Improved iPod battery life

[riksweeney]

Hi,

I was just wondering what was done to improve the battery life on the iPods. I've had a quick look at a diff on the system-pp5002.c file and the comment on one of the "offending" lines is

/* don't clock unused PP5002 hardware components */

I was just wondering what this means. I know this is geeky but this is the kind of thing that interests me (the same way that I find what people do when coding MAME more interesting than the games themselves).

[LambdaCalculus]

Don't worry about it being geeky. We're all geeks here!

Which number line is that in the code? I'm going to take a look at it in a few minutes.

[GodEater]

I was just wondering what was done to improve the battery life on the iPods.

Bear in mind, this isn't ALL ipods, only 1st, 2nd and 3rd Gen iPods.

It doesn't include either generation Mini, the Nano, or the 4th, 5th and 5.5th Gen iPods.

[LambdaCalculus]

I know it won't help my iPod out, but it's always good to learn something every day.

These are the lines riksweeney's inquiring mind wants to know about:

#if defined(IPOD_1G2G) | | defined(IPOD_3G)
 Â         DEV_EN = 0x0b9f; /* don't clock unused PP5002 hardware components *?
 Â         outl(0x0035, 0xcf005004); /* DEV_EN2 ? *?
#endif

Could it be the pointers given that are the cause?

[riksweeney]

#if defined(IPOD_1G2G) | | defined(IPOD_3G)
          DEV_EN = 0x0b9f; /* don't clock unused PP5002 hardware components *?
          outl(0x0035, 0xcf005004); /* DEV_EN2 ? *?
#endif

I guess that it's the outl() function that is improving the battery life. I'll have a look in the function to see if I can understand what it does myself!

[LambdaCalculus]

I think you may be on to something there, my friend!  

[AlexP]

I believe that by disasembling the original firmware for 1,2,3rd gen iPods, Amiconn discovered some additional hardware initialisation routines to turn off uneeded hardware in the portalplayer chip and therefore save power.

Now whilst it isn't for sure, most people believe similar additional hardware needs turning off to improve the battery life for newer iPods.  However it is a similar yet different and more complicated chip in the newer iPods, hence the improvements cannot be ported straight across.

[elborak]

It's simply turning off parts of the chip that aren't used by Rockbox.

Without detailed docs on a component, figuring out what port controls what and what bits do what is very taxing work.

[LambdaCalculus]

I remember this being discussed before. We need to look at precisely what needs to be turned off in the newer PortalPlayer chips that isn't needed by Rockbox.

BigBambi, a reading of system-pp502x.c shows me no lines that are similar to the lines pointed out from system-pp5002.c by myself and riksweeney, so your belief may be true. The process seems a bit different. Any ideas?

(Note: If I'm wrong, I'm wrong. I'm taking this as a learning process.)

[GodEater]

BigBambi, a reading of system-pp502x.c shows me no lines that are similar to the lines pointed out from system-pp5002.c by myself and riksweeney, so your belief may be true. The process seems a bit different. Any ideas?

That would be because the register addresses on the 502x aren't the same as the 5002, and therefore we :

a) don't know where to write the values to turn stuff off
and
b) don't know what values to write even if we did know the write register addresses.

[LambdaCalculus]

I had a hunch it was due to lack of register addresses. Thanks for that information, GodEater. You and BigBambi are bringing back my C knowledge from the depths of my subconscious!

[AlexP]

Ha, I know next to no C, I just pick up the broad reasons.  

[LambdaCalculus]

Don't feel bad. My C knowledge is fair, but rusty. I'm out of practice and I'm trying to get back into the swing of things. But it's always good to learn and share information with my fellow peers.