Hacker News Reader

Comment by kelnos | Hacker News Reader

The issue is that Windows doesn't go into S3 sleep when you close the lid; it goes into a weird not-really-asleep state for... some reason that I don't agree with[0]. I guess that state is "hot" enough that you could damage the laptop by putting it in a bag while in that state.

Fortunately I've been running Linux on my XPS 13, which I have put into a bag hundreds of times (maybe even over a thousand) while suspended. Linux's suspend does actually use the "cold" S3 suspend state, so this is safe.

[0] It's called "Modern Standby" (https://docs.microsoft.com/en-us/windows-hardware/design/dev...), which... honestly sounds really stupid to me. "Modern Standby" presumably drains the battery orders of magnitude faster than a normal S3 standby. I can't imagine ever wanting this behavior. I'm entirely fine with my laptop taking ~10 seconds to wake up and reconnect to whatever network I'm near.

View on HN

Oh wow. i did NOT know that. Even worse...

>Switching between S3 and Modern Standby cannot be done by changing a setting in the BIOS. Switching the power model is not supported in Windows without a complete OS re-install.

and on Reddit

>Warning: if your laptop is newer than 2019, there is a high chance, your OEM removed any S3 code from the bios, and your laptop will crash entering S3 and you have to force hold power key to restart and then delete the registry entry again to revert back to modern standby.

https://www.reddit.com/r/Dell/comments/h0r56s/getting_back_s...

so not only is microsoft screwing around.. its getting all the manufacturers to screw around as well?

Spoiler: I have an XPS 9500 under Linux, which sleeps fine under S3 but then it never wakes up. So the problem isn't with Windows.

What does "never wakes up" mean? I thought I had the same problem but then I discovered that closing and opening the lid is the only way to get it to wake. Neither the power button (it turns it off) nor keypresses (they do nothing) work.

I think I tried everything, from buttons to lid to magic network packets. Once in S3 it's dead Jim, and there's nothing in the BIOS to change that behavior.

The irony is that I have a 4 years old XPS 13 where everything works like a charm, even under Linux. Talk about progress ...

I’ve had a problem like this with my XPS and I was able to solve it by holding the power button for 25-40 seconds to reset something:

https://www.dell.com/support/kbdoc/en-us/000125880/how-to-re...

A comment above mentions manufacturers removing S3 support from their bios with the exact symptoms you mentioned

So you’re actually "putting it to sleep", when you put it to sleep.

I have an XPS 9500 and it sleeps+wakes, as expected, with S3.

It's terrible. I remember being woken up one night because my laptop decided to wake up and resume the Netflix video I'd be watching earlier. No input devices connected, lid closed. Another time it cooked in my bag because windows wanted to do an update on shut down so I slept it instead.

I have a 2017 XPS model so I the firmware supports S3 sleep. There is no modal in the BIOS though. I had to trick windows into thinking it didn't support modern sleep.

Based on posts from others trying to do the same thing, it used to be a single registery key edit. Now it's convoluted and required fiddling with WiFi power controls. The end result it sleeps properly but takes anywhere from 10 seconds (good) to more than a minute (bad) to connect to WiFi when woken up.

Of course on Linux it just works.

Back when I was looking into this I found someone who'd found a way to patch the acpi table to enable it for the newer models (at risk of bricking the device entirely). Alas I can't find the link any more.

I won't buy another Dell until this is fixed.

My work laptop is an X1 carbon and it supports only S3 sleep so it just works.

Seems to be mostly a Dell/Microsoft problem, haven't heard of other vendors removing S3 functionality (yet, anyway).

ASUS UM325

Weird, I got a newer Asus (Flow X13) and it does S3 just fine.

I've got a Lenovo Yoga Slim 7 which doesn't advertise S3. It's a really nice machine, but having to wait several months after getting it for someone out there who understands these things to come up with the right combination of kernel code and ACPI patches so I could shut the lid and not have it just reset when I opened it back up again was something I could have done without. Fortunately it's got the capability, it just needed turning on.

hi i have the Yoga Slim 7 too (the white carbon fiber one). How do you enable this ?

The DSDT mod at https://github.com/jrandiny/yoga-slim7-ubuntu worked for me to stop it resetting, but only on the 5.14 kernel. YMMV.

> Switching the power model is not supported in Windows without a complete OS re-install

That's absolutely insane.

Intel just removed S3 sleep completely from the latest Tigerlake chips.

I don't think it's the OEMs / manufacturers, it's Intel.

I have dell XPS, it had perfectly functional sleep and could be put in a bag. Then 'modern standby' was added in an update and it ruine the laptop - if I place it in the bag in the morning, by linchtime it will be crazy hot and the battery will be gone. And you cannot diaable this feature. I don't get wtf these people are smoking.

On my Dell, this problem happens when you suspend the laptop from software (either with pm-suspend or systemctl suspend). If you simply close the lid, it seems to enter in a different "mode" of suspend, the real one, where you can carry it without overheating and the battery lasts for weeks.

I purchased one of those 1st gen Windows on ARM laptops (HP Envy) specifically for the "instant on" / modern standby feature. It worked perfectly, just like a smartphone, until I received a major Windows update. Now it wakes slowly ruining the only good thing about that device. I've now been burned twice in a row by MS and will likely not purchase another product from them again.

The very latest (9310) XPS 13 doesn't have S3 at all, though, from what I'm gathering. So even on Linux, it won't sleep well.

Changing the disk to AHCI in BIOS allows my 9310 to reach C10, fixing the sleeping heater issue.

https://bugzilla.kernel.org/show_bug.cgi?id=211879#c24

I’m running linux kernel 5.14.3 for the QCA6390 support as my laptop is the Windows edition.

I have the same machine for work: early this year sleep mode was not very useful (it wouldn't pass a weekend), but after an Ubuntu upgrade a few months ago it got much better. Yet, it's still worse than my 6 year old ThinkPad.

> "some reason that I don't agree with[0]"

On my side I am quite happy that the hardware+software can support some kind of fast wake-up (similar to Macbooks and smartphones instead of needing >10 seconds to wake up when I open the lid). I just do a real shutdown when I don't need the laptop anymore.

And precisely : as it was said by someone else above, on modern computers with SSDs, I would rather disable "fast startup" so that the computer really shutdowns when I ask him to (which is also useful when you have dual-boot and your Linux systems wants to get r/w access to the NTFS partition)

But I admit that the computer could/should automatically enter in suspend-to-disk after being in "modern standby" for more than e.g. 5 minutes, and this should be configurable.

The value of fast wakeup is real. But manufacturers and OS vendors need to actually get their act together and implement this, not fake it in a way that passes reviews but is actually damaging to hardware in the real world.

Exactly. It's entirely possible to implement fast wakeups, but it requires thoughtful design. Smartphone go to sleep opportunistically whenever you shutdown the screen. Even an Intel CPU could probably do this in less than 500ms (and that is a boatload of cycles).

The "fast startup" cause a lot of problems for us. If you "shutdown" you're not doing a full shutdown.... You’re really only doing a full shutdown when you select “Restart” instead as evidenced by the uptime clock in task manager.

It's like Windows broke the simplest button ... just shut it down!

In case anyone is curious I wrote up how I use suspend (amongst other things) with my XPS 13 running Debian: http://h2.jaguarpaw.co.uk/posts/how-i-use-debian-dell-xps-13...

Nice write-up, but I think the point is on the newer XPS' the S3 option is completely gone.

On the other hand ssh sessions staying opened are nice. I agree it doesn't outweigh the mess it created though

> On the other hand ssh sessions staying opened are nice.

Probably a convenience feature for the NSA folks logged in trough the Intel ME.

Just wait. The newer laptops don't have S3 in the bios at all.

True but you can enable it on some laptops by telling the firmware setup that your OS is Linux. I'm not sure what Linux S2idle is in ACPI lingo, but that's what Linux uses on my newest lapop by default. S3 (suspend-to-RAM) is not available to the kernel. And yet it's a power hog compared to just not using the laptop and letting the display backlight turn off, roughly 1% battery life loss per hour in S2idle. Whereas it's about 0.5% battery life loss per hour in whatever Windows 10 is doing for laptop lid close.

Reply on news.ycombinator.com