• Return from Computer Hell

    From jason_warren@jason_warren@ieee.org to alt.comp.os.windows-11 on Thu Dec 4 10:39:20 2025
    From Newsgroup: alt.comp.os.windows-11

    I posted earlier about how the disk numbering changed:
    disk zero became disk one and vice versa. I still don't
    know how that happened, but after a month I think I'm back
    from Computer Hell.

    The Alienware R9 suddenly began to run very slowly. I ran
    every diagnostic I could find - the built-in ones and
    others. All passed, including the long versions that took
    hours to complete.

    I am diligent about backups and so I restored the SSD and
    spinning SCSI drive, but that didn't solve the problem.
    The machine still took half an hour just to boot, and I
    could hear the SCSI disk rattling a lot throughout.

    Today I did a "clean everything" fresh Win 11
    installation. The install proceeded quickly (I'd tried
    before and it was very slow). Forcing the Win installation
    to use the SSD cut install time from more than an hour to
    15 minutes.

    By and by, Windows reported that there was a slew of
    updates to apply, about 30. That's not surprising
    considering how long it had been since the last update.
    But what I noticed, and hadn't seen before, was the fact
    that four of the updates were for the (Intel) SCSI
    adapter! I've seen plenty of updates over the years, but
    they've always been for drivers.

    After the installation and required reboot, the machine
    took off and is now running just as it always had.
    There's more cleanup to do (and a fresh backup!). So, I'm
    left wondering what happened.

    I have an idea (please don't laugh). I've been a ham radio
    amateur since the 60's. We are now experiencing solar
    cycle 25 and it's been the strongest in years. Cycles are
    on an ~11-year cycle. A strong CME (Coronal Mass Ejection)
    floods nearby space (us!) with a LOT of energetic
    particles that are known to be able to disrupt sensitive
    circuitry, sometimes momentarily, other times actually
    damaging the minute gates in today's super-dense chips.
    Could the SCSI adapter have gotten dinged by a charged
    particle?


    --- Synchronet 3.21a-Linux NewsLink 1.2
  • From knuttle@keith_nuttle@yahoo.com to alt.comp.os.windows-11 on Thu Dec 4 12:20:11 2025
    From Newsgroup: alt.comp.os.windows-11

    On 12/04/2025 10:39 AM, jason_warren wrote:
    I posted earlier about how the disk numbering changed:
    disk zero became disk one and vice versa. I still don't
    know how that happened, but after a month I think I'm back
    from Computer Hell.

    The Alienware R9 suddenly began to run very slowly. I ran
    every diagnostic I could find - the built-in ones and
    others. All passed, including the long versions that took
    hours to complete.

    I am diligent about backups and so I restored the SSD and
    spinning SCSI drive, but that didn't solve the problem.
    The machine still took half an hour just to boot, and I
    could hear the SCSI disk rattling a lot throughout.

    Today I did a "clean everything" fresh Win 11
    installation. The install proceeded quickly (I'd tried
    before and it was very slow). Forcing the Win installation
    to use the SSD cut install time from more than an hour to
    15 minutes.

    By and by, Windows reported that there was a slew of
    updates to apply, about 30. That's not surprising
    considering how long it had been since the last update.
    But what I noticed, and hadn't seen before, was the fact
    that four of the updates were for the (Intel) SCSI
    adapter! I've seen plenty of updates over the years, but
    they've always been for drivers.

    After the installation and required reboot, the machine
    took off and is now running just as it always had.
    There's more cleanup to do (and a fresh backup!). So, I'm
    left wondering what happened.

    I have an idea (please don't laugh). I've been a ham radio
    amateur since the 60's. We are now experiencing solar
    cycle 25 and it's been the strongest in years. Cycles are
    on an ~11-year cycle. A strong CME (Coronal Mass Ejection)
    floods nearby space (us!) with a LOT of energetic
    particles that are known to be able to disrupt sensitive
    circuitry, sometimes momentarily, other times actually
    damaging the minute gates in today's super-dense chips.
    Could the SCSI adapter have gotten dinged by a charged
    particle?


    Remember when everything has been eleminated what is left probably is
    the reason.

    OR just a misfortunate occurance of random variables.
    --- Synchronet 3.21a-Linux NewsLink 1.2
  • From MikeS@mikes@is.invalid to alt.comp.os.windows-11 on Thu Dec 4 21:54:43 2025
    From Newsgroup: alt.comp.os.windows-11

    On 04/12/2025 15:39, jason_warren wrote:
    I posted earlier about how the disk numbering changed:
    disk zero became disk one and vice versa. I still don't
    know how that happened, but after a month I think I'm back
    from Computer Hell.

    The Alienware R9 suddenly began to run very slowly. I ran
    every diagnostic I could find - the built-in ones and
    others. All passed, including the long versions that took
    hours to complete.

    I am diligent about backups and so I restored the SSD and
    spinning SCSI drive, but that didn't solve the problem.
    The machine still took half an hour just to boot, and I
    could hear the SCSI disk rattling a lot throughout.

    Today I did a "clean everything" fresh Win 11
    installation. The install proceeded quickly (I'd tried
    before and it was very slow). Forcing the Win installation
    to use the SSD cut install time from more than an hour to
    15 minutes.

    By and by, Windows reported that there was a slew of
    updates to apply, about 30. That's not surprising
    considering how long it had been since the last update.
    But what I noticed, and hadn't seen before, was the fact
    that four of the updates were for the (Intel) SCSI
    adapter! I've seen plenty of updates over the years, but
    they've always been for drivers.

    After the installation and required reboot, the machine
    took off and is now running just as it always had.
    There's more cleanup to do (and a fresh backup!). So, I'm
    left wondering what happened.

    I have an idea (please don't laugh). I've been a ham radio
    amateur since the 60's. We are now experiencing solar
    cycle 25 and it's been the strongest in years. Cycles are
    on an ~11-year cycle. A strong CME (Coronal Mass Ejection)
    floods nearby space (us!) with a LOT of energetic
    particles that are known to be able to disrupt sensitive
    circuitry, sometimes momentarily, other times actually
    damaging the minute gates in today's super-dense chips.
    Could the SCSI adapter have gotten dinged by a charged
    particle?

    No idea what went wrong but I am certain it was not because your SCSI
    adapter was dinged by a charged particle. You just installed Win 11 on
    an SSD so why would the SCSI adapter affect its boot time with or
    without new drivers?
    --- Synchronet 3.21a-Linux NewsLink 1.2
  • From Paul@nospam@needed.invalid to alt.comp.os.windows-11 on Thu Dec 4 18:39:16 2025
    From Newsgroup: alt.comp.os.windows-11

    On Thu, 12/4/2025 4:54 PM, MikeS wrote:
    On 04/12/2025 15:39, jason_warren wrote:
    I posted earlier about how the disk numbering changed:
    disk zero became disk one and vice versa. I still don't
    know how that happened, but after a month I think I'm back
    from Computer Hell.

    The Alienware R9 suddenly began to run very slowly. I ran
    every diagnostic I could find - the built-in ones and
    others. All passed, including the long versions that took
    hours to complete.

    I am diligent about backups and so I restored the SSD and
    spinning SCSI drive, but that didn't solve the problem.
    The machine still took half an hour just to boot, and I
    could hear the SCSI disk rattling a lot throughout.

    Today I did a "clean everything" fresh Win 11
    installation. The install proceeded quickly (I'd tried
    before and it was very slow). Forcing the Win installation
    to use the SSD cut install time from more than an hour to
    15 minutes.

    By and by, Windows reported that there was a slew of
    updates to apply, about 30. That's not surprising
    considering how long it had been since the last update.
    But what I noticed, and hadn't seen before, was the fact
    that four of the updates were for the (Intel) SCSI
    adapter! I've seen plenty of updates over the years, but
    they've always been for drivers.

    After the installation and required reboot, the machine
    took off and is now running just as it always had.
    There's more cleanup to do (and a fresh backup!). So, I'm
    left wondering what happened.

    I have an idea (please don't laugh). I've been a ham radio
    amateur since the 60's. We are now experiencing solar
    cycle 25 and it's been the strongest in years. Cycles are
    on an ~11-year cycle. A strong CME (Coronal Mass Ejection)
    floods nearby space (us!) with a LOT of energetic
    particles that are known to be able to disrupt sensitive
    circuitry, sometimes momentarily, other timesĀ  actually
    damaging the minute gates in today's super-dense chips.
    Could the SCSI adapter have gotten dinged by a charged
    particle?

    No idea what went wrong but I am certain it was not because
    your SCSI adapter was dinged by a charged particle. You just
    installed Win 11 on an SSD so why would the SCSI adapter affect
    its boot time with or without new drivers?

    Disks receive analysis before they are mounted. Windows uses
    automount in a sense. The reason Windows can CHKDSK so many
    partitions sequentially at startup, is due to autochk.

    (ControlSet001 here is likely to be CurrentControlSet)

    HKEY_LOCAL_MACHINE\SYSTEM\ControlSet001\Control\Session Manager
    BootExecute REG_MULTI_SZ autocheck autochk *

    You would need to find a log of SCSI sense codes, to get some
    idea what the "complaint" was on this subsystem. On Windows,
    SCSI CDB (control data block) are used for "foreign" devices,
    and you would be surprised how many subsystems use that. SCSI is
    the interface proxy for unknown device types, allowing hobbyists
    to glue things to an OS.

    The first outboard SATA controller, the SIL3112, it used
    a two-deep driver stack, and one of the layers, its job was
    to translate a SCSI CDB into something a SATA interface could use.
    This means, one of those driver layers is pretty thin.

    The "secret" to storage, is good quality logging.

    Even a Linux boot takes three minutes, when it is
    distracted by some secondary drive having a bad hair day.
    The "sudo dmesg" on Ubuntu, shows you what is going on.
    That's not a fancy logging system, but many people use that
    in lieu of learning how the "latest hot thing" in logging works.
    In the same way that nobody here really wants to use EventVwr
    for anything that matters.

    Paul
    --- Synchronet 3.21a-Linux NewsLink 1.2
  • From jason_warren@jason_warren@ieee.org to alt.comp.os.windows-11 on Thu Dec 4 19:24:59 2025
    From Newsgroup: alt.comp.os.windows-11

    In the mid sixties, NCAR in Boulder installed a 360/65
    system. NCAR had been 100% Cray, so this was considered a
    Big Deal. The trouble was, the machine kept failing with
    random machine checks (parity errors and the like).
    Someone suggested that the cause might be cosmic rays, so
    IBM installed a mod 65 on top of a mountain near Boulder.
    It failed all the time until shielding was added on top of
    the machine which was normally open to allow cooling air
    to flow. The machine stopped failing, and shielding was
    added to the design.


    --- Synchronet 3.21a-Linux NewsLink 1.2
  • From MikeS@mikes@is.invalid to alt.comp.os.windows-11 on Fri Dec 5 10:16:10 2025
    From Newsgroup: alt.comp.os.windows-11

    On 05/12/2025 00:24, jason_warren wrote:
    In the mid sixties, NCAR in Boulder installed a 360/65
    system. NCAR had been 100% Cray, so this was considered a
    Big Deal. The trouble was, the machine kept failing with
    random machine checks (parity errors and the like).
    Someone suggested that the cause might be cosmic rays, so
    IBM installed a mod 65 on top of a mountain near Boulder.
    It failed all the time until shielding was added on top of
    the machine which was normally open to allow cooling air
    to flow. The machine stopped failing, and shielding was
    added to the design.


    1. Micro electronics have come a long way since the mid sixties.
    2. I doubt you live on top of a mountain.
    3. Have you considered that if you are correct about the effect of a
    solar burst on your SCSI adapter there would have been widespread
    reports about similarly affected devices in sensitive applications?
    --- Synchronet 3.21a-Linux NewsLink 1.2