Occasional problems powering on iPod 5.5 80GB

[Zardoz]

Sometimes when I try to turn on my ipod, a short press of menu or select simply doesn't work and a hard reset is required. There is no apparent pattern to this and it has always happened across many different revisions (I'm currently using r18236.) It's not really a big issue for me, but I wondered if there was a reason for it or if perhaps the scrollwheel on my device was a bit dodgy. Has anyone else experienced similar behaviour?

[soap]

It is a known, and common, bug with Apple's own bootloader.
It is one you rarely see when using only Apple's own firmware because the iPod is most often sleeping, and rarely fully off.  Rockbox always powers all the way off, so the chances of running into this bug are multiplied dramatically.

[Zardoz]

OK I see - thanks. Is the 'deep sleep' feature of the Apple firmware implementable or even desirable in Rockbox? Or is it very wasteful of battery power? Thanks for the quick reply by the way

[soap]

Clearly it is more wasteful than a full power off - but perhaps someone will take the time to implement something similar in Rockbox one day...

[Zardoz]

thanks for the reply. something i wondered: does a full 'power on' routine cost more in terms of battery than a 'resume from sleep' - ok it obviously does, but where is the optimal point of trade off? (that's a stupid construction  ) I mean would it be better to have a unit in a deep sleep for two days and resume or to power down completely and restart?

[soap]

No easy way to test that with current conditions.
Obviously waking from sleep takes less power than booting - but sleeping takes more power than being off.
Where the balance is is hard to say.

You could put a (logging) ammeter inline with the battery of your ipod and measure the draw of the iPod sleeping in Apple OS mode.
For giggles call it .025ma.
Then measure the draw of Rockbox booting.
For giggles call it 30ma for 30 seconds.

In this example situation it would take only 10 hours of sleep to equal one boot, so when you let it sleep less than 10 hours you have a net gain.
(I think my math is right - it's late)

But outside someone doing the measurements it is all jibber jabber.

[Zardoz]

Thing is I never use the Apple OF (never have even once.) Your suggestion makes sense. I'll try that.... I should really set up a situation where I use the Apple firmware over 3 days and use Rockbox over 3 days too, under same conditions.

[Llorean]

That's not really indicative of anything, though, since it would still depend on your use habits. What might be more battery efficient for you may not be for others.

[Zardoz]

That's what I was querying - use habits.

[Llorean]

I don't understand the question then.

Basically: The longer you leave the player before resuming playback, the more likely full shutdown will benefit you. If you turn on and off more often then sleep will benefit. The point where sleep stops being beneficial and real poweroff does needs testing more precise than just guess and check.

[Zardoz]

I don't understand the question then.

Basically: The longer you leave the player before resuming playback, the more likely full shutdown will benefit you. If you turn on and off more often then sleep will benefit. The point where sleep stops being beneficial and real poweroff does needs testing more precise than just guess and check.

I was talking about the distance between restarts as applied in real world. Of course it varies. It's immeasurable. Sorry for annoying you

[Llorean]

I'm not mad, I was just curious what you were attempting to discover here since Soap had been talking about a realistic means for measuring the effectiveness and it seemed like you were proposing to just measure your personal use habits instead (which aren't really relevant overall).

[Zardoz]

True. My habits are irrelevant. I was trying to give my original post some 'resolution' rather than leave it open-ended. I've found out all I need to know. Thanks and sorry again for the 'noise'