Open GoogleCodeExporter opened 8 years ago
Battery monitoring is valid and listed already in Issue 4.
I am not really sure where the hwmonitor has the "power sensor" data from. My
guess is, it uses a reading of the average performance-state (P-state) or just
the (average) multiplier and interpolates over a power consumption table with
reference values for each p-state or multiplier. But this would yield more of a
theoretical power consumption value, than an actual measurement.
Some mainboards have actual current and voltage monitoring chips in the voltage
regulator circuits, and from these the power consumption could be calculated as
well. But hwmonitor shows these seperately (and often with completely different
values).
The hwmonitor power value is only useful, if you can not find two different CPU
stress tools which produce the same power value in hwmonitor but different
steady state core temperatures. Maybe you can try LinX and some other less
agressiv stress tool (which loads only one core) and see if you can get the
same power value, but obviously different core temperatures. Perhaps you need
to disable turbo mode for this.
Original comment by moel.mich
on 19 Jul 2010 at 7:58
Original comment by moel.mich
on 3 Oct 2010 at 5:45
Original issue reported on code.google.com by
tobias.h...@googlemail.com
on 19 Jul 2010 at 6:46Attachments: