Odd battery/rockbox behavior with r16320

[bzavala]

Hello everyone!

I'm posting this on behalf of my cousin who has a 30 GB iPod Photo (4G I beleive), since I was the one who convinced him to try out Rockbox  

Anyways... He had installed SVN 16218 from Feb 05, and he told me that after completely charging the iPod, the battery would drop impressively fast and it would turn off in about 10 mins or so. So I told him to get the latest SVN and do a battery bench.

Last sunday night we installed SVN 16320 on it. I told him to first do a test with the apple firmware. So he played an album on repeate on Monday (yesterday) and it lasted about 7 hrs. That would confirm that the battery is not bad at all.  After that, he charged the iPod again and did a battery benchmark test with Rockbox.

He sent me the battery_bench.txt file which I am attaching. Looks like it lasted for about 20 mins before it turned itself off.

Isn't this weird?  Do you spot anything wrong?

Thanks in advance,


Ben

[soap]

There is something weird going on.  Rockbox believes the iPod is not plugged in, but is charging.  Also the shutoff appeared to happen very early.  Seeing the low starting/initial voltage also makes me wonder if the battery is dying.  

< speculation >
Perhaps Rockbox is either shutting off the iPod at a higher voltage cutoff point, or perhaps Apple averages a couple of voltage readings before deciding to call the battery empty.
< /speculation >

EDIT:  Though looking again the voltage is not dipping severely on HDD access.
It is dropping like a non-linear rock, though.

[bzavala]

Soap,

Thanks for replying. So.. anything we can do? It's not a setting issue, correct? Could it be a hardware problem? Because like I mentioned, his battery lasted about 7 hrs with the apple software.. it does sound weird.

Thanks!

Ben

[soap]

I'm at the limits of my knowledge, and will try to think of who's attention should be brought to this issue.