Rockbox Technical Forums
Rockbox General => Rockbox General Discussion => Topic started by: silentseraph on September 14, 2007, 08:51:44 PM
-
Hello. I was scanning through the manual for rockbox on my video ipod 5.5 and I was wondering if there was a sleep mode that could be toggled in and out of like in the original firmware, since shutting down and starting up takes a lot of battery power. Also, if it is supported, on what builds?
-
Rockbox doesn't have a sleep mode.
-
Hello. I was scanning through the manual for rockbox on my video ipod 5.5 and I was wondering if there was a sleep mode that could be toggled in and out of like in the original firmware, since shutting down and starting up takes a lot of battery power. Also, if it is supported, on what builds?
What makes you think shutting down and starting up uses more battery power than leaving your ipod in sleep mode would ?
-
The apple os has a kind of sleep mode. like if you pause during a song and leave it sitting for a certain amount of time (time settings set by user) then it will not shut off but go into a "sleep" mode. Next time you press a button, the ipod kicks up where you left off, instead of rebooting. like on the song you left it on. I think that is what he means.
-
We know what he means, but why does he think sleep (where some of the hardware is on) takes less power than being off (when the hardware is off). I can see a cross over period, but would imagine it is very short.
-
well, i mean, thats the point of sleep mode. It doesnt eat up a lot of battery, but its still "on". It just kills the battery to boot it up again and again, and while in school, ill have to turn it off in between classes, which kills it a lot faster than it would If I could put it to sleep. Im not saying that leaving it off takes more power, but booting it up repeatedly just kills it really fast.
-
well, i mean, thats the point of sleep mode. It doesnt eat up a lot of battery, but its still "on". It just kills the battery to boot it up again and again, and while in school, ill have to turn it off in between classes, which kills it a lot faster than it would If I could put it to sleep. Im not saying that leaving it off takes more power, but booting it up repeatedly just kills it really fast.
you're definitely right. :)
-
Again - what are you basing this on ?
-
I know what sleep mode does and what turning on and off is. However, you cannot say which takes more power without some measurements and calculations. You need to do:
Power required to turn off, then on again = x
Power required to stay in sleep = y/min
Minutes in sleep = z
Therefore if z is low enough that y*z < x then sleep is worthwhile, otherwise it is better to turn off then on again.
Just saying you are definately right doesn't mean anything, the only way you can say for sure is by doing some measurements.
-
And this experiment is impossible to do, as there's no sleep mode to get variable y.
However, we could estimate it I suppose, and it's surely 0 < y < (power consumed while idle, backlight off, not updating the screen).
(The rest of this is on an iPod Video, all I have available to test on. While the same could be done for all other players, the hard number apply only to the iPod Video, and actually technically only to MY iPod Video)
As far as battery usage for booting, on my iPod 5G Video it appears to be under 1%, possibly around 0.5%, on a battery that's seen over 18 months of nearly daily use, but the iPods tend to report the battery levels strangely right on bootup (At least mine does. It said 15% right away, and after 30 seconds or so jumped to 24%. It was this 24%, which shrunk from 26% over 7 reboots, that I used)
I haven't done a battery test lately, but while using the ipod it's around 6 hours from just my noting how much time I've been listening and how much battery is left (something I do moderately often), so I'll use 6*60/100 = 3.6 minutes per percent while running. As I don't tend to leave my ipod on and not playing music, this number is likely low, but by how much I have no idea. Let's say 5 minutes to lose a percent in the lowest mode you could get in without sleep mode.
So, if I'm going to leave my ipod on for more than 2.5 minutes, I'm better turning it off. That's interesting to me, I frequently pause my ipod for upwards of 30 minutes at work, sometimes hitting that 30 minutes, which is where I have Rockbox set to turn it off.
Can sleep mode do significantly better? 10x better or more? If it can't do at least 20 times better, you can't make it through an hour class without wasting more battery.
-
My point is solely that we don't know one way or another so saying things like:
well, i mean, thats the point of sleep mode. It doesnt eat up a lot of battery, but its still "on". It just kills the battery to boot it up again and again, and while in school, ill have to turn it off in between classes, which kills it a lot faster than it would If I could put it to sleep. Im not saying that leaving it off takes more power, but booting it up repeatedly just kills it really fast.
you're definitely right. :)
is meaningless.
This is all a bit academic, seeing as we don't know the hardware well enough to initialise everything properly so we don't use so much battery when running properly, let alone sleep components.
-
Ive done some research and for the type of hard drive the 5g ipod video uses (toshiba MK3008GAL) there is a large (i dont really know power measurements though) difference in the amount of energy required to start up and to sleep
http://sdd.toshiba.com/main.aspx?Path=HardDrivesOpticalDrives/1.8-inchHardDiskDrives/MK3008GAL/MK3008GALSpecifications
heres the page and if you scroll down to power consumption you see that starting up the hard drive requires 1.8 watts of energy while sleeping requires a considerably smaller amount (only .07 watts of energy)
-
But the 0.7 watts of energy whilst sleeping is a *constant* drain - whilst spinning up is a one off.
-
Ive done some research and for the type of hard drive the 5g ipod video uses (toshiba MK3008GAL) there is a large (i dont really know power measurements though) difference in the amount of energy required to start up and to sleep
http://sdd.toshiba.com/main.aspx?Path=HardDrivesOpticalDrives/1.8-inchHardDiskDrives/MK3008GAL/MK3008GALSpecifications
heres the page and if you scroll down to power consumption you see that starting up the hard drive requires 1.8 watts of energy while sleeping requires a considerably smaller amount (only .07 watts of energy)
Yes, of course. I'm not saying sleep isn't useful in some circumstances, I'm just saying without measuring numbers you cannot make sweeping statements like you originally did. Without measurements, we have no way of knowing the cut off point. I seem to remember back in the day someone did just such a measurement for an Archos device, and concluded about four minutes was the cutoff point between it being worth turning off and not. This was between sitting idle and turning off, and doesn't apply to the different hardware of the iPod anyway, but again:
My point is not that in some circumstances sleep cannot save power, it can. My point is only that you do not know those circumstances without measurements. Is it five minutes, is it ten minutes, is it an hour, is it a day?
-
Also: 0.07 watts for 90 seconds is more energy than 1.8 watts for 3 seconds. So we'd need to still know how long this "start" drain takes place over, and how long the drive sleeps.
Seriously, if shutting down the drive was energy inefficient, why do you think battery life *increases* when we enable HD power off?