5 ms·

Didn't read it all, but regarding the battery and sleep issue: I can totally relate to such issues with my Dell XPS 15 9550 (2015) and 9510 (2021). Windows has

• Robinxd 5y ago
  Ironically, I have had the same experience with MacBook Pro 2017 model. It’s got some other funky bugs too, but it’s not my main driver either.
• mcv 5y ago
  I still don't understand how Windows could possibly have become so bad at sleeping. Both desktops and laptops have a tendency to wake up at weird times and then never go to sleep. This stuff used to work. What happened?
  • hughrr 5y ago
    Microsoft replaced QA with telemetry.
  • pizza234 5y ago
    This is possibly the so-called "Connected Standby" (https://en.wikipedia.org/wiki/InstantGo https://en.wikipedia.org/wiki/InstantGo), an infamous Windows anti-feature. I believe, although I'm not 100% sure, that in addition to turning on the computer without the user consent, it also puts the machine in a strange state where it's neither on nor off. According to the spec: https://en.wikipedia.org/wiki/InstantGo https://en.wikipedia.org/wiki/InstantGo, it's allowed to suck up to 7.5% of battery every day. As a matter of fact, my Surface device would lose more than 5% every day, and I'd find odd periods of wakeup in the sleep log (when the machine was supposedly fully shutdown).
    • buran77 5y ago
      My old iMac was constantly waking from sleep for online tasks because of the (on by default) Power Nap feature introduced around the same time as InstantGo. By the time I paid any attention to it my HDD had ~500.000 head park cycles which also lead to its early demise. Ended up using it with an external SSD which was a massive boost in performance anyway.
  • q3k 5y ago
    The introduction of S0ix, a.k.a. pretend to sleep. Why? So that apps can communicate with the Internet while the machine is in sleep mode, or do periodic wakeups for other reasons. Another attempt to shoehorn the smartphone paradigm into desktops/laptops.
    • walterbell 5y ago
      The beatings will continue until telemetry and location data improve. Does S0ix-related heat increase or decrease when laptop is in a faraday bag?
      • p_l 5y ago
        "Modern Standby" requires that power draw (and thus thermals) is comparable or lower to S3. It's less about telemetry etc. (you could do it with age old ACPI timers), it's about people complaining laptop doesn't boot immediately from sleep when they open the lid - but also that on a system where power management support is heavily tested (read: not Linux) "Modern Standby" has less moving parts and is harder to break.
    • p_l 5y ago
      Modern Standby has less moving parts that S3 sleep (and IIRC also avoids certain security related issues)
  • sundvor 5y ago
    Windows desktop refuses to even let the monitors sleep if it detects eg game controllers such as Heusinveld Sprints (pedals). It's truly notoriously bad. I suppose it's designed that way to ensure a game won't black screen, but how hard would it be to make it a bit smarter? Ie when VSCODE or Terminal has focus, allow sleep.
  • hollowdene 5y ago
    I used to review laptops for a living and it's baffled me for a decade how Windows is still so bad at this. Last Windows laptop I had was an XPS 13 and I resorted to turning it off to avoid the "heat the bag" problem. M1 MacBook Air is basically the only laptop 95% of people should buy (assuming they can afford it)
    • m_st 5y ago
      I second the MacBook Air M1 and recommend it to about everyone that can run a Mac. What a great piece of hardware. Honestly it just works, is super fast and doesn't even have a fan.
  • npteljes 5y ago
    Microsoft is shifting to remotely managing your system. The technology itself works well, it's just what the software tells it to do.
  • jseliger 5y ago
    Long discussion of these issues here: https://news.ycombinator.com/item?id=28639952 https://news.ycombinator.com/item?id=28639952, but the short version appears to be that Intel removed "S3 Sleep" from recent chips, at Microsoft's behest, in favor of a mode that is supposed to be able to wake up while nominally sleeping in order to get emails, texts, and that sort of thing. Except this mode doesn't seem to work properly and instead seems implemented in such a way as to frustrate people into choosing AMD or Apple.
• ps 5y ago
  >> Being a happy MacBook user since they were called PowerBook, I never ever had any such issue with an Apple laptop. My early 2020 16inch required lots of tweaking to prevent complete discharge every night and battery is sub 7500mAh now. It still happens several times a year and I am not sure why. I am probably not paying over 4000 EUR next time. User since the 2006 MB Pro.
  • mholm 5y ago
    FWIW, I haven't seen this issue at all on M1 Macbooks. Used to happen once on a month on my Intels.
• altbdoor 5y ago
  +1, received a Dell XPS 15 9500 for work, and it is horrible. Countless complains in the Dell forums, and still so many issues after all the updates. The whole situation really corrodes my trust in Dell.
  • Octabrain 5y ago
    I bought an XPS 17 9710 for my GF on Xmass. Horrible experience and definitely the last Dell laptop I get. From now on I will stick to MacBook pro's and Lenovo Thinkpads.
  • hackerfromthefu 5y ago
    Dell quality control is russian roulette but with half the bullets loaded. They consider an actually working laptop to be a premium feature reserved for the latitude business line.
• canbus 5y ago
  Don't get me started with the 9550...
  • m_st 5y ago
    Oh the 9510 also has its surprises. First the Microsoft bluetooth mouse constantly went sleeping for about 500ms when not moving it. This annoyed the hell out of me. After lots of research I found some guys recommending to disable all sleep features on all the drivers in the device manager. This sure fixed it. But now, I have the exact same issue with the internal track pad, when running on battery power! Meanwhile the MacBook Air M1 just works. For each and every feature. It's sad, but as a Windows Dev, the XPS15 still is a great machine. You just have to know and adapt to its bad parts.
• Cthulhu_ 5y ago
  > Windows has this setting "When I close the lid" which is set to "Sleep". In fact it should be renamed as "Drain the battery and heat the bag". That's still a thing? I thought they would have sorted that out a decade ago; my laptop from my first job (regular dell thing, I don't remember) was cooking in the bag and that's been over ten years ago.
  • EVa5I7bHFq9mnYK 5y ago
    As an elder, I can attest that 2.5 decades ago it all worked like a charm. Then millennials arrived.
• fauigerzigerk 5y ago
  >I never ever had any such issue with an Apple laptop. There is just such an issue right now with macOS 12.2: https://www.macworld.com/article/610175/macos-monterey-12-2-battery-drain-bluetooth.html https://www.macworld.com/article/610175/macos-monterey-12-2-... I currently have to shut down my Mac mini every night instead of putting it to sleep because it routinely crashes and reboots in the middle of the night, lighting up the room
• jcrawfordor 5y ago
  With the current ACPI S0 implementation ("modern standby" in Microsoft parlance) there are a huge set of devices and events which can cause the system to wake. This is in theory a good thing, because it allows the system to perform a lot of background tasks while sleeping and allows "instant" resume in response to user input. It can cause a lot of frustration with "mystery wakes" though. I haven't found this to be especially common on my ThinkPads running Windows although it does happen from time to time, which I eventually tracked down to the Bluetooth controller causing a wake whenever my wireless earbuds connected. Under Linux it's been much more of a problem, presumably due to the Linux drivers having a poorer/less validated understanding of what various device events mean. A frustration is that Linux does not offer as good of tools to diagnose this as Windows does (powercfg). That said you can do a lot of tweaking of the Linux behavior using the acpi sysfs, and on my X1 running Fedora I ended up blacklisting just about every device from causing wake because the USB controller seemed to be generating almost constant mystery wake events. The short story is this: many devices are now active in a low-power state during sleep and can generate a huge variety of events that the OS is expected to receive and make a decision on. Sometimes the OS doesn't understand the event or there's a requirements issue (implementers did not understand/foresee the use case) and these can both result in the system waking for seemingly no good reason. For example, my issue with Linux seems to be the USB controller generating an event the driver did not understand or handle correctly. My issue with Windows seems to be more that a well-intentioned decision (newly connecting Bluetooth devices should wake up the system, which tends to make sense in the case of a mouse/keyboard) had an irritating implication in real-world use (BT headset that connects to multiple devices simultaneously would connect to and wake up my laptop when I was using it with my phone). Unfortunately the way events are implemented on Linux and I assume on Windows as well, there isn't really any support for "if the lid is still closed the system should stay asleep" and so once the system wakes up for an event, and there's no handler that causes it to go back to suspend, it tends to just sit there awake even though the lid is closed. There's no "new event" for the OS to handle since the lid state remains the same. I'm not sure if it's practical for the system to check the lid state in real-time instead of waiting for ACPI events about it. For what it's worth I'd say my touchbar MacBook Pro does this more often than my Windows ThinkPads do, to the extent that pulling it out of my bag and discovering that the battery has gone completely flat overnight is probably a weekly experience. I haven't put much effort into diagnosing this so I don't know if it's a result of some odd software I'm using or something.
• fleddr 5y ago
  Hear hear, nothing about Window's shutdown makes any sense. Other than "sleep" doing the exact opposite, I have a few more: Sometimes an update is pending and the shutdown menu gives a choice "update and shutdown" or just "shutdown". So I pick "shutdown". It then updates anyway, including full restarts. Sometimes there's no special options in the shutdown and upon shutting down, an update is forced. Even ones I did not postpone for too long. Further, the "restart" option is really too close to "shutdown", especially on a very large screen. And there's no canceling once it starts.