On Sunday 10 February 2008 09:27, Tom Patton wrote:
...
You missed my point, that if there were nothing there, and (since) the hardware refreshes it all anyway...then THAT constitutes an energy waste...so it costs nothing extra for LINUX to hang onto possibly useful information, at NO additional expense. So the "advantage" goes to LINUX, and windows WASTES energy, refreshing empty ram.
Yes. I misunderstood. And while I still think it's a tiny energy issue, it's not null. It would be interesting to know how this compares to the energy used for extra disk activity needed to refill that RAM later (as Windows' inferior algorithms require). While the main energy consumption in a hard drive is from the spindle motor, high-performance disks with fast seek times presumably consume non-negligible energy in their voice-coil disk arm actuators.
It was rather tongue-in-cheek anyway. Considering the odds of ram cache being stale, my hypotheses could easily be disproved in a long-term test comparison.
I still think these quantities are in the noise, so it would probably be difficult to get a meaningful measurement and there would be many confounding factors (ambient temperature fluctuation causing fan speed changes, e.g., or different CPU / IO balance).
Tom in NM
Randall Schulz -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org For additional commands, e-mail: opensuse+help@opensuse.org