Active Topics
-
Porting apps to Leste (32)
to Maemo 7 / Leste by Arno_11 - 4 days, 9 hrs ago -
[ANNOUNCE] CSSU-thumb thread - stable Thumb2 on N900 (2,264)
to Maemo 5 / Fremantle by panjgoori - 6 days, 4 hrs ago - more...
|
2011-12-03
, 11:40
|
|
Posts: 1,680 |
Thanked: 3,685 times |
Joined on Jan 2011
|
#2422
|
eh? Since kp49 smart reflex hoons like a boss. These battery drain issues are not SR related. I am just interested in your case as you seem to be a unique one.
__________________
N900: One of God's own prototypes. A high-powered mutant of some kind never even considered for mass production. Too weird to live, and too rare to die.
|
|
2011-12-03
, 18:59
|
|
Posts: 3,074 |
Thanked: 12,960 times |
Joined on Mar 2010
@ Sofia,Bulgaria
|
#2423
|
Originally Posted by vi_
Well, what some devices cannot take is DSP overclocking and SR enabled. More precisely:
Does that suggest SR is still a kind of folklore? Some devices can take it while some cannot? Perhaps some kind of 'fudge factor' set in /sys/power/ which is added to SR value. i.e. SR value+fudge factor (1, 2, 3, etc) to accommodate individual pieces of hardware?
We need results in order to conclude anything.
dsp kernel profile sets DSP clock to be 600 for MPU 805. Some devices cannot handle that. The solution is to move DSP overclocking to higher MPU frequencies, for example: (720,480) (805,560) (850,600) , where first comes MPU frequency, second - DSP. And maybe some devices cannot handle 600 for DSP at all or at least not bellow MPU@900. Nothing to be done here.
| The Following 4 Users Say Thank You to freemangordon For This Useful Post: | ||
|
|
2011-12-03
, 21:20
|
|
Posts: 638 |
Thanked: 1,692 times |
Joined on Aug 2009
|
#2424
|
With my device, (2204) i confirm that all the problems are solved with vdd1=0 and the default profile.(no overclock)
NO instability, no problems recording/playing video, perfectly smooth and no battery drain.
As soon as i enable vdd1, i cannot play any video, cannot record, many dsp faults and battery drain...
Going back and setting vdd1=1 all the issues start again...
So, same phone, same apps, same clock profile.... only one bit of difference between one working phone and one apparently faulty device.
Last edited by xes; 2011-12-03 at 21:24.
NO instability, no problems recording/playing video, perfectly smooth and no battery drain.
As soon as i enable vdd1, i cannot play any video, cannot record, many dsp faults and battery drain...
Going back and setting vdd1=1 all the issues start again...
So, same phone, same apps, same clock profile.... only one bit of difference between one working phone and one apparently faulty device.
Last edited by xes; 2011-12-03 at 21:24.
| The Following User Says Thank You to xes For This Useful Post: | ||
|
|
2011-12-03
, 22:52
|
|
Posts: 1,808 |
Thanked: 4,272 times |
Joined on Feb 2011
@ Germany
|
#2425
|
@xes,
There you go. Now enjoy your working phone!
There you go. Now enjoy your working phone!
|
|
2011-12-04
, 07:51
|
|
Posts: 3,074 |
Thanked: 12,960 times |
Joined on Mar 2010
@ Sofia,Bulgaria
|
#2426
|
Originally Posted by xes
Well, that is strange, seems my SR re-calculations will need an adjustment for the next KP. In that regard could you please post your "/sys/power/efuse_vdd1" values, along with voltages SR calculates when enabled.
With my device, (2204) i confirm that all the problems are solved with vdd1=0 and the default profile.(no overclock)
NO instability, no problems recording/playing video, perfectly smooth and no battery drain.
As soon as i enable vdd1, i cannot play any video, cannot record, many dsp faults and battery drain...
Going back and setting vdd1=1 all the issues start again...
So, same phone, same apps, same clock profile.... only one bit of difference between one working phone and one apparently faulty device.
|
|
2011-12-04
, 08:49
|
|
Posts: 112 |
Thanked: 92 times |
Joined on Nov 2010
|
#2427
|
Same here, as i previously posted, my device too (2104) cant pay video steadily with SR even for stock clocks, SD video just quits while playing back.
Also i didn't notice any improvement on battery life for the past couple of weeks with SR on, seemed it drained faster.
SR is not working for some devices for sure, so I'm back to using without SR until you hopefully update with new calibrations...etc
The issue is severe when battery level is low (close to emply) video just don't play for more than a few seconds or don't play at all.
I'm not too sure how power regulators are programmed/designed and whether there any load line calibration in place, instability might be caused because of a too high v drop when the mpu and dsp is loaded, i suspect this since the volts supplied is already very low.
Last edited by nkirk; 2011-12-04 at 15:54. Reason: :}
Also i didn't notice any improvement on battery life for the past couple of weeks with SR on, seemed it drained faster.
SR is not working for some devices for sure, so I'm back to using without SR until you hopefully update with new calibrations...etc
The issue is severe when battery level is low (close to emply) video just don't play for more than a few seconds or don't play at all.
I'm not too sure how power regulators are programmed/designed and whether there any load line calibration in place, instability might be caused because of a too high v drop when the mpu and dsp is loaded, i suspect this since the volts supplied is already very low.
Last edited by nkirk; 2011-12-04 at 15:54. Reason: :}
|
|
2011-12-04
, 11:40
|
|
Posts: 1,225 |
Thanked: 1,905 times |
Joined on Feb 2011
@ Quezon City, Philippines
|
#2428
|
Turning on SRVDD1 with kp49 from the repos makes DSP bomb out on undervoltage. Doesn't reboot the system though, so I'll give it that.
Oh well, it seems that my N900 just can't take SRVDD1.
Oh well, it seems that my N900 just can't take SRVDD1.
__________________
N9 PR 1.3 Open Mode + kernel-plus for Harmattan
@kenweknot, working on Glacier for Nemo.
N9 PR 1.3 Open Mode + kernel-plus for Harmattan
@kenweknot, working on Glacier for Nemo.
|
|
2011-12-05
, 10:10
|
|
Posts: 561 |
Thanked: 75 times |
Joined on Jan 2010
@ Spain
|
#2429
|
Originally Posted by pali
Is not it possible to adapt NTFS-3G or any of the modules that support modules on today in the distributions?
multiboot-kernel-power is obsolated by package kernel-power-bootimg. So you need to install only kernel-power-bootimg!
Only if sombody write kernel driver for that...
NTFS-3G
http://en.wikipedia.org/wiki/NTFS-3G
|
|
2011-12-05
, 10:25
|
|
Posts: 11 |
Thanked: 6 times |
Joined on Sep 2011
@ Stockport, UK
|
#2430
|
Hi,
I don't know if this is the correct thread, but as there were some battery drain problems in this thread before, I decided to post it here first.
This Friday evening, I charged my phone completely.
I then had to do a 999 call during the night at 2. Afterwards, I changed my profile to General in case they wanted to call me back.
When I woke up the next morning, my battery was drained.
Batterygraph showed a constant CPU usage.
Top showed the process pulseaudio to consume app. 13% of CPU.
I obviously cannot reproduce this (fake 999 calls are not really what I want to do), just to note it down somewhere so somebody can look into it.
Regards,
Jan
I don't know if this is the correct thread, but as there were some battery drain problems in this thread before, I decided to post it here first.
This Friday evening, I charged my phone completely.
I then had to do a 999 call during the night at 2. Afterwards, I changed my profile to General in case they wanted to call me back.
When I woke up the next morning, my battery was drained.
Batterygraph showed a constant CPU usage.
Top showed the process pulseaudio to consume app. 13% of CPU.
I obviously cannot reproduce this (fake 999 calls are not really what I want to do), just to note it down somewhere so somebody can look into it.
Regards,
Jan
 |
| Tags |
| battery-status, bq27x00_battery, kernel, kernel-power, misiak4king, noobs-cant-read, pali4president, patches, readdirections, revolverspinyou |
«
Previous Thread
|
Next Thread
»
|
All times are GMT. The time now is 21:06.
That's what I've done. No VDD1, no VDD2. My battery behaves consistently good (up to 5 days with little use). KP47, LV, 805Mhz max and no CSSU or crap-patches.
For all I care, kernel support as well as hardware itself is buggy when it comes to SmartReflex.
I know it's unsatisfying (not being able to fix a bug), but I would suggest forgetting about SmartReflex. Perhaps this would open more "resources" to fix and further develop compcache.