I've been trying powenowd for a few hours now, it does seem to work better (frequency does not change in a second, and I like the scaling approach better, as opposed to cpudyn's maximum or minimum thing).
However, I still have not seen a reduction in temperature, which is the main reason I wanted this. Plus, I still find the discrepancy between the frequency reported by /sys/devices/system/cpu/cpu0/cpufreq/scaling_setspeed (which matches the one reported in /proc/cpuinfo) and the one reported by x86info, culled from the /dev/cpu/0/cpuid, which does not go down very much (only occasionally and then only to around 1200MHz at the most). Which one should I trust? On Mon, Apr 26, 2004 at 10:29:15PM +0100, Pete Wright wrote: > I found that using powernowd instead of cpufreq worked like a charm. I > also found that gkrellm was not a good indicator of the processors > current speed - you get much better results by just 'cat'ing the > relevent file in /sys/devices/system/cpu/cpu0/cpufreq > > Hope that helps some > > Pete > -- Ivan Fernández [EMAIL PROTECTED]
