply take a huge amount of energy.
When I got home, I re-charged the device fully and did a static test using the same apps. Music playing, live GPS tracking (I use Nokia Sports Tracker) and IM via Gizmo. I left the device alone and didn’t use it. After 110 minutes it switched off indicating a 2.1W average drain from the 3.7wh battery.
I then did the same test and turned off UMTS (running on the old, lower power GPRS/GSM standard instead.) This time I got 3 hours indicating an average drain of 1.2W.
Standby power on a Nokia N82 is 15mw
Estimated ‘normal phone use’ is about 36 hours. (A few emails, a bit of music, calendar activities, some SMS’s and 5 phone calls) which is about a 100mw average drain.
The results:
Always-on Internet applications increases the average power requirement of a smartphone by a minimum of 1000% and up to 3000%
Let me take a reality kick. 2.1W average drain is amazing. ARM and Nokia should be applauded for squeezing so much out of a device that can do so much. The best x86-based UMPCs take about 8W to do the same.
I’ll take a second reality kick because not many people actually do this in real life.Granted, this is a fairly leading-edge test.
I’ll take a third kick too because technology will improve.
BUT….
If you do want to be ‘always connected’ on a high speed network today (and many many people do) you need a HUGE battery. For the average 16 hour day on a smartphone, you need nine batteries which would weigh 200g making your phone, using the smallest possible components today, over 300gm in weight. [*1] That’s close to the size of the Intel MIDs.
Even using low-speed Internet access (like the iPhone does) you would need a device looking something like this:
And putting it into ‘real’ terms. If you went online for 4 hours per day, which I think will be the norm once people start to get used to web, email, IM and YouTube on the go, you would need a battery three times as big. 10wh. If it was possible to do it, the usage figures would sky-rocket once people learn how to be live on their tailored facebook widget or available on their Skype service all day.
And another thing, You won’t want this all-day device in your pocket all-day! It would be at least twice the volume of a current smartphone and trust me, 3W generates a lot of heat when it’s in your pocket.
The last point I want to make here is that despite the big talk about low-power CPU’s from ARM and Intel, it doesn’t really matter any more for Mobile Internet. In average ‘background activity’ mobile Internet conditions. the biggest power drains are the radios. (Don’t even get me started on simultaneous 3G and Wifi use!) and other components that both Intel and ARM has similar access too. Neither has exclusive rights on low-power radio tech. Neither has rights to magic backlighting technology either. Considering Intel now has CPU’s that idle down to fractions of a watt and average less than a watt drain, the playing field is all of a sudden very level.
The next time you hear a Nokia or Intel exec talking about all-day always-connected Internet in your pocket, think about those batteries!
*1 – I used a new Nokia N82 in this test.
