Reply
Thread Tools
Benson's Avatar
Posts: 4,930 | Thanked: 2,272 times | Joined on Oct 2007
#191
Originally Posted by dubiousmike View Post
Someday, my kids will wonder how we ever went less than 6 months between charges.
Understood you're being facetious, but no, they won't. Devices today have the same battery life devices had in the nineties, because the power consumption increases with battery technology. Partly this is how technology works, and partly there's accepted cultural notions of what battery life should be.

If you built an N800 to run for a week of active use, you'd sell maybe a tenth as many, because most people wouldn't like the tradeoff, whether in performance, or in portability. Likewise, if it only lasted 30 minutes, but had an x86 CPU for better compatibility and more performance, or to save practically all the battery weight, few people would buy it.

So while there's quite a lot of leeway technically (right now), people prefer something like current battery life. I, and I think many others, would gladly trade the weight/size for about twice the battery life, but even so I'd think it crazy to build one that lasts for a week.

If battery/power-saving technology improves more than power-consuming capabilities, some of the improvement will go to increased battery life, but a lot will go to shifting it up the existing performance/power curve.
 
aflegg's Avatar
Posts: 1,463 | Thanked: 81 times | Joined on Oct 2005 @ UK
#192
Originally Posted by Benson View Post
Understood you're being facetious, but no, they won't. Devices today have the same battery life devices had in the nineties, because the power consumption increases with battery technology. Partly this is how technology works, and partly there's accepted cultural notions of what battery life should be.
I beg to differ: my Psion 5mx went for a month or two between recharges of its 2 AA batteries - and that's being used every day. But you're right: consumers nowadays have an expectation on the amount of time between recharges that are acceptable; but that acceptance varies on what they're getting in return.

There's a problem with the NITs: the over-reliance on power management to deliver marketing's "always-on" vision just doesn't work in the real world. There aren't the tools available for power users (or even developers) to identify what's killing the battery.

My wife's N810 lasts less than a day, and doesn't have any additional software installed which would run in the background (tomorrow I'm going to have a good look); whereas my N810 can vary from needing to be charged every night to going a couple of days between charges - without my usage changing from one day to the next.

Even Nokia's own software, such as modest; the RSS home reader applet and so on can drain the battery by doing braindead things. And with the increasing ease with which programmers can knock up things which run continually within hildon-desktop in high-level languages like Python the system need much more aggressive power management.

Personally, I'd like to see apps have to opt in to not being frozen when they go into the background or the screen goes off: that way the developer is making a conscious choice rather than just trying to update the screen every 50ms.
__________________
Andrew Flegg -- mailto:andrew@bleb.org | http://www.bleb.org
Now known as
Jaffa
 
Benson's Avatar
Posts: 4,930 | Thanked: 2,272 times | Joined on Oct 2007
#193
OK, you're right -- they actually go downhill. HP 48sx vs. 49g+ comes to mind as well. And my Dad had an HP200lx -- two AAs lasted forever, with it shut down when not in use. (At least a month, maybe two.)

But I was thinking more of PC-compatible laptops, and cell-phones, being two examples I'm familiar with. I haven't seen an order-of-magnitude change in those. (Considering mainstream devices, and talk time for cell phones; even dirt-cheap phones now have everlasting standby.)

You're right, the use of power-management rather than shutting it down when not in use does open the door for unpredictable battery abuse. A useful tool for analyzing such things might be some sort of logging top, but then that would kill the battery on its own.

The "default to sanity" you speak of would be good; app frameworks for the tablets, like pygame, should include that. But there's no way to enforce such a thing in general, more's the pity.
 
tso's Avatar
Posts: 4,783 | Thanked: 1,253 times | Joined on Aug 2007 @ norway
#194
powertop anyone?
 
Posts: 7 | Thanked: 0 times | Joined on Mar 2008
#195
If you go back further, then an Apple Newton (surely the GreatGrandaddy of the NIT) would run for months on a charge. Black and white screen, and limited connectivity helped, but you never wondered if the battery would last.

And I don't want a "phone" in my tablet, I want a data connection that is instant and just there all the time. It is like Climate Control in cars, yes you can get the same result with air-con - but once you are used to just setting the temperature to 19degrees and forgetting it, you find it difficult to go back.
 
Reply


 
Forum Jump


All times are GMT. The time now is 20:46.