Pyra standby time


_wb_

Microbe
Staff member
Joined
Apr 5, 2012
Messages
5,390
Age
42
Location
Brussels, Belgium
I was reading some old thread and noticed this remark by notaz:
 

That said there are some things that can't be properly turned off on pandora, like the nubs or USB host VBUS power supply. I complained about that to MWeston but it was too late, when I got power saving stuff working ED was already setting up rebirth production and trying to fix that would have resulted in months of delays. When the board is changed some prototypes need to be made to be sure changes are good, the bare board has delay over 1 month, then GC would need to reprogram machines just for a few test boards and so on (maybe ED can clarify), so it would be even harmful to do now, not to mention extra money needed for this.

Edit: it's unclear how much benefit those fixes would give though, nubs supposedly have their own automatic power saving modes and USB 5V supply in not loaded when nothing is connected.
Is this something that will be taken into account for the Pyra?

The Pandora currently consumes ~110mW when being completely idle and the lid is closed. Or ~95mW if the power led is disabled. What would it take to bring that number further down on the Pyra?

It would be awesome if the standby time of the Pyra could be even better than that of the Pandora.

We will have more RAM, but it will be lower power RAM, so what will the net effect be?

What about the other components?
 
OMAP5 doesn't support LPDDR2. Not the one still in production anyway.

It seems I was wrong

Although the SoC isn't fixed yet, OMAP5 once again seems like the most likely one?
 
Last edited by a moderator:
OMAP5 doesn't support LPDDR2. Not the one still in production anyway.

Although the SoC isn't fixed yet, OMAP5 once again seems like the most likely one?
Are you sure? I thought the OMAP5430 only supported LPDDR2, while the OMAP5432 supports both LPDDR2 and DDR3L. I could be wrong though.
 
OMAP5 doesn't support LPDDR2. Not the one still in production anyway.

Although the SoC isn't fixed yet, OMAP5 once again seems like the most likely one?
Are you sure? I thought the OMAP5430 only supported LPDDR2, while the OMAP5432 supports both LPDDR2 and DDR3L. I could be wrong though.
Really? I can't say I'm 100% sure about it not supporting it, I did not see any mentioning of it when I last read about it though.

EDIT: It seems I was wrong. Nevermind.
 
Last edited by a moderator:
OMAP5 doesn't support LPDDR2. Not the one still in production anyway.

Although the SoC isn't fixed yet, OMAP5 once again seems like the most likely one?
Are you sure? I thought the OMAP5430 only supported LPDDR2, while the OMAP5432 supports both LPDDR2 and DDR3L. I could be wrong though.
Really? I can't say I'm 100% sure about it not supporting it, I did not see any mentioning of it when I last read about it though.

EDIT: It seems I was wrong. Nevermind.
Are you sure? I think I was actually wrong, I can't find anything claiming OMAP5432 supports LPDDR2.

Anyway, how does DDR3L compare to the Pandora RAM ?
 
Last edited by a moderator:
wiki says it supports both, think the best bet is LPDDR2-533mhz

http://en.wikipedia.org/wiki/OMAP#OMAP_5

I think the pandora only had DDR(1st gen) -333mhz so significantly faster ram and less power.

can't find any direct comparisons on power consumption but I'm guessing it will be significant

EDIT: LPDDR2 works at 1.2v where DDR3L is 1.35v

http://en.wikipedia.org/wiki/Mobile_DDR#LPDDR2

http://en.wikipedia.org/wiki/DDR3_SDRAM#DDR3L_and_DDR3U
Actually the wikipedia and all sources it references say that OMAP5430 supports LPDDR2 and OMAP5432 supports DDR3/DDR3L. See e.g. page 3 of http://focus.ti.com/pdfs/wtbu/OMAP5_2011-7-13.pdf
 
The difference in voltage seems rather small, DDR3L might be the best option there anyway if LPDDR2 is still compatible, the extra bandwidth would justify it I think
 
Last edited by a moderator:
Voltage is not the only thing; LPDDRx has more power-saving states than DDRxL; the former is aimed at smartphones and tablets, while the latter is aimed at laptops and embedded systems.

But I assume DDR3L should still be more efficient than what we have in the Pandora.
 
the specification document (which is still sometimes wrong) page 303 said:
The OMAP5432 device is offered in a 754-ball, 17 × 17 -mm package, which does not support LPDDR2

memory stacking but instead can interface DDR3 on-board memories.
So no, 5430 supports LPDDR2 (but not DDR3) and 5432 supports DDR3 (but not LPDDR2)
 
Does the Pandora currently support hibernation? Is it possible to save everything to swap for better power saving? Or would that not give much advantage over simply shutting down the whole unit?
 
I was reading some old thread and noticed this remark by notaz:

That said there are some things that can't be properly turned off on pandora, like the nubs or USB host VBUS power supply. I complained about that to MWeston but it was too late, when I got power saving stuff working ED was already setting up rebirth production and trying to fix that would have resulted in months of delays. When the board is changed some prototypes need to be made to be sure changes are good, the bare board has delay over 1 month, then GC would need to reprogram machines just for a few test boards and so on (maybe ED can clarify), so it would be even harmful to do now, not to mention extra money needed for this.


Edit: it's unclear how much benefit those fixes would give though, nubs supposedly have their own automatic power saving modes and USB 5V supply in not loaded when nothing is connected.
Is this something that will be taken into account for the Pyra?
I would like to know if it will be taken into account too.That would be useful if we can turn off USB ports whenever we want, for usb modules for example.
 
Last edited by a moderator:
Back
Top