Save iPod battery life by turning off in Apple OS?

[desertmusic]

Hi,

    I have an ipod 5.5g 80g with rockbox installed and have been loving it.  I've followed the forum discussions on battery life for the ipods and read through the rockbox manual.  I've noticed battery drain between turning off the ipod in rockbox and restarting rockbox.  I know this has been covered elsewhere.  However, I didn't see any mention of my specific question, so I thought I'd ask.

   Suppose I reboot into the Apple OS and shut down from there?  Can I assume that this would eliminate battery drain during off time?   Other than whatever normal drain the original Apple OS causes?  I'm just wondering if I can avoid the sometimes dramatic battery drainage I seem to encounter every so often between turning off and on inside Rockbox.

Thanks,
John.

[bluebrother]

Suppose I reboot into the Apple OS and shut down from there?  Can I assume that this would eliminate battery drain during off time?

No. When Rockbox shuts down the player it is powered off. AppleOS puts it into a mode known as "deep sleep". In that mode the player is still running but at a very low clock speed, thus requiring still power. And when it's turned of its ... off

[riksweeney]

No. When Rockbox shuts down the player it is powered off. AppleOS puts it into a mode known as "deep sleep". In that mode the player is still running but at a very low clock speed, thus requiring still power. And when it's turned of its ... off

My God, Apple do all kinds of crazy things don't they?! Shouldn't off mean off?

Anyway, hasn't the Rockbox battery life on the iPod been solved (or significantly improved), or doesn't it apply to this model?

[GodEater]

The battery life fixes applied only to 1st, 2nd and 3rd gen ipods.

[bluebrother]

My God, Apple do all kinds of crazy things don't they?! Shouldn't off mean off?

Well, that heavily depends on what you want -- waking up from a sleep mode is *much* faster than a complete restart, similar to booting a laptop vs. waking up from suspend to RAM / disk. So regarding the user experience this makes quite sense.

[countach]

Hi,

    I have an ipod 5.5g 80g with rockbox installed and have been loving it.  I've followed the forum discussions on battery life for the ipods and read through the rockbox manual.  I've noticed battery drain between turning off the ipod in rockbox and restarting rockbox.  I know this has been covered elsewhere.  However, I didn't see any mention of my specific question, so I thought I'd ask.

   Suppose I reboot into the Apple OS and shut down from there?  Can I assume that this would eliminate battery drain during off time?   Other than whatever normal drain the original Apple OS causes?  I'm just wondering if I can avoid the sometimes dramatic battery drainage I seem to encounter every so often between turning off and on inside Rockbox.

Thanks,
John.

I confirm that shutting down in rockbox mode and leaving it alone for some days, drains all the battery. The three lasts weekends I left the iPod untouched on the desk and the monday morning was totally drained.
The keypad-off switch was enabled, so there was no incidental power-on, and if for any reason it powered by itself, it didn't start en Apple Mode, despite being activate that switch.
So I think this is clearly some kind of bug.

Sometimes I find the device playing by itself. Even if I set the start-up mode to 'Main Menu'. In that case, if I switch it on manually (pressing the MENU key), it goes to the Main Menu, that's ok. But when starting by 'itself', it starts playing last song, so there it is the drain cause.

For the moment, the only way to secure the battery in a long-time non-using is entering in iPod mode before and then shutdown from there.

[Llorean]

If it starts up playing, it sounds like you've got an alarm set, or there's a bug with the alarm... Which SVN revision was the last one you experienced this problem with?

[countach]

I use the last one DevilG Build, evilg-5G-64-20070914.zip file.

About the alarm, the Screen Displays:

Alarm Time: 23:15

SELECT=Set MENU=Cancel

I never set up the alarm, and pressing MENU an 'Alarm Disabled' TEXT appears, but entering again, it seems that the Alarm is still enabled, as the screen remains with the main data.

There is a video I made right now:
http://download.yousendit.com/F9533D732F08914A

Notice the fourth icon in the upper line: ¿Is is the alarm-on indicator or the random playlist setting?

But I can assure that no alarm was set up ever.

[Llorean]

That is an UNSUPPORTED BUILD. Please do not post in the official forums about unofficial builds, even if your problem is similar. This is in the posting guidelines.

[countach]

Sorry for that, I know. I'll install the last official build and wachout if it happen too, but I pretty sure that will do...

About that unsupported build... "please please please... (smilies in knees) " do include the cover art thingy in the official builds !!  
I know you always says "No   " every time is asked... I suppose there have to be some great reason for that big nono...

[Llorean]

The reason has been discussed in many places, we've made no secret of why we don't include it yet. Please use the search feature, as per the posting guidelines.

In fact, go and read them now, it seems you've forgotten a few of them.

It really doesn't matter if you're sure it will happen with the official build or not, unsupported builds are running older code, and have different code, so they can have completely different causes for the same problems.

[countach]

Ok, let me apologize for my laziness about finding the reason about the album art issue. I've made now a deep search and I found a lot of people asking the same, and I think this thread explain pretty well the situation:

http://forums.rockbox.org/index.php?topic=9392.0

In a few words, I understand that the RB team is not happy about the quality of the current art solution and the one that would be (if so) will be different. Great news, so let's wait.

About the posting guidelines I've readed them again and aside the non-sopported build question, (for wich one I apologize, in fact, i always used its thread to discuss issues about the build), I think I'm following them. As a non-english speaker, maybe my text is weird sometimes, but I try to keep it clean...



So, returning to the topic, I will install the official build and will keep track of this self-startup thing, posting the results next week in this thread.

Again, thank you for the effort, I only want to contribute in any way I can. I'm a coder too, and for sure if I had time will be looking at the sources... in the meanwhile, I'm making some Themes.

[TexasRockbox]

Some current is needed to keep the clock running and accurate also the on/off isn't like a physical on/off switch to connectg/disconnect the battery but instead senses on when a button is pressed.  That type of on/off takes continous current as well.

[scharkalvin]

Some insight into battery powered devices from a software developer.....

Most computer controlled devices do not really power down but enter a VERY low power mode.  Depending on the hardware this can be somewhere between a few milliamperes to tens of microamperes.   The cpu makers call this mode 'power down', 'power save', or 'sleep' mode.  In the lowest power modes the cpu clocks stop completely and a few interrupt inputs are left armed.  The cpu comes back on line when an external interrupt is detected (from the user interface).  If time keeping is desired, an external cmos tod clock chip is used.  Higher power sleep modes just slow down the clock.  Since the usual clock speed used is in the range of megahertz to hundreds of megahertz a real slow clock rate of 32khz is sometimes used.  At this rate the cpu can keep track of the TOD, but not much else, but power consumption in this mode might be in the milliampere range.

Aside from the power down current drain we also have to worry about the batteries self discharge rate.  This is quite high for NiCads and Nimh types resulting in a shelf life of about a month or so (for AA size batteries).  LiIon types have a much lower self discharge rate and can hold a charge for a year or more.  But LiIon batteries can't be allowed to drain below a critical point or they will require reforming before they will accept a full charge again.  This makes a good battery monitor / smart charger circuit a critical part of any device powered by a LiIon battery.  As a result most LiIon powered devices do NOT implement the charger in software, a separate charger chip is used.