Windows Modern Standby Bug Reporting

[AlexTheGreatish]

Hello folks,

 

I had a meeting with Microsoft and they are taking our video about Windows Modern Standby very seriously.  If your Windows laptop experiences this problem (extreme battery drain while asleep, finding your laptop super warm in your bag with the battery drained) then there are a couple of things you can do.

 

For your own personal troubleshooting you can create a Sleep Study.  To look at this open Command Prompt as an admin and type in "powercfg /sleepstudy".  This creates a log of your devices battery life and sleep states, maybe there is something interesting in there.

 

More importantly, you can send feedback to Microsoft so they can try and diagnose the problem.  Here is how that works:

 

The Feedback Hub is the best way to provide detailed feedback on issues to the Windows engineering team. The tool gathers detailed logs and can run additional diagnostics to help us diagnose and fix issues.

 

Launch the Feedback Hub and give your feedback a relevant title.

 

 

 

Click “Report a Problem” then provide more information on the specific issue. Any additional details are valuable.

 

 

 

Click Next then choose Power and Battery >> Sleep (it may automatically select this based on what information you provided in the description). Selecting the right area is important as it directs the issue to the appropriate engineering team and gathers relevant telemetry from the system.

 

 

Click Next, then select New feedback.

 

 

On the Add more details section, mark as High Severity if you’ve hit the battery drain issue, then choose one of the items below, I’d pick Inability to use my PC if your battery is dead.

 

 

Continuing in section 4 – this is the critical part – gathering additional data, without this we will not have enough data to diagnose the issue.

 

For the battery drain issues please select Sleep (it should be selected automatically). Click Start Recording, button turns read, it will gather data, wait 10s or so then press stop (you don’t need to go through the actual sleep process, nor do you need the screenshots) it may take a minute or two to process after stopping the recording. You can also put your computer to sleep during the process then reawaken it – it will collect data across the process.

 

 

Click Submit.

 

 

The most difficult part of diagnosing Windows Modern Standby problems is getting good data, so with any luck the LTT community can send them boat loads of data to try and get it fixed.

[Imbadatnames]

Best fix I’ve found personally is to just get a mac 

[jaslion]

Wouldn't a temporary (and permanent) fix be that windows 11 MANDATES S3 state as available? Till s0 gets fixed (if ever).

[Kid.Lazer]

8 minutes ago, jaslion said:

Wouldn't a temporary (and permanent) fix be that windows 11 MANDATES S0 state as available? Till s3 gets fixed (if ever).

I think you have that backwards? S3 is the old-school sleep state that normally works ok(ish). S0 is the current broken model.

[jaslion]

9 minutes ago, Kid.Lazer said:

I think you have that backwards? S3 is the old-school sleep state that normally works ok(ish). S0 is the current broken model.

Yup! Switched em around

[Senzelian]

53 minutes ago, AlexTheGreatish said:

The most difficult part of diagnosing Windows Modern Standby problems is getting good data

I honestly believe that. I can't imagine a single person on this planet actually sends Microsoft any usable feedback. Most of which is probably just people sending stuff like "F*ck Microc*ck!". On the other hand it should've never been a problem from the beginning.

[Kilrah]

I've had the issue with every laptop that uses modern standby/S0, and everything pointed to "it's by design". For example as was mentioned in the video, the whole thing about unplugging the charger waking the system up is a MS spec, it's a requirement for the machine to signal the change of state - but then it handles it poorly and instead of waking up, going "oh I was disconnected, good to know, let's go back to sleep" it just stays on.

My XPS 7390 2in1 (same Linus has/had) would be absolutely terrible for that, even after a fresh reinstall. 

 

Will try to do this on my current Asus which doesn't even obey the sleep settings, I've set it to sleep after 20 minutes but often I leave, come back 2 hours later and it never went to sleep. When it does go to "sleep" i.e. power indicator blinking it'll often be running fully, with the fan spinning and will drain the battery in half a day. It's only _less_ of a problem with the Asus than with the Dell because the Asus is AMD and has really low idle drain, with the XPS it'd be dead in 3 hours.

 

But I would also recommend Microsoft just talk with their partners because they're obviously aware of it. Dell for example has a hard rule of "never put your laptop in a backpack without fully turning it off" which is obviously absolutely ridiculous and was never a problem with S3.

 

 

I'm not necessarily entirely against S0 being an option, but the ability to choose S3 sleep and disable S0 in the BIOS if desired should be mandatory. It's highly frustrating that a no-name laptop that doesn't implement S0 gives a better experience than every major brand. 

[AlexTheGreatish]

4 hours ago, jaslion said:

Wouldn't a temporary (and permanent) fix be that windows 11 MANDATES S3 state as available? Till s0 gets fixed (if ever).

 

3 hours ago, Kilrah said:

~snip~

In talking to Microsoft about the problem I at least get why they don't want to keep S3 as an option.  With S3 sleep the firmware of each device  is responsible for how it goes to sleep, and of course many devices don't have regular firmware updates or are just broken.  By using S0 it is way more secure and overall has less problems sleeping and waking everything, but as we know S0 has it's own problems.

[Kilrah]

32 minutes ago, AlexTheGreatish said:

With S3 sleep the firmware of each device  is responsible for how it goes to sleep, and of course many devices don't have regular firmware updates or are just broken.

Yeah but even if it could in theory that's never really been a widespread problem, unlike S0 that's almost always broken and not working as expected  

 

I'm doing the recording right now of my laptop being turned on, no app launched, left unattended with the power plan set to sleep after 15 minutes and never going to sleep - just left it for a couple of hours before that to check, it never did and battery went from 60 to 30%...

 

When running Ubuntu on that laptop it can stay for days in standby as expected.

[Imbadatnames]

2 minutes ago, Kilrah said:

Yeah but even if it could in theory that's never really been a problem, unlike S0 that's almost always broken and not working as expected  

 

I'm doing the recording right now of my laptop being turned on, no app launched, left unattended with the power plan set to sleep after 15 minutes and never going to sleep - just left it for a couple of hours before that to check, it never did and battery went from 60 to 30%...

 

 

Isn’t it something to do with it not swapping sleep state if you plug it in, sleep it then unplug it? 
 

TBH I don’t see Microsoft ever fixing it if they even can without manufacturer firmware updates. Basic shit like the taskbar still having a ton of issues where it derps out carried over for on windows 10 hasn’t been fixed for example. 

[Kilrah]

5 minutes ago, Imbadatnames said:

Isn’t it something to do with it not swapping sleep state if you plug it in, sleep it then unplug it? 

That was one of the suppositions in the video, but the requirement for notification by firmware of the change of state on connection/disconnection of the adapter would precisely be so that it can take note of the change... could be a bug of course but it'd have been there for years.

Can't find the page in the MS docs that related to this a couple of years ago unfortunately.

[Imbadatnames]

17 minutes ago, Kilrah said:

That was one of the suppositions in the video, but the requirement for notification by firmware of the change of state on connection/disconnection of the adapter would precisely be so that it can take note of the change... could be a bug of course but it'd have been there for years.

Can't find the page in the MS docs that related to this a couple of years ago unfortunately.

Could be a security thing with not being able to reassign sleep state while sleeping 

[birdhuman]

I have learnt to unplug before close the lid anyway...

[SunnyZ]

I have a Toshiba Satellite S70t-A.

The battery drains over night when in sleep, also when powered off completely.
I thought this was a battery issues, so bought a new one (this laptop has a battery you can just pop out).

Same battery drain issue on the new battery...

 

My current solution is to just take the battery out when not in use...

[kramertorium]

I'd like to see where my Framework 12th-gen stands on this issue - I've experienced some of the same problems as many other users.

 

However, it seems this problem's reporting methodology among users is all over the map, understandably. Perhaps it would be good to make a suite of scenarios and battery run-down tools we can all try so we can get consistent information?

 

I'd be happy to spend some time going through specific steps, even if it takes a long time, but only if there's a way to aggregate such data into something useful for troubleshooting.

 

I'm particularly interested to see the results from other Framework 12th-gen users; there's plenty of data here about the problem (and on community.frame.work) but it's not structured enough for comparisons.

 

[I'm not very active here in the forums, apologies if this isn't the right place to put this comment]

[voldemarz]

@AlexTheGreatish Posting here in hope you can forward some of this to Microsoft or at least for some visibility and upvotes in Feedback Hub. Neither of my reports in Feedback Hub over the years have resulted in anything being fixed. MS in the new laptop is driving me mad while neither of preious laptops with s3 had any issues.

 

1. Another issue with Modern Standy is that some "sleep" related OS settings just don't work. For instance untoggling "Allow this device to wake computer" in Device Manager does nothing, and hence can't disable mouse input from waking the machine. I have disabled this for all devices, USB controllers etc, it still wakes. Here is Feedback Hub report link - https://aka.ms/AAip0id

 

2. When OS is using s0 sleep, there is now way to initiate it via command line. All methods discussed online or even suggested directly by Microsft result in hibernation or do nothing if hibernation is disabled. Feedbac Hub report - https://aka.ms/AAip0hu

 

The reason for needing a way to trigger it via command line is because it would help troubleshooting with Windows Performanca Analyzer by placing WPA tracer start and sleep initiation in a script. Because it doesn't work, I had to resort to initiate WPA trace using Scheduled Task triggered on Kernel-Power event with ID 506, but that triggers also for hibernation, not just sleep. Then upon login after wake I have a task to save the trace.

 

@Kilrah I have also seen the issue with laptop waking after pluging in power. Annoyingly it mis-reported the wake reason as "Input Touchpad". https://aka.ms/AAj4fi4

 

MS wants people to submit reports via Feeback Hub, but at the same time haven't even made sensible related report selection. If you try creating report titled "Battery drained during sleep" and description "Battery drains 5% per hour." in the "Find Similar Feedback" step all suggested feebacks are very obviously totaly unrelated.  Not a single post relateing to "sleep", or "battery".

 

Also, there might be another issue to troubleshoot. Today laptop had died completely while in sleep, it didn't even hibernate at 5% like it should.

 

 

[Dustball72]

I had a strange version of the issue... I closed the laptop, left it on my coffee table (unplugged throughout); then about 10 minutes later the thing tried momentarily to takeoff (fans on full). I hadn't noticed it before, but I'm not routinely next to it after shutting it. 

[Smir]

I submitted a report the other day Dell XPS 13 2 in 1 9310 closed lid around 60-70% opened up the next day to 2%.

[Remixthis]

Just wanted to drop in to say I noticed my hibernate settings were set to an odd number. All my sleep settings were normal but when I went into my power options under "hibernate after:" the plugged in setting was set to 120 min while the, on battery was set to something like 342759463 minutes. Not sure why because I don't set times like that. I don't know exactly what the number was and I'm not saying this will fix all these issues but it might help someone or highlight a different issue. I've still reported the issue a couple times before this setting was changed.

[wseaton]

I have never been a fan of battery dribble modes. Fast Start caused enough problems.

 

I've always used hibernation mode and always will.  With a decent SSD the extra time starting more than cancels out the myriad of problems battery dribble modes cause. 

 

Let's keep it real. PC based portables are pieces of  sh_t. They always have been pieces of sh_t and likely always will be. Ive never understood the neurotic obsession with PC based laptops. 

 

Get a Mac, use a desktop or use hibernation mode and stop whining like my ex. 

[Scaredyfish]

 I finally thought I had this cracked - at least enough for day to day usage. I ran the sleep study report a few weeks ago, and found that it was waking, doing a bunch of dns stuff  relating to my work domain which was failing, causing it to keep waking every minute. I got into the habit of going into airplane mode before putting the laptop to sleep, and that worked pretty well for the last few weeks.

 

Then, just today, I opened up a dead laptop. I ran the sleep study report, and the reason for waking and battery drain was "TimerForAudioPlayback". Why would that even be a thing? You had to urgently wake up to play Spotify?

 

And my sysadmin has feedback hub disabled, so sorry MS, I can't send you my report.

[Isuck Assimov]

I also heard from dozens of people who claim their devices lose most of their charge while the device is completely powered down and not even in sleep mode.

To be fair i still assume those people accidently put the devices in sleep or have usb devices connected to the always on ports, which would make a lot more sense that trickle discharge while 100% shut down.

[BlueChinchillaEatingDorito]

Mhmm we're actually dealing with a whole slew of issues with Lenovo ThinkPads where they can be unresponsive after going to sleep. Looking at Event Viewer and most of the time the last entries before everything just stops responding is the device entering connected standby. After that the machines are visually "sleeping" (i.e. power LED pulsates) but you cannot get the thing to wake up no matter what you do. No response from power button, keyboard, mouse,  etc. Only solution is to do an emergency reset on the device. Sometimes they even have a tendency to cook their own guts out as the CPU appears to be going full power but the fans don't kick in because it's still in "sleep". 

[Kilrah]

Gahhh this is so frustrating.

Laptop set to sleep after 20 minutes. 

Leave it open on a table doing nothing, walk away, come back 3 hours later, not in sleep. 

Start a 70GB transfer to a network share, walk away, come back and it's gone to sleep mid-way of the transfer 

[Mark Kaine]

On 2/9/2023 at 2:11 AM, BlueChinchillaEatingDorito said:

Mhmm we're actually dealing with a whole slew of issues with Lenovo ThinkPads where they can be unresponsive after going to sleep.

sounds like a hardware issue almost?  i have a Lenovo ideapad i always use sleep / hibernation (both) and there doesn't seem to be any issues waking up, *except* sometimes it'll run out of battery way sooner than it should have (almost like it would wake up by itself doing stuff and drain the battery,  but i never noticed it doing that either, and it happens really very rarely,  still mysterious...)