maemo.org - Talk - Battery life not good -- problem?

maemo.org - Talk (https://talk.maemo.org/index.php)

- Nokia N800 (https://talk.maemo.org/forumdisplay.php?f=25)

- - Battery life not good -- problem? (https://talk.maemo.org/showthread.php?t=25683)

Re: Battery life not good -- problem?

seems cheap SD cards can have a real impact.

removed mine and im looking at .5% pr hour so far, vs the 3% i was seeing before...

Re: Battery life not good -- problem?

Quote:

Originally Posted by tso (Post 251712)

still, that does not explain how im getting a calculated standby time of 10 days, a figure i have no chance of reaching...

Calculated standby time is based on a constant factor of 10%/day or so, and is independent of actual idle draw. Calculated active time is based on current draw. (Of course, this is all done in closed source, so I may be wrong, but that's what it looks like to me.)

Re: Battery life not good -- problem?

Quote:

Originally Posted by tso (Post 251808)

seems cheap SD cards can have a real impact.
removed mine and im looking at .5% pr hour so far, vs the 3% i was seeing before...

I didn't have to remove mine (yet) - just disabling swap was enough, although I didn't see a change until I rebooted (after installing pyBattery, a really nifty tool !).

Re: Battery life not good -- problem?

well im going to pop those cards in tomorrow, and see if there is much of a change.

(or maybe tonight as i may need my ebook fix)

Re: Battery life not good -- problem?

Hmm and why your devices are reading/writing to swap while sleeping...? :)

Re: Battery life not good -- problem?

Same here. Battery duration declined since Diablo. In the meantime, my original charger has died, and I'm using a cheap third-party replacement, it does the work, but I'm wondering if it's not at the expense of my batteries' lifespan.
With so many factors affecting battery life, I find it difficult to determine the cause of the poor performance.

Re: Battery life not good -- problem?

yep, especially when the tablet cant run tools like powertop :(

Re: Battery life not good -- problem?

I have been having battery life problem since the first update to diablo. When I updated to diablo originally, I was seeing about 5 days of battery life with an average of 20-25 minutes of wifi usage and another 20 minutes or so of tablet usage without wifi every day. Some time after the the first SSU to diablo, my battery life dropped to about 2 days with the same usage pattern. After 2 days I would get a warning about battery level being very low and I used to plug it into charger at this point. I would also get low battery warning after about 1.5 hour of wifi usage even if battery was fully charged before I started using wifi.

My wife's N800 is running original diablo release and she sees 4+ days of battery life. I swapped batteries with her to see if my battery was just getting old, but her battery in my N800 also lasted just about 2 days. To narrow down which software was causing battery drain, I reflashed to the latest diablo release (5.2008.43-7), restored my old settings from backup but did not install all the software I had on my N800 before reflash. I installed advanced-power to monitor battery level. I noticed that after 7 hours of up time and about 1 hour 25 minutes of wifi usage, my battery level was still about 65%. I continued to use wifi and 10 minutes later, I got a low battery warning. Advanced-power showed 0.3% battery level. Battery level dropping from 65% to 0.3% over 10 minutes was just too suspicious. I decided to keep using it until N800 just dies to see how long the battery would last after the warning. I used wifi for another 20 minutes and battery stayed at 0.3%. I then shut down wifi and left N800 in standby overnight. Next morning advanced power showed a battery level of 37%!! I again used wifi for another 10 minutes and battery power slowly dropped to 36.1%.

Based upon my observation, I am not sure if charge level being reported is correct. Advanced power applet gets its data from D-Bus. So it is not getting good data from dbus service.

Re: Battery life not good -- problem?

so basically your saying there is some bug running around inside the battery driver, or something like that?

could be, altho i recall being told that adv power uses the mah report to calculate the percentage. but i could have misunderstood or remembered wrong.

Re: Battery life not good -- problem?

Quote:

Originally Posted by fpp (Post 251143)

You mean this kind ? :

Code:

menelaus 1-0072: Setting voltage 'VMMC' to 3000 mV (reg 0x0a, val 0xb8)

I have some of those too. But almost all my dmesg entries (especially since the last SSU) are thus :

Code:

EAC mode: play enabled, rec enabled

Which app does that ?
(and again, my SD and swap were fine with Chinook...)

Those messages are annoying. I patched the kernel to get rid of them:

Quote:

diff -Naur kernel-source.orig/drivers/i2c/chips/menelaus.c kernel-source/drivers/i2c/chips/menelaus.c
--- kernel-source.orig/drivers/i2c/chips/menelaus.c 2008-12-07 17:11:33.000000000 +0000
+++ kernel-source/drivers/i2c/chips/menelaus.c 2008-12-07 16:31:50.000000000 +0000
@@ -477,7 +477,7 @@
if (ret < 0)
return ret;

- dev_info(&menelaus.client.dev,
+ dev_dbg(&menelaus.client.dev,
"Setting voltage '%s' to %d mV (reg 0x%02x, val 0x%02x)\n",
vtg->name, mV, vtg->vtg_reg, val);
diff -Naur kernel-source.orig/sound/arm/omap/eac.c kernel-source/sound/arm/omap/eac.c
--- kernel-source.orig/sound/arm/omap/eac.c 2008-12-07 17:11:33.000000000 +0000
+++ kernel-source/sound/arm/omap/eac.c 2008-12-07 17:08:07.000000000 +0000
@@ -746,11 +746,13 @@
{
struct omap_eac *eac = dev_get_drvdata(dev);

+#if 0
#ifdef DEBUG
printk(KERN_DEBUG "EAC mode: play %s, rec %s\n",
play ? "enabled" : "disabled",
rec ? "enabled" : "disabled");
#endif
+#endif
BUG_ON(eac == NULL);
mutex_lock(&eac->mutex);
if (play || rec) {

(the menulaus patch is by fanoush)