Hi there,

One of my tasks on the Desktop Power Consumption blueprint
(https://blueprints.launchpad.net/ubuntu/+spec/desktop-p-power-consumption)
was to analyse mouse power consumption.


I have run a set of tests and the data (.ods Spreadsheet) with some
analysis (results.txt) are available here:
http://zinc.canonical.com/~cking/power-benchmarking/mouse-movement/


The bottom line is that halving the mouse rate does not really save much
power at all. It appears most of the power consumption is in the
rendering of the mouse movement which varies depending on the graphics
driver and this really isn't influenced by the mouse input.


My recommendation is not to reduce the mouse rate. Instead we need to
see why some systems seem to consume a lot of cycles just rendering a
mouse movement. This is a separate task and maybe out of scope for this
blueprint.


Colin



--
kernel-team mailing list
kernel-team@lists.ubuntu.com
https://lists.ubuntu.com/mailman/listinfo/kernel-team

[Being re-sent as I'm unsure it went out last time I tried]

Hi there,

One of my tasks on the Desktop Power Consumption blueprint
(https://blueprints.launchpad.net/ubuntu/+spec/desktop-p-power-consumption)
was to analyse mouse power consumption.


I have run a set of tests and the data (.ods Spreadsheet) with some
analysis (results.txt) are available here:
http://zinc.canonical.com/~cking/power-benchmarking/mouse-movement/


The bottom line is that halving the mouse rate does not really save much
power at all. It appears most of the power consumption is in the
rendering of the mouse movement which varies depending on the graphics
driver and this really isn't influenced by the mouse input.


My recommendation is not to reduce the mouse rate. Instead we need to
see why some systems seem to consume a lot of cycles just rendering a
mouse movement. This is a separate task and maybe out of scope for this
blueprint.


Colin

--
kernel-team mailing list
kernel-team@lists.ubuntu.com
https://lists.ubuntu.com/mailman/listinfo/kernel-team

Hello Colin,

Colin Ian King [2011-12-09 19:07 +0000]:
> I have run a set of tests and the data (.ods Spreadsheet) with some
> analysis (results.txt) are available here:
> http://zinc.canonical.com/~cking/power-benchmarking/mouse-movement/

Many thanks for doing this!

> My recommendation is not to reduce the mouse rate. Instead we need
> to see why some systems seem to consume a lot of cycles just
> rendering a mouse movement. This is a separate task and maybe out of
> scope for this blueprint.

Agreed, so let's not fiddle with the mouse frequency, it might cause
more damage than improvement.

Martin
--
Martin Pitt | http://www.piware.de
Ubuntu Developer (www.ubuntu.com) | Debian Developer (www.debian.org)

--
kernel-team mailing list
kernel-team@lists.ubuntu.com
https://lists.ubuntu.com/mailman/listinfo/kernel-team