Standby Modes

[onurbi]

Hi,

Rockbox can configure so much things (phantastic!), so am wondering, that the familiar standby mode seems not be possible, where the player consumes few power, but it comes back immediately, when switching on again. Would be good for "short time switch offs".

Did I discover this configuration not yet, or is really not possible?

Regards,

onurbi

[Llorean]

If it were possible, it would be in the manual. Much like your previous question, it's that nobody who's interested in such a feature has done the associated work. It's not as easy as it seems for this one.

[MarcGuay]

http://www.rockbox.org/tracker/task/5112

[onurbi]

>http://www.rockbox.org/tracker/task/5112

Thanks for that link. Now I know more about some opinions!

[MarcGuay]

There's definitely no shortage of those...

[dreamlayers]

What's the difficulty with implementing this?  I guess reinitializing hardware that's been affected by standby?  One way to make this easier would be to have standby happen at one particular point (eg. the main menu) rather than anywhere.

Instant wakeup is the main and maybe only significant advantage the original firmware has over Rockbox now.

[Llorean]

Tell me, do you have some special insight into how to put various parts of the hardware into standby mode rather than shutting it down?

[ryran]

dreamlayers:
What Llorean is saying.. is that we don't know how to put the ipods into standby / sleep mode. Or at least, if people have ideas about the coding, none of the work has been done.

[dreamlayers]

I'm not claiming to have special insight or information here.

I suppose that for standby you need to only power the PortalPlayer and RAM and configure them for standby.   

The power part is done using the PCF50605/PCF50607 power control unit.  It is not put into standby like currently because that turns off the output and just puts that chip in standby waiting for triggers that turn it on.  Instead it needs to be left on to provide those outputs.  According to the PCF50606 datasheet, the maximum current draw is 1.5mA, which isn't too bad.  I suppose associated components could draw more. 

There are some comments in firmware/drivers/pcf50605.c regarding what the different outputs do.  I bet the 1.8V unknown is for the RAM because that's what the RAM datasheet says.  I thought you'd also use ECO mode on the power control unit but the comments say it's not stable.  I suppose it might be stable with the low power draw encountered in suspend.

The K4S56163PF RAM datasheet talks about temperature compensated self-refresh, which I assume should be used.  The current for refreshing the whole array at 40C is 200uA (I assume µA, microamps) which seems quite reasonable. 

The PortalPlayer is the problem.  The description says it supports standby but if there's no documentation, that is a problem.  The only solution I can think of now is reverse-engineering the Apple firmware.

[Llorean]

So you've answered your own question: "What's the difficulty with implementing this?" "There's no documentation for PortalPlayer hardware."

[dreamlayers]

I wonder if it's possible to not use the PortalPlayer standby feature.  Just shutdown power to everything (including the PortalPlayer) except the RAM, and when the PortalPlayer is powered up again get control back to Rockbox via a patch to the flash. 

I just learned that some iPods can hibernate:
http://docs.info.apple.com/article.html?artnum=60945

Also an idea for figuring out how to sleep: Why not look at the diagnostics rather than the main firmware?  They are smaller and they must be easier to figure out, and they do test sleep.