3 ms·

It’s a bit of column A, a bit of column B sort of situation. Microsoft should be more careful, but also such blatantly garbage hardware never should’ve shipped

It’s a bit of column A, a bit of column B sort of situation. Microsoft should be more careful, but also such blatantly garbage hardware never should’ve shipped. So I’d say hold MS responsible, but also push back against manufactured e-waste like that.

gruez 1y ago
>Microsoft should be more careful What does "more careful" even mean? For instance, if your crappy SSD has some bug that causes data corruption if there are more than 16 writes in flight at a time, and microsoft pushes an update that causes it to do writes more aggressively, should it really be microsoft's fault that your crappy SSD broke? Should microsoft keep throttle windows IO at 2000s HDD throguhput on the off chance that some IDE drive bricks itself when it encounters SSD level workloads?
- cosmic_cheese 1y ago
  It’s not really MS’ fault that crappy SSDs broke, but at the same time it’s only ever Windows that sees these issues. This suggests that this kind of strain on the hardware is unnecessary given the task, which then paints a picture of negligent sloppiness in engineering and QA culture at MS.
  - gruez 1y ago
    >It’s not really MS’ fault that crappy SSDs broke, but at the same time it’s only ever Windows that sees these issues see sibling comment for something similar happening on zfs. linux is also filled with workarounds for crappy SSDs, so "SSDs breaking due to software" isn't exactly exclusively a windows problem. It just gets more attention because of the bigger user base. >This suggests that this kind of strain on the hardware is unnecessary given the task, which then paints a picture of negligent sloppiness in engineering and QA culture at MS. "straining hardware" shouldn't be a thing for PC components. If it can't handle a given workload, it should throttle/queue the requests, not brick itself. I can grant some leeway for long term wear (eg. heat damage or electromigration), but that's clearly not what's happening here. Can you imagine a network card that bricks itself if you send too many packets, and you have to baby it with how many packets you send in any given amount of time?
    - Ekaros 1y ago
      GPUs and CPUs control their clock frequency perfectly well depending on load and temperature. I see no reason why every single other component could not do the same.
  - recursive 1y ago
    Windows has a lot of deployments so it makes sense to see it there more often. I don't think Windows has the most hardware compatibility issues.
- ragequittah 1y ago
  There was an NVIDIA update recently [1] that caused problems with temperatures being properly reported. Entirely software, killed some video cards. Though entirely nvidia's fault I wouldn't say it's because they make garbage hardware in any way. It seems increasingly these problems (including the NVIDIA one) are only cropping up on 24H2 machines which is why I've decided to stick with 23H2 until it's out of its support window. Unfortunately that's coming up pretty soon. [1]https://www.neowin.net/news/nvidia-fixes-windows-11-24h2-driver-issues-but-completely-breaks-gpu-temperature-reading/ https://www.neowin.net/news/nvidia-fixes-windows-11-24h2-dri...

gruez 1y ago

>Microsoft should be more careful What does "more careful" even mean? For instance, if your crappy SSD has some bug that causes data corruption if there are more than 16 writes in flight at a time, and microsoft pushes an update that causes it to do writes more aggressively, should it really be microsoft's fault that your crappy SSD broke? Should microsoft keep throttle windows IO at 2000s HDD throguhput on the off chance that some IDE drive bricks itself when it encounters SSD level workloads?

cosmic_cheese 1y ago
It’s not really MS’ fault that crappy SSDs broke, but at the same time it’s only ever Windows that sees these issues. This suggests that this kind of strain on the hardware is unnecessary given the task, which then paints a picture of negligent sloppiness in engineering and QA culture at MS.
- gruez 1y ago
  >It’s not really MS’ fault that crappy SSDs broke, but at the same time it’s only ever Windows that sees these issues see sibling comment for something similar happening on zfs. linux is also filled with workarounds for crappy SSDs, so "SSDs breaking due to software" isn't exactly exclusively a windows problem. It just gets more attention because of the bigger user base. >This suggests that this kind of strain on the hardware is unnecessary given the task, which then paints a picture of negligent sloppiness in engineering and QA culture at MS. "straining hardware" shouldn't be a thing for PC components. If it can't handle a given workload, it should throttle/queue the requests, not brick itself. I can grant some leeway for long term wear (eg. heat damage or electromigration), but that's clearly not what's happening here. Can you imagine a network card that bricks itself if you send too many packets, and you have to baby it with how many packets you send in any given amount of time?
  - Ekaros 1y ago
    GPUs and CPUs control their clock frequency perfectly well depending on load and temperature. I see no reason why every single other component could not do the same.
- recursive 1y ago
  Windows has a lot of deployments so it makes sense to see it there more often. I don't think Windows has the most hardware compatibility issues.
ragequittah 1y ago
There was an NVIDIA update recently [1] that caused problems with temperatures being properly reported. Entirely software, killed some video cards. Though entirely nvidia's fault I wouldn't say it's because they make garbage hardware in any way. It seems increasingly these problems (including the NVIDIA one) are only cropping up on 24H2 machines which is why I've decided to stick with 23H2 until it's out of its support window. Unfortunately that's coming up pretty soon. [1]https://www.neowin.net/news/nvidia-fixes-windows-11-24h2-driver-issues-but-completely-breaks-gpu-temperature-reading/ https://www.neowin.net/news/nvidia-fixes-windows-11-24h2-dri...