<freemangordon>
blue curve is predicted, red one is measured
<freemangordon>
this is idle discharge up to 40%
<freemangordon>
and then max load to 30%
<freemangordon>
then again idle
mro has joined #maemo-leste
mro has quit [Remote host closed the connection]
mro has joined #maemo-leste
pere has quit [Ping timeout: 248 seconds]
<tmlind>
freemangordon: cool, maybe blue is measured though looking at the low-voltage range?
<freemangordon>
can't parse
<tmlind>
freemangordon: you're saying blue curve is predicted, i'm saying it looks like blue curve is measured?
arno11 has joined #maemo-leste
<freemangordon>
'predicted' as in 'calculated by the algo'
<freemangordon>
'measured' is reported by the driver
mro has quit [Remote host closed the connection]
<freemangordon>
and no, blue curve is bad at low SoC values
<freemangordon>
we have error > 10%
<tmlind>
so you're sure the red curve is measured then?
<freemangordon>
mhm
<tmlind>
ok interesting
<freemangordon>
why?
<freemangordon>
this is not OCV/Soc graphic
<tmlind>
oh what is the vertical scale then?
<freemangordon>
but SoC over time (unique values only)
<freemangordon>
vertical is %
<tmlind>
ok
<freemangordon>
horizontal is sample number
<tmlind>
ok
norayr has left #maemo-leste [Error from remote client]
<tmlind>
i thought the decline below 3.5V would be steeper
<freemangordon>
it does not reach 3.5 V
<tmlind>
oh
<freemangordon>
20% is maybe 3.7 or something
<tmlind>
ok
<tmlind>
i posted some frankenbattery photos to muru.com/linux/d4/der-franken-battery
<tmlind>
maybe somebody copy them over somewhere and i'll remove them from my low-bandwidth server
Pali has joined #maemo-leste
<freemangordon>
Wizzup: ^^^
<freemangordon>
tmlind: looks almost like mine
<tmlind>
i did not have kapton tape or black rtv silicone around.. so i used electrical tape and hotglue :)
<freemangordon>
besides I used hot silicone
<tmlind>
yeah that's better
<freemangordon>
me neither, but I salvaged some kapton from e960 itself
<tmlind>
i also soldered the wrong side of the e960 connectors first before soldering the connector to en960..
<tmlind>
massive extra blobs of solder on the photo on top :)
<freemangordon>
yeah
<tmlind>
oh well seems to work
<freemangordon>
should give you more that d day idle with chimaera
<tmlind>
probably a good idea to cut the leads in a kettle with lid handy though
<freemangordon>
day == 24hours
<freemangordon>
sec, GF's calling
<tmlind>
yeah has not recalibrated yet
<bencoh>
oh, thanks for the pictures
<tmlind>
looks like a huge amount of the m-l timers keeping system from idling properly are the neigh periodic work for network interfaces, looks like sudo ifconfig sit0 down; sudo ifconfig usb0 down cuts them down quite a bit
<tmlind>
did not ifconfig wlan0 down though
<tmlind>
i think there's some sysfs entry for disabling neigh work
arno11 has left #maemo-leste [#maemo-leste]
<bencoh>
does it keep sending neigh on usb0 even though it has no carrier (not connected)?
<tmlind>
bencho: np, maybe write some proper instructions if you decide to try it :)
mro has joined #maemo-leste
<tmlind>
sorry i mean bencoh
<tmlind>
not sure but my guess is it happens for each network interface that's up
<tmlind>
oh the neigh stuff is in /proc if somebody wants to figure out how to disable the neigh work
mro has quit [Remote host closed the connection]
parazyd has quit [Ping timeout: 246 seconds]
parazyd has joined #maemo-leste
Daaanct12 has joined #maemo-leste
Daanct12 has quit [Ping timeout: 248 seconds]
mro has joined #maemo-leste
moparisthebest has joined #maemo-leste
<tmlind>
heh maybe this helps to keep the neigh stuff low in powertop as a test hack: find /proc/sys/net/ -name interval_probe_time_ms | xargs -I% sh -c 'echo -1 > %'