It’s the stanby/quick restart that’s doing it. It’s literally staying on. Best to use the power/shut down option in the menu vs just hitting the power button. Same was happening to me before I figured it out.
Thanks for that. I’ve been thinking of it as a sleep/hibernate function but obviously it’s not. Linux still hasn’t seemed to figure that one out for some reason. Not Steam’s fault.
Linux has a hibernate function, but the power button activates Sleep, which is higher-consumption. There may be a way to set it to actually hibernate on button press, but I’ve only seen Hibernation via Desktop Mode.
Sleep = keep processes in memory = more power consumption.
Hibernation = write most processes to disk and keep only vital system processes in memory = less power consumption.
At some point SteamOS has major issues crashing when waking up from hibernation, which is probably why it hasn’t been added as an option.
Which is annoying, because if you run out of battery, the deck just dies. At the very least, it should force-hibernate itself before dying.
This kinda describes where Linux has been at with sleep/hibernation for quite a few years. I don’t understand the deeper implications but it’s never seemed like a priority for Linux devs, vs how Windows and Mac have solved it long ago. Maybe because Linux hasn’t traditionally focused on portable devices but arm (etc) seems to be changing that.
Windows hibernation is about as broken as linux hibernation, i.e. they both mostly work most of the time, but there is good reason both hide them away by default (if you can really say linux hides anything, with these things being decided by distros and not kernel devs). It is naive to say windows has “solved” hibernation. Either you don’t use it much or have very basic hardware and software needs.
Edit: as a side note, neither iOS nor android devices use anything similar to hibernate, so I am a bit lost with what you mean by arm causing hibernation implementation pressure.
idk, for me hibernation has mostly worked fine so long as i don’t hibernate with a game running, which seems to be more of a GPU issue than hibernate itself.
This is good intel actually. I used HoloISO for a long time on my gaming rig but I never thought to mess around with those settings because I’ve always just thought of Linux battery use as ass (have run various distros on tons of different laptops as well). Would be good to take the deck deeper hibernation settings for a spin, but it would be kinda a shame if the deck devs haven’t already explored these things in ways I’ll never understand as a lay user, frankly. You’d think they’d be tweaking this stuff mercilessly for the UX and battery life.
It’s the stanby/quick restart that’s doing it. It’s literally staying on. Best to use the power/shut down option in the menu vs just hitting the power button. Same was happening to me before I figured it out.
Thanks for that. I’ve been thinking of it as a sleep/hibernate function but obviously it’s not. Linux still hasn’t seemed to figure that one out for some reason. Not Steam’s fault.
Linux has a hibernate function, but the power button activates Sleep, which is higher-consumption. There may be a way to set it to actually hibernate on button press, but I’ve only seen Hibernation via Desktop Mode.
Sleep = keep processes in memory = more power consumption.
Hibernation = write most processes to disk and keep only vital system processes in memory = less power consumption.
At some point SteamOS has major issues crashing when waking up from hibernation, which is probably why it hasn’t been added as an option. Which is annoying, because if you run out of battery, the deck just dies. At the very least, it should force-hibernate itself before dying.
This kinda describes where Linux has been at with sleep/hibernation for quite a few years. I don’t understand the deeper implications but it’s never seemed like a priority for Linux devs, vs how Windows and Mac have solved it long ago. Maybe because Linux hasn’t traditionally focused on portable devices but arm (etc) seems to be changing that.
Windows hibernation is about as broken as linux hibernation, i.e. they both mostly work most of the time, but there is good reason both hide them away by default (if you can really say linux hides anything, with these things being decided by distros and not kernel devs). It is naive to say windows has “solved” hibernation. Either you don’t use it much or have very basic hardware and software needs.
Edit: as a side note, neither iOS nor android devices use anything similar to hibernate, so I am a bit lost with what you mean by arm causing hibernation implementation pressure.
idk, for me hibernation has mostly worked fine so long as i don’t hibernate with a game running, which seems to be more of a GPU issue than hibernate itself.
That’s because system firmware is designed and tested on Windows, so the supply of new and exciting hardware bugs that need workarounds is endless.
This is good intel actually. I used HoloISO for a long time on my gaming rig but I never thought to mess around with those settings because I’ve always just thought of Linux battery use as ass (have run various distros on tons of different laptops as well). Would be good to take the deck deeper hibernation settings for a spin, but it would be kinda a shame if the deck devs haven’t already explored these things in ways I’ll never understand as a lay user, frankly. You’d think they’d be tweaking this stuff mercilessly for the UX and battery life.
Try putting a laptop running Windows to sleep for a week and see if it has any battery left.