Windows 10 Waking from S3 Sleep oddities

[evievi]

Hi all,

 

I have a new Ryzen 7000 system I built and I'm having an interesting experience waking the computer from sleep. I double checked powercfg and confirmed that my motherboard only has S3 sleep and Hibernate enabled. I find that when my device goes to sleep, I cannot wake it using my GMMK pro keyboard, but my OG Logitech MX 518 (still the best mouse to ever exist) is able to wake it up. I know that the MX 518 released towards the end of the PS/2 to USB era, and Windows says "USB/PS/2 Mouse" when I plug in the device, but Devices and Printers pretty clearly says that it's working over USB.

 

 

Is there any reason why this might be? My old motherboard had PS/2 and I always thought that sleep disabled USB devices and the reason why I could wake is because I kept my MX 518 connected over PS/2 and it could send interrupts. Now I'm getting the impression this problem might be with my GMMK Pro? Is there a good test I can run to see if my keyboard gets disabled in sleep? Otherwise I have no idea, ACPI is weird and it doesn't really bother me, because I have no plans on ditching my MX 518 anytime soon.

 

[manikyath]

the logitech MX518 (along with all logitech mice i'm aware of) performs all its mouse functions trough a bog standard mouse device, and all it's special features are trough a separate 'device' as seen by windows.

 

perhaps your keyboard doesnt present as a device that sleep mode understands once windows has loaded, perhaps it's just some bios setting, or perhaps S3 sleep is just an inherently broken piece of utter garbage.

[SquintyG33Rs]

yeah sleep has never worked as intended for me... ever.

 

the computer even sometimes has specific ports that it keeps awake and you have to know which ones that is (the internal keyboard connector of laptops is still also usb protocol very often)

in some rare cases it's just bios settings for wake from sleep over usb only from mouse or only from keyboard (usually all USB is on a single setting though...)

 

it could also simply be a question of the GMMK keyboard using a fancy driver to do extra things with it's controller that doesn't play well with windows sleep state. while the OG mouse has never changed for 20 years.

 

sometimes it's also just windows being extremely dumb in how it kills background services that don't see activity for a long period (even without sleep)

[evievi]

20 hours ago, manikyath said:

the logitech MX518 (along with all logitech mice i'm aware of) performs all its mouse functions trough a bog standard mouse device, and all it's special features are trough a separate 'device' as seen by windows.

 

perhaps your keyboard doesnt present as a device that sleep mode understands once windows has loaded, perhaps it's just some bios setting, or perhaps S3 sleep is just an inherently broken piece of utter garbage.

Alright yeah that's what I assumed, thanks for clarifying. I did some googling around after this post and PS/2 over USB isn't a thing that exists so it's definitely a standard USB mouse in the device tree. Have to assume it's down to a special case of which USB controllers stay active in sleep and they're both connected to different controllers. S3 sleep is definitely broken though, so it always could just be S3 sleep being buggy.

 

[evievi]

19 hours ago, SquintyG33Rs said:

the computer even sometimes has specific ports that it keeps awake and you have to know which ones that is (the internal keyboard connector of laptops is still also usb protocol very often)

in some rare cases it's just bios settings for wake from sleep over usb only from mouse or only from keyboard (usually all USB is on a single setting though...)

 

Kind of a nitpick, but is this true? It was my understanding that a lot of laptop makers still generally use PS/2, I2C, or a proprietary interconnect for laptop keyboards and mice. As someone who used to Hackintosh, I know that VoodoPS2 exists as a kext because so many laptops use PS/2 for trackpads. I'm sure, though, that plenty of laptops still use USB. I actually just checked out my old Intel Mac and confirmed the keyboard and mouse do in fact are under the USB device tree, just connected to the T2 chip. I think Apple never supported PS/2 though, so that could be an Apple "thinking different" moment.

 

19 hours ago, SquintyG33Rs said:

the computer even sometimes has specific ports that it keeps awake and you have to know which ones that is (the internal keyboard connector of laptops is still also usb protocol very often)

in some rare cases it's just bios settings for wake from sleep over usb only from mouse or only from keyboard (usually all USB is on a single setting though...)

 

it could also simply be a question of the GMMK keyboard using a fancy driver to do extra things with it's controller that doesn't play well with windows sleep state. while the OG mouse has never changed for 20 years.

 

sometimes it's also just windows being extremely dumb in how it kills background services that don't see activity for a long period (even without sleep)

I think these 3 options are probably the culprit in some way. I just confirmed that my keyboard and mice are plugged into the same controllers, so most likely the issue is with Windows or the GMMK driver because all of the ports under the parent controller would stay equally awake during sleep. Guess this is score another point to my MX 518; makes me want to go and buy a backup one.

[SquintyG33Rs]

7 hours ago, evievi said:

 

Kind of a nitpick, but is this true? It was my understanding that a lot of laptop makers still generally use PS/2, I2C, or a proprietary interconnect for laptop keyboards and mice. As someone who used to Hackintosh, I know that VoodoPS2 exists as a kext because so many laptops use PS/2 for trackpads. I'm sure, though, that plenty of laptops still use USB. I actually just checked out my old Intel Mac and confirmed the keyboard and mouse do in fact are under the USB device tree, just connected to the T2 chip. I think Apple never supported PS/2 though, so that could be an Apple "thinking different" moment.

 

I think these 3 options are probably the culprit in some way. I just confirmed that my keyboard and mice are plugged into the same controllers, so most likely the issue is with Windows or the GMMK driver because all of the ports under the parent controller would stay equally awake during sleep. Guess this is score another point to my MX 518; makes me want to go and buy a backup one.

the controller can choose which ports to sleep individually. and not wake up properly depending on the device.

 

my 2 front panel USB3, one has a bluetooth dongle in it and so stays awake but the other goes to sleep and no matter what i plug in it won't notice. Except if i move the bluetooth dongle into it...