The way this is usually done is with a hardware battery emulator to
remove the variability of the battery over lifetime and temperature.
Even if the emulator holds a constant voltage, it can tell you if
current consumption during a particular test has changed significantly.
--m.
On 13-02-01 10:33 AM, Andrew Halberstadt wrote:
Battery life regression is something that is very tricky to test. It
is very dependent on the hardware you use and how you use it. Since
all batteries degrade over time, it's also difficult to tell if there
was a regression, or the battery needs to be replaced. I'm not saying
it's impossible or anything, but it would be tricky to interpret the
data correctly and it would be very difficult to get tests like these
automated properly. We can't really even talk about battery life tests
until we address the issue Naoki brought up in
https://bugzilla.mozilla.org/show_bug.cgi?id=814193#c1
This is the kind of thing that I imagine each OEM would have testing
for (and probably in a much better environment/setup than we could
hope to manage). The easiest thing we could do would be to communicate
with them and make sure they let us know if they notice any worrying
regressions.
-Andrew
On 01/31/2013 04:14 PM, Gary Kwong wrote:
I was wondering if we have (or already have) plans for testing battery
life on some standard phone hardware now that we are about to have the
first release of Firefox OS finalised.
Battery life is a large concern for mobile/smartphone users and it will
be interesting to have some areweusingtoomuchbattery.com site (or
something similar) to track battery life of different Firefox OS
versions going forward.
Thoughts? Has any bug been filed, if needed?
Regards,
-Gary
_______________________________________________
dev-b2g mailing list
[email protected]
https://lists.mozilla.org/listinfo/dev-b2g
_______________________________________________
dev-b2g mailing list
[email protected]
https://lists.mozilla.org/listinfo/dev-b2g
