The Following 4 Users Say Thank You to Lehto For This Useful Post: | ||
|
2010-04-11
, 15:15
|
Posts: 946 |
Thanked: 1,650 times |
Joined on Oct 2009
@ Germany
|
#2282
|
The Following 2 Users Say Thank You to titan For This Useful Post: | ||
|
2010-04-11
, 15:29
|
Posts: 126 |
Thanked: 327 times |
Joined on Nov 2009
@ Finland
|
#2283
|
I also believe that 125Mhz isn't an advantage but we should run tests to verify it on the N900. BTW, do you whether any Sholes overclockes have already fried their devices?
The Following 7 Users Say Thank You to Lehto For This Useful Post: | ||
|
2010-04-11
, 15:30
|
Posts: 171 |
Thanked: 114 times |
Joined on Feb 2010
|
#2284
|
The Following 2 Users Say Thank You to arbitrabbit For This Useful Post: | ||
|
2010-04-11
, 15:35
|
Posts: 126 |
Thanked: 327 times |
Joined on Nov 2009
@ Finland
|
#2285
|
Doesn't particularly surprise me. That wouldn't be the case if SmartReflex worked as intended but given that it doesn't, the best solution may very well be to use 500MHz as the minimum frequency (or whatever highest we can achieve at the lowest possible voltage) and then go back to idle asap
The Following 3 Users Say Thank You to Lehto For This Useful Post: | ||
|
2010-04-11
, 15:40
|
|
Posts: 83 |
Thanked: 299 times |
Joined on Aug 2009
@ Germany, Oldenburg
|
#2286
|
titan your time_in_state script ain't working, when i execute it i get this:
...
1200 MHz: overall 0.0 %, when not idle awk: cmd. line:1: Division by zero
The Following 2 Users Say Thank You to evilJazz For This Useful Post: | ||
|
2010-04-11
, 15:49
|
Posts: 171 |
Thanked: 114 times |
Joined on Feb 2010
|
#2287
|
The Following User Says Thank You to arbitrabbit For This Useful Post: | ||
|
2010-04-11
, 15:53
|
Posts: 946 |
Thanked: 1,650 times |
Joined on Oct 2009
@ Germany
|
#2288
|
echo 500000 > /sys/devices/system/cpu/cpu0/cpufreq/scaling_min_freq
cat /sys/class/power_supply/bq27200-0/current_now
BTW for 1100mhz you can use flag 0x43.
EDIT: BTW, Im now running my phone with 500mhz idle mode on my ulv voltages and so far I've noticed that the battery life is all the same but the phone pretty much has stopped lagging and being unresponsive. Hate to say it but almost "apple-like" experience with everything just happening instantly. Apple runs their iPhone on idle at some 415mhz (I've read that on google).
The Following 5 Users Say Thank You to titan For This Useful Post: | ||
|
2010-04-11
, 15:57
|
Posts: 197 |
Thanked: 101 times |
Joined on Dec 2009
@ Netherlands
|
#2289
|
#!/bin/sh cur=`cat /sys/devices/system/cpu/cpu0/cpufreq/scaling_cur_freq` max=`cat /sys/devices/system/cpu/cpu0/cpufreq/scaling_max_freq` min=`cat /sys/devices/system/cpu/cpu0/cpufreq/scaling_min_freq` let cur=$cur/1000 let max=$max/1000 let min=$min/1000 echo "Current $cur MHz min $min max $max" #bat=`cat /sys/class/power_supply/bq27200-0/current_now` #echo "Bat $bat" cat /sys/devices/system/cpu/cpu0/cpufreq/stats/time_in_state | awk ' { freq[NR] = $1; cnt[NR] = $2 ; sum += $2} END { ni = NR while (cnt[ni] == 0) { ni-- } nonidle = sum - cnt[ni] for (i = 1; i <= NR; i++) { printf "%4d MHz overall %6d %4.1f%%", freq[i]/1000, cnt[i], cnt[i]*100.0/sum if (i<ni) { printf " nonidle %4.1f%%", cnt[i]*100.0/nonidle } printf "\n" } } '
Current 250 MHz min 125 max 800 1200 MHz overall 0 0.0% nonidle 0.0% 1100 MHz overall 0 0.0% nonidle 0.0% 1000 MHz overall 0 0.0% nonidle 0.0% 950 MHz overall 0 0.0% nonidle 0.0% 900 MHz overall 0 0.0% nonidle 0.0% 850 MHz overall 0 0.0% nonidle 0.0% 810 MHz overall 30019 5.6% nonidle 48.8% 750 MHz overall 386 0.1% nonidle 0.6% 700 MHz overall 675 0.1% nonidle 1.1% 600 MHz overall 7960 1.5% nonidle 12.9% 550 MHz overall 475 0.1% nonidle 0.8% 500 MHz overall 22007 4.1% nonidle 35.8% 250 MHz overall 473821 88.5% 124 MHz overall 0 0.0%
The Following User Says Thank You to mirakels For This Useful Post: | ||
|
2010-04-11
, 16:00
|
Posts: 946 |
Thanked: 1,650 times |
Joined on Oct 2009
@ Germany
|
#2290
|
The Following 3 Users Say Thank You to titan For This Useful Post: | ||
|
My OC kernel collection