My Snapdragon Dev Kit was healthy and working fine until a Windows update failed

Original Article ↗ Hacker News Discussion ↗

Was Windows Update to Blame?

  • One camp argues the timing is too coincidental: the device worked, an update failed, and it never booted reliably again.
  • They point out the update reportedly contained UEFI patches; a bad firmware push or mis‑targeted handheld firmware could plausibly trash the boot environment.
  • Others counter that this looks like classic hardware failure coinciding with a big update: random freezes at different boot stages and unpredictable reboots suggest flaky hardware more than a clean software/boot-chain bug.
  • Several suggest the reverse causality: failing hardware caused the Windows update to fail, not the other way around.

Hardware Failure Theories and Debugging Ideas

  • Common suspects: dying SSD (especially after heavy writes), bad RAM, marginal power rails, degraded capacitors, or thermal issues (fans ramping hard, device known to run hot).
  • Multiple commenters note hardware often “fails” on reboot or under unusual code paths, after having been technically broken for months.
  • Suggestions include:
    • Running Memtest86 from UEFI.
    • Swapping the SSD or trying a fresh Windows install on another ARM system and moving the drive back.
    • Booting an Ubuntu ARM ISO with experimental Snapdragon support.
    • Checking cooling and voltages if dev‑kit documentation/test points are available.
  • There’s discussion of why updates are inherently stressful: recompiles, heavy disk I/O, high sustained CPU usage, and firmware flashes all push marginal components over the edge.

Windows on ARM, Snapdragon, and Linux

  • The specific Snapdragon dev kit is confirmed canceled/recalled, but Snapdragon laptops and newer X‑series chips are still active.
  • Many are skeptical of Windows on ARM as a platform: emulation gaps, weaker game support, and concerns that Microsoft isn’t investing enough to make it first‑class.
  • Debate over Linux support:
    • Some say “it can’t run Linux”; others correct this as false but concede that current ARM laptops run Linux poorly compared to x86 or Apple Silicon.
    • View that, for non‑Apple vendors, serious hardware must have solid Linux support; otherwise it’s not worth buying.

Updates, Firmware, and Support Culture

  • A few advocate disabling Windows Update entirely on EOL/experimental hardware to avoid firmware surprises.
  • Others criticize the broader industry norm of opaque firmware pushes, short support windows, and shrugging off bricked dev kits, contrasting this with older vendors that supported obsolete systems diligently.