• molasses, last few days?

    From J. P. Gilliver@3:633/10 to All on Sunday, November 15, 2026 01:43:21
    In the last day or two, my computer (Windows 10, 64 bit) has started to
    run very slowly - but only in _some_ things.

    I first noticed it in Thunderbird. For example, spacing through the
    posts in a newsgroup proceeds smoothly as normal - but when I get to the
    end of the last one, the "move to next unread in <next 'group>" message
    appears oddly: the window it's going to appear in appears, but the "move
    ..." text, and yes and no buttons, don't for a few seconds. Restarting
    TB didn't cure it. (I'm posting this with TB; touch wood, going OK.)

    Restarting Windows - including a shutdown then startup, not just a
    "restart" - didn't.

    After a restart, and before browser or TB opened: the normal clock (at
    right end of taskbar), which I _think_ normally increments seconds much
    like my wristwatch, is now changing only every about 6 seconds,
    sometimes longer. (With the right time, i. e. it's not gaining or losing significantly.) But my chiming clock (ClockSmithLite), which chimes on
    the quarter, does so at the right moment, even if that moment is between
    pauses in the displayed time. Task Manager seems to update about once a
    second (its default?) without hesitation. Above all, the mouse cursor is totally responsive - if I move the mouse, the cursor moves with no
    hesitation.

    Sometimes it changes to the blue circle (what would have been an
    hourglass in earlier Windows) when over the taskbar, which also
    sometimes seems to take ages to update - for example, if I run task
    manager it shows an icon in the taskbar with a blue line under it, but
    if I close TM - which closes almost instantly - that icon doesn't
    disappear from the taskbar for 10-12 seconds. During which I can
    continue typing this text no problem.

    Obvious things:

    CrstalDiskInfo shows the SSD at 95%: not sure if that's good or bad
    news, but the main thing is it's been at that for weeks if not months, certainly well before the slowdown. (That was my main concern: back in I
    think Windows 98 days, a friend's PC was fine for some activities, very
    slow for others, which eventually traced to a fying HD: anything
    operating from RAM, such as emails most of the time, was fine, but e. g. booting took a quarter of an hour!)

    Speedfan shows the GPU (it only shows temperature for the GPU and "HDD")
    as well below temperatures it was operating at happily during the
    summer. ("HDD" too - that's in the 30s [C].)

    Task Manager doesn't show high CPU or RAM load, and is very responsive.

    BitMeter2 shows minimal network activity.

    I am not aware of having installed or upgraded anything recently -
    certainly not that I can correlate with the change in behaviour. (Of
    course, some things may have "up"graded themselves, such as the OS,
    though for W10-64 that shouldn't be doing much by now, although I am on
    the extended thingy. The only other thing I can think of is AVG
    [whatever you get for free]; I've just tried to find a history of
    updates in it, but can't find one, but am not _aware_ of it having done
    so - it tells me viruses are up to date but AVG itself has a new version available dated 30 October, so obviously _that_ hasn't updated in the
    last few days.)

    Loading a 3.6G video file (Casablanca!) went and plays fine (it took a
    while to start, but I think a file that bog always did); I could jump to
    a point half way through or near the end and that was almost instant.

    PC remains usable, with occasional irritating waits for some things to respond/happen, but obviously I don't want it to get worse, so if anyone
    knows what's happened and how to fix it ...
    --
    J. P. Gilliver. UMRA: 1960/<1985 MB++G()ALIS-Ch++(p)Ar++T+H+Sh0!:`)DNAf

    "If god doesn't like the way I live, Let him tell me, not you." - unknown


    --- PyGate Linux v1.5.19
    * Origin: Dragon's Lair, PyGate NNTP<>Fido Gate (3:633/10)
  • From J. P. Gilliver@3:633/10 to All on Sunday, November 15, 2026 12:50:11
    On 2026/11/15 4:2:27, Paul wrote:
    On Sat, 11/14/2026 8:43 PM, J. P. Gilliver wrote:
    In the last day or two, my computer (Windows 10, 64 bit) has started to
    run very slowly - but only in _some_ things.

    I first noticed it in Thunderbird. For example, spacing through the
    posts in a newsgroup proceeds smoothly as normal - but when I get to the
    end of the last one, the "move to next unread in <next 'group>" message
    appears oddly: the window it's going to appear in appears, but the "move
    ..." text, and yes and no buttons, don't for a few seconds. Restarting
    TB didn't cure it. (I'm posting this with TB; touch wood, going OK.)

    Restarting Windows - including a shutdown then startup, not just a
    "restart" - didn't.

    After a restart, and before browser or TB opened: the normal clock (at
    right end of taskbar), which I _think_ normally increments seconds much
    like my wristwatch, is now changing only every about 6 seconds,
    sometimes longer. (With the right time, i. e. it's not gaining or losing
    significantly.) But my chiming clock (ClockSmithLite), which chimes on
    the quarter, does so at the right moment, even if that moment is between
    pauses in the displayed time. Task Manager seems to update about once a
    second (its default?) without hesitation. Above all, the mouse cursor is
    totally responsive - if I move the mouse, the cursor moves with no
    hesitation.

    Sometimes it changes to the blue circle (what would have been an
    hourglass in earlier Windows) when over the taskbar, which also
    sometimes seems to take ages to update - for example, if I run task
    manager it shows an icon in the taskbar with a blue line under it, but
    if I close TM - which closes almost instantly - that icon doesn't
    disappear from the taskbar for 10-12 seconds. During which I can
    continue typing this text no problem.

    Obvious things:

    CrstalDiskInfo shows the SSD at 95%: not sure if that's good or bad
    news, but the main thing is it's been at that for weeks if not months,
    certainly well before the slowdown. (That was my main concern: back in I
    think Windows 98 days, a friend's PC was fine for some activities, very
    slow for others, which eventually traced to a fying HD: anything
    operating from RAM, such as emails most of the time, was fine, but e. g.
    booting took a quarter of an hour!)

    Speedfan shows the GPU (it only shows temperature for the GPU and "HDD")
    as well below temperatures it was operating at happily during the
    summer. ("HDD" too - that's in the 30s [C].)

    Task Manager doesn't show high CPU or RAM load, and is very responsive.

    BitMeter2 shows minimal network activity.

    I am not aware of having installed or upgraded anything recently -
    certainly not that I can correlate with the change in behaviour. (Of
    course, some things may have "up"graded themselves, such as the OS,
    though for W10-64 that shouldn't be doing much by now, although I am on
    the extended thingy. The only other thing I can think of is AVG
    [whatever you get for free]; I've just tried to find a history of
    updates in it, but can't find one, but am not _aware_ of it having done
    so - it tells me viruses are up to date but AVG itself has a new version
    available dated 30 October, so obviously _that_ hasn't updated in the
    last few days.)

    Loading a 3.6G video file (Casablanca!) went and plays fine (it took a
    while to start, but I think a file that bog always did); I could jump to
    a point half way through or near the end and that was almost instant.

    PC remains usable, with occasional irritating waits for some things to
    respond/happen, but obviously I don't want it to get worse, so if anyone
    knows what's happened and how to fix it ...


    You should mention as well, that you're on the ESU and are still getting security updates. And your symptoms are showing up, right after Patch Tuesday.
    KB5126657. The people who are not on ESU, may not be seeing this.

    Paul

    I did say "I am on the extended thingy" )I couldn't remember what it was called). But I'm interested to hear if anyone else _has_ experienced a slow-to-do-_some_-things lately. (FWIW, I think mine started rather
    _after_ Tuesday, but I guess such updates may not happen to everybody on
    the actual Tuesday.)

    (Is there an easy - that's easy for me, not easy for Paul! - way of
    trying to undo that KB, assuming my system has installed it, to see if
    it's the cause of the slowdown?)
    --
    J. P. Gilliver. UMRA: 1960/<1985 MB++G()ALIS-Ch++(p)Ar++T+H+Sh0!:`)DNAf

    The best way to achieve immortality is by not dying.

    --- PyGate Linux v1.5.19
    * Origin: Dragon's Lair, PyGate NNTP<>Fido Gate (3:633/10)
  • From Paul in Houston TX@3:633/10 to All on Sunday, November 15, 2026 08:39:56
    J. P. Gilliver wrote:
    On 2026/11/15 4:2:27, Paul wrote:
    On Sat, 11/14/2026 8:43 PM, J. P. Gilliver wrote:
    In the last day or two, my computer (Windows 10, 64 bit) has started to
    run very slowly - but only in _some_ things.
    ...

    (Is there an easy - that's easy for me, not easy for Paul! - way of
    trying to undo that KB, assuming my system has installed it, to see if
    it's the cause of the slowdown?)

    System Restore?

    --- PyGate Linux v1.5.19
    * Origin: Dragon's Lair, PyGate NNTP<>Fido Gate (3:633/10)
  • From Frank Slootweg@3:633/10 to All on Sunday, November 15, 2026 16:14:29
    J. P. Gilliver <G6JPG@255soft.uk> wrote:
    [...]
    (Is there an easy - that's easy for me, not easy for Paul! - way of
    trying to undo that KB, assuming my system has installed it, to see if
    it's the cause of the slowdown?)

    Isn't 'Uninstall updates' on the (bottom of the) 'Update history'
    page? (It's in Windows 11 and AFAIK also in 10, but I can't check
    anymore.)

    And please don't follow the System Restore advice. System Restore
    should only be used as an utter last resort as it's broken by design.

    --- PyGate Linux v1.5.19
    * Origin: Dragon's Lair, PyGate NNTP<>Fido Gate (3:633/10)
  • From J. P. Gilliver@3:633/10 to All on Sunday, November 15, 2026 16:34:02
    On 2026/11/15 16:14:29, Frank Slootweg wrote:
    J. P. Gilliver <G6JPG@255soft.uk> wrote:
    [...]
    (Is there an easy - that's easy for me, not easy for Paul! - way of
    trying to undo that KB, assuming my system has installed it, to see if

    it's the cause of the slowdown?)

    Isn't 'Uninstall updates' on the (bottom of the) 'Update history'
    page? (It's in Windows 11 and AFAIK also in 10, but I can't check
    anymore.)

    And please don't follow the System Restore advice. System Restore
    should only be used as an utter last resort as it's broken by design.

    It tells me there are no restore points anyway.

    I had a look in add/remove prog.s. or whatever it's called under 10
    (Uninstall or change a program), as I had memories of seeing KBs in
    there in the past, but no luck there; clicking the Installed On column
    heading to sort by that, the only things it _shows_ as restored this
    month are Edge and Copilot on the 15th (today) and Chrome on the 11th,
    which must all be updates, as I haven't installed any of them this
    month. I have tried uninstalling Copilot (as I never installed and don't
    use it), and that seemed to disappear fine, but the slowdown is still
    there (though I haven't done a restart yet).
    --
    J. P. Gilliver. UMRA: 1960/<1985 MB++G()ALIS-Ch++(p)Ar++T+H+Sh0!:`)DNAf

    Eddie [Waring] underook elocution lessons in Leeds. After four weeks he
    was asked to leave - all members in his class had begun to speak like
    him. Stuart Hall, RT 7-13 August 2010

    --- PyGate Linux v1.5.19
    * Origin: Dragon's Lair, PyGate NNTP<>Fido Gate (3:633/10)
  • From Frank Slootweg@3:633/10 to All on Sunday, November 15, 2026 16:52:27
    J. P. Gilliver <G6JPG@255soft.uk> wrote:
    On 2026/11/15 16:14:29, Frank Slootweg wrote:
    J. P. Gilliver <G6JPG@255soft.uk> wrote:
    [...]
    (Is there an easy - that's easy for me, not easy for Paul! - way of
    trying to undo that KB, assuming my system has installed it, to see if
    it's the cause of the slowdown?)

    Isn't 'Uninstall updates' on the (bottom of the) 'Update history'
    page? (It's in Windows 11 and AFAIK also in 10, but I can't check
    anymore.)

    You didn't respond to this part. Did you check whether 'Uninstall
    updates' is on the 'Update history' page (of *Windows Update*, *not* of 'Uninstall or change a program')?

    I.e. in Windows 11 and probably in 10: Settings -> Windows Update ->
    Update history -> (bottom of page ) Uninstall updates

    And please don't follow the System Restore advice. System Restore
    should only be used as an utter last resort as it's broken by design.

    It tells me there are no restore points anyway.

    I had a look in add/remove prog.s. or whatever it's called under 10 (Uninstall or change a program), as I had memories of seeing KBs in
    there in the past, but no luck there;...)

    Yes, it had 'View installed updates' and still has, but now it
    redirects back to 'Uninstall updates' of the 'Settings' UI, not of
    Control Panel (again for Windows 11, not sure about 10).

    [...]

    --- PyGate Linux v1.5.19
    * Origin: Dragon's Lair, PyGate NNTP<>Fido Gate (3:633/10)
  • From J. P. Gilliver@3:633/10 to All on Monday, November 16, 2026 05:09:48
    SORTED!

    On 2026/11/15 16:52:27, Frank Slootweg wrote:
    []
    You didn't respond to this part. Did you check whether 'Uninstall
    updates' is on the 'Update history' page (of *Windows Update*, *not* of 'Uninstall or change a program')?

    I.e. in Windows 11 and probably in 10: Settings -> Windows Update ->
    Update history -> (bottom of page ) Uninstall updates

    Thanks for that. I eventually found that list, and there were only two
    this month - a stack one on -11, and a security one on -12. I tried uninstalling the later (thinking it's probably best to do them in
    reverse order); after much messing about it failed, and when I looked
    again, they showed as stack -11, security -15. I tried again, and it
    seemed (though taking ages) to go, but was still there. (I tried the
    earlier one, but uninstall wasn't an option, presumably because of the
    later one.)

    And please don't follow the System Restore advice. System Restore
    should only be used as an utter last resort as it's broken by design.

    It tells me there are no restore points anyway.

    I had a look in add/remove prog.s. or whatever it's called under 10
    (Uninstall or change a program), as I had memories of seeing KBs in
    there in the past, but no luck there;...)

    Yes, it had 'View installed updates' and still has, but now it
    redirects back to 'Uninstall updates' of the 'Settings' UI, not of
    Control Panel (again for Windows 11, not sure about 10).

    [...]
    I gave up, and restored C: from an image - fortunately my monthly one
    was only from the 3rd. Did this with a lot of help from Google's AI, to
    save and restore my settings for Chrome, Thunderbird, and System
    Scheduler. (Incidentally, it said do more than just the profile, in
    order to get a couple of files that were from the directory above.)

    All went well - though almost certainly unnecessary:

    Although the image was made before the slowdown, it was still in
    evidence after the restore :-(. With considerable help from Google AI
    again, we eventually found the culprit: ShowBatteryBar, a utility I'd
    had in for ages. Well, the _actual_ culprit was/is my battery, which
    seems to have died altogether (or, I hope, just jarred loose). But
    apparently ShowBatteryBar - which is apparently quite old - not only interrogates the battery, but updates the taskbar - but, if it doesn't
    receive any reply from the battery, holds up all sorts of things you'd
    think unrelated (such as Thunderbird, when switching to the next
    newsgroup; things started from the taskbar; and the start menu).
    Eventually these time out, but it slows down the system considerably to
    the point of making it almost unusable - the most obvious being that the taskbar clock was only updating (with the right time, but only) about
    every six seconds. (That was the cue that we'd found the culprit, as it reverted to counting seconds normally once I'd turned SBB off.)

    Ah well - experience!
    --
    J. P. Gilliver. UMRA: 1960/<1985 MB++G()ALIS-Ch++(p)Ar++T+H+Sh0!:`)DNAf

    "What happens if I press this button?"
    "I wouldn't ..." (pinggg!) "Oh!" "What happened?"
    "A sign lit up, saying `please do not press this button again'!"(s1f2)

    --- PyGate Linux v1.5.19
    * Origin: Dragon's Lair, PyGate NNTP<>Fido Gate (3:633/10)
  • From Paul@3:633/10 to All on Monday, November 16, 2026 00:38:00
    On Mon, 11/16/2026 12:09 AM, J. P. Gilliver wrote:
    SORTED!

    On 2026/11/15 16:52:27, Frank Slootweg wrote:
    []
    You didn't respond to this part. Did you check whether 'Uninstall
    updates' is on the 'Update history' page (of *Windows Update*, *not* of
    'Uninstall or change a program')?

    I.e. in Windows 11 and probably in 10: Settings -> Windows Update ->
    Update history -> (bottom of page ) Uninstall updates

    Thanks for that. I eventually found that list, and there were only two
    this month - a stack one on -11, and a security one on -12. I tried uninstalling the later (thinking it's probably best to do them in
    reverse order); after much messing about it failed, and when I looked
    again, they showed as stack -11, security -15. I tried again, and it
    seemed (though taking ages) to go, but was still there. (I tried the
    earlier one, but uninstall wasn't an option, presumably because of the
    later one.)

    And please don't follow the System Restore advice. System Restore
    should only be used as an utter last resort as it's broken by design.

    It tells me there are no restore points anyway.

    I had a look in add/remove prog.s. or whatever it's called under 10
    (Uninstall or change a program), as I had memories of seeing KBs in
    there in the past, but no luck there;...)

    Yes, it had 'View installed updates' and still has, but now it
    redirects back to 'Uninstall updates' of the 'Settings' UI, not of
    Control Panel (again for Windows 11, not sure about 10).

    [...]
    I gave up, and restored C: from an image - fortunately my monthly one
    was only from the 3rd. Did this with a lot of help from Google's AI, to
    save and restore my settings for Chrome, Thunderbird, and System
    Scheduler. (Incidentally, it said do more than just the profile, in
    order to get a couple of files that were from the directory above.)

    All went well - though almost certainly unnecessary:

    Although the image was made before the slowdown, it was still in
    evidence after the restore :-(. With considerable help from Google AI
    again, we eventually found the culprit: ShowBatteryBar, a utility I'd
    had in for ages. Well, the _actual_ culprit was/is my battery, which
    seems to have died altogether (or, I hope, just jarred loose). But
    apparently ShowBatteryBar - which is apparently quite old - not only interrogates the battery, but updates the taskbar - but, if it doesn't receive any reply from the battery, holds up all sorts of things you'd
    think unrelated (such as Thunderbird, when switching to the next
    newsgroup; things started from the taskbar; and the start menu).
    Eventually these time out, but it slows down the system considerably to
    the point of making it almost unusable - the most obvious being that the taskbar clock was only updating (with the right time, but only) about
    every six seconds. (That was the cue that we'd found the culprit, as it reverted to counting seconds normally once I'd turned SBB off.)

    Ah well - experience!


    And you had fun, right ?

    It makes you wonder how all those things were tied together like that.

    Once you'd IDed the problem, did the LLM-AI write an explanation
    of how those items are locked together ?

    Paul


    --- PyGate Linux v1.5.19
    * Origin: Dragon's Lair, PyGate NNTP<>Fido Gate (3:633/10)
  • From Frank Slootweg@3:633/10 to All on Monday, November 16, 2026 11:22:51
    J. P. Gilliver <G6JPG@255soft.uk> wrote:
    SORTED!

    Glad you found the culprit!

    On 2026/11/15 16:52:27, Frank Slootweg wrote:
    []
    You didn't respond to this part. Did you check whether 'Uninstall updates' is on the 'Update history' page (of *Windows Update*, *not* of 'Uninstall or change a program')?

    I.e. in Windows 11 and probably in 10: Settings -> Windows Update -> Update history -> (bottom of page ) Uninstall updates

    Thanks for that. I eventually found that list, and there were only two
    this month - a stack one on -11, and a security one on -12. I tried uninstalling the later (thinking it's probably best to do them in
    reverse order); after much messing about it failed, and when I looked
    again, they showed as stack -11, security -15. I tried again, and it
    seemed (though taking ages) to go, but was still there. (I tried the
    earlier one, but uninstall wasn't an option, presumably because of the
    later one.)

    Some after the fact advice: It seems that you used 'Uninstall
    updates', but didn't do a 'Pause updates' (or whatever the Windows 10
    term is) *before* doing the 'Uninstall updates'. If so, Windows Update
    will happily do its automatic check-for-updates routine and re-install
    the update(s) you just unininstalled.

    [...]

    [Report of successful (but unneeded) recovery deleted.]

    --- PyGate Linux v1.5.19
    * Origin: Dragon's Lair, PyGate NNTP<>Fido Gate (3:633/10)
  • From Carlos E.R.@3:633/10 to All on Tuesday, November 17, 2026 12:05:17
    On 2026-11-16 06:09, J. P. Gilliver wrote:

    ...

    Although the image was made before the slowdown, it was still in
    evidence after the restore ?. With considerable help from Google AI
    again, we eventually found the culprit: ShowBatteryBar, a utility I'd
    had in for ages. Well, the_actual_ culprit was/is my battery, which
    seems to have died altogether (or, I hope, just jarred loose). But
    apparently ShowBatteryBar - which is apparently quite old - not only interrogates the battery, but updates the taskbar - but, if it doesn't receive any reply from the battery, holds up all sorts of things you'd
    think unrelated (such as Thunderbird, when switching to the next
    newsgroup; things started from the taskbar; and the start menu).
    Eventually these time out, but it slows down the system considerably to
    the point of making it almost unusable - the most obvious being that the taskbar clock was only updating (with the right time, but only) about
    every six seconds. (That was the cue that we'd found the culprit, as it reverted to counting seconds normally once I'd turned SBB off.)

    Ah well - experience!

    Interesting.

    Did you try to see CPU load, while it was going on?


    As I was reading the thread, I had a suspicion, which now is unimportant
    as you found the culprit:

    CrstalDiskInfo shows the SSD at 95%: not sure if that's good or bad
    news, but the main thing is it's been at that for weeks if not months, certainly well before the slowdown.

    95% what? Life, or used size?

    SSDs do not like being full, they turn slow as they try to find space
    free and actually cleared out to write again.

    Life percent has no effect, till really old and goes R/O.

    --
    Cheers, Carlos.
    ES??, EU??;

    --- PyGate Linux v1.5.19
    * Origin: Dragon's Lair, PyGate NNTP<>Fido Gate (3:633/10)
  • From J. P. Gilliver@3:633/10 to All on Tuesday, November 17, 2026 20:33:58
    On 2026/11/16 5:38:0, Paul wrote:
    On Mon, 11/16/2026 12:09 AM, J. P. Gilliver wrote:
    SORTED!
    []
    apparently ShowBatteryBar - which is apparently quite old - not only
    []
    Ah well - experience!


    And you had fun, right ?

    It makes you wonder how all those things were tied together like that.

    I gave up some years ago trying to understand how - let alone why -
    various parts of Windows are interconnected; I leave that to experts
    like you, Paul (though I have recently got the sense that even you are
    giving up a bit, and certainly are exasperated).

    Once you'd IDed the problem, did the LLM-AI write an explanation
    of how those items are locked together ?

    I think it did, but it's quite good at detecting my mindset, so probably
    didn't push any explanation. Although I _know_ it's just software, I
    _do_ quite enjoy chatting to it - whether troubleshooting a computer
    problem as here, or finding an alternative to a food product, or
    discussing why some instructions say "hot, but not boiling". Though I do
    have to learn not to trust it so much; I specifically asked it about
    saving Chrome things - tabs, bookmarks, extensions, and settings -
    before doing the image restore, and it assured me that saving the folder
    "User Data" (and all below, of course) would save everything; which
    turned out to be only true in a limited sense. To be fair, I didn't
    explicitly tell it that restoring from an image would restore a(n only slightly, but enough) earlier version of Chrome which then argued with
    the saved data from the later version (not on format but on encryption);
    but you'd have thought it might have thought of that ...

    Paul

    John
    --
    J. P. Gilliver. UMRA: 1960/<1985 MB++G()ALIS-Ch++(p)Ar++T+H+Sh0!:`)DNAf

    There are two rules for success:
    1. Never reveal everything you know.

    --- PyGate Linux v1.5.19
    * Origin: Dragon's Lair, PyGate NNTP<>Fido Gate (3:633/10)
  • From J. P. Gilliver@3:633/10 to All on Tuesday, November 17, 2026 20:37:15
    On 2026/11/16 11:22:51, Frank Slootweg wrote:
    J. P. Gilliver <G6JPG@255soft.uk> wrote:
    SORTED!
    []
    Thanks for that. I eventually found that list, and there were only two
    this month - a stack one on -11, and a security one on -12. I tried
    uninstalling the later (thinking it's probably best to do them in
    reverse order); after much messing about it failed, and when I looked
    again, they showed as stack -11, security -15. I tried again, and it
    seemed (though taking ages) to go, but was still there. (I tried the
    earlier one, but uninstall wasn't an option, presumably because of the
    later one.)

    Some after the fact advice: It seems that you used 'Uninstall
    updates', but didn't do a 'Pause updates' (or whatever the Windows 10
    term is) *before* doing the 'Uninstall updates'. If so, Windows Update
    will happily do its automatic check-for-updates routine and re-install
    the update(s) you just unininstalled.

    Yes, that explains why after the restart required by removing the
    November 12 one, it reappeared but now dated -15.

    [...]

    [Report of successful (but unneeded) recovery deleted.]

    Indeed. Turned out not to be the updates at all; just my luck that my
    battery died (or shook loose) just after Patch Tuesday.
    --
    J. P. Gilliver. UMRA: 1960/<1985 MB++G()ALIS-Ch++(p)Ar++T+H+Sh0!:`)DNAf

    There are two rules for success:
    1. Never reveal everything you know.

    --- PyGate Linux v1.5.19
    * Origin: Dragon's Lair, PyGate NNTP<>Fido Gate (3:633/10)
  • From J. P. Gilliver@3:633/10 to All on Tuesday, November 17, 2026 20:49:42
    On 2026/11/17 11:5:17, Carlos E.R. wrote:
    On 2026-11-16 06:09, J. P. Gilliver wrote:
    []
    apparently ShowBatteryBar - which is apparently quite old - not only
    []
    Ah well - experience!

    Interesting.

    Did you try to see CPU load, while it was going on?

    Yes, that was one of the things I looked at (and the AI suggested). It
    wasn't railing anything.

    As I was reading the thread, I had a suspicion, which now is unimportant
    as you found the culprit:

    CrstalDiskInfo shows the SSD at 95%: not sure if that's good or bad
    news, but the main thing is it's been at that for weeks if not months,
    certainly well before the slowdown.

    95% what? Life, or used size?

    CrystalDiskInfo displays a box near the top left of its window, labelled
    Health Status, and containing

    Good
    95 %

    It also shows SMART parameter CA as "Lifetime Remaining" which it also
    shows as 95 (Current 95, Worst 95, Threshold 1, Raw Value 000000000005 -
    I don't know what units).

    It _is_ very full (a lot of surveillance camera videos I was reviewing
    for a friend and hadn't deleted); that _was_ something the AI thought
    could be contributing, but I deleted lots of them and it didn't make any difference. the only thing that did - and that immediately - was turning
    off the old battery monitor utility.

    SSDs do not like being full, they turn slow as they try to find space
    free and actually cleared out to write again.

    Life percent has no effect, till really old and goes R/O.

    It's a "CT480BX500SSD1 : 480.1 GB" according to CrystalDiskInfo, which
    the AI said is a budget Crucial model, with no RAM (or some such statement).
    --
    J. P. Gilliver. UMRA: 1960/<1985 MB++G()ALIS-Ch++(p)Ar++T+H+Sh0!:`)DNAf

    There are two rules for success:
    1. Never reveal everything you know.

    --- PyGate Linux v1.5.19
    * Origin: Dragon's Lair, PyGate NNTP<>Fido Gate (3:633/10)
  • From Carlos E.R.@3:633/10 to All on Tuesday, November 17, 2026 22:10:59
    On 2026-11-17 21:49, J. P. Gilliver wrote:
    On 2026/11/17 11:5:17, Carlos E.R. wrote:
    On 2026-11-16 06:09, J. P. Gilliver wrote:
    []
    apparently ShowBatteryBar - which is apparently quite old - not only
    []
    Ah well - experience!

    Interesting.

    Did you try to see CPU load, while it was going on?

    Yes, that was one of the things I looked at (and the AI suggested). It
    wasn't railing anything.

    As I was reading the thread, I had a suspicion, which now is unimportant
    as you found the culprit:

    CrstalDiskInfo shows the SSD at 95%: not sure if that's good or bad
    news, but the main thing is it's been at that for weeks if not months,
    certainly well before the slowdown.

    95% what? Life, or used size?

    CrystalDiskInfo displays a box near the top left of its window, labelled Health Status, and containing

    Good
    95 %

    It also shows SMART parameter CA as "Lifetime Remaining" which it also
    shows as 95 (Current 95, Worst 95, Threshold 1, Raw Value 000000000005 -
    I don't know what units).

    Must be %.


    It _is_ very full (a lot of surveillance camera videos I was reviewing
    for a friend and hadn't deleted); that _was_ something the AI thought
    could be contributing, but I deleted lots of them and it didn't make any difference. the only thing that did - and that immediately - was turning
    off the old battery monitor utility.

    Deleting files in an SSD doesn't work immediately. The sectors have to
    be actually cleared which is done by a process in background. I think it
    runs inside the disk firmware. It may be called "trim".

    Ask the Google AI "what does "trim" do on an SSD disk?" and you get a
    good answer about it :-)



    SSDs do not like being full, they turn slow as they try to find space
    free and actually cleared out to write again.

    Life percent has no effect, till really old and goes R/O.

    It's a "CT480BX500SSD1 : 480.1 GB" according to CrystalDiskInfo, which
    the AI said is a budget Crucial model, with no RAM (or some such statement).

    Can't say, I'm not really an expert on SSDs :-)

    --
    Cheers, Carlos.
    ES??, EU??;

    --- PyGate Linux v1.5.19
    * Origin: Dragon's Lair, PyGate NNTP<>Fido Gate (3:633/10)
  • From Paul@3:633/10 to All on Wednesday, November 18, 2026 04:57:57
    On Wed, 11/18/2026 3:46 AM, Daniel70 wrote:
    On 18/11/2026 8:10 am, Carlos E.R. wrote:
    On 2026-11-17 21:49, J. P. Gilliver wrote:
    On 2026/11/17 11:5:17, Carlos E.R. wrote:
    On 2026-11-16 06:09, J. P. Gilliver wrote:

    <Snip>

    CrstalDiskInfo shows the SSD at 95%: not sure if that's good or bad
    news, but the main thing is it's been at that for weeks if not months, >>>>> certainly well before the slowdown.

    95% what? Life, or used size?

    CrystalDiskInfo displays a box near the top left of its window, labelled >>> Health Status, and containing

    ˙˙˙˙Good
    ˙˙˙˙95 %

    It also shows SMART parameter CA as "Lifetime Remaining" which it also
    shows as 95 (Current 95, Worst 95, Threshold 1, Raw Value 000000000005 - >>> I don't know what units).

    Must be %.

    It _is_ very full (a lot of surveillance camera videos I was reviewing
    for a friend and hadn't deleted); that _was_ something the AI thought
    could be contributing, but I deleted lots of them and it didn't make any >>> difference. the only thing that did - and that immediately - was turning >>> off the old battery monitor utility.

    Deleting files in an SSD doesn't work immediately. The sectors have to be actually
    cleared which is done by a process in background. I think it runs inside the
    disk firmware. It may be called "trim".

    Ask the Google AI "what does "trim" do on an SSD disk?" and you get a good answer about it :-)

    Does 'Defrag' achieve anything with a SSD disk?? Would it even shift all
    the remaining Files on the Disk towards the start of the SSD Disk??


    SSDs have "zero seek time", while on HDD it takes time for the head
    to go from one place to another. Even a head switch costs 1 millisecond.
    A seek could cost 10 milliseconds. Whereas an SSD is maybe 0.1 millisecond
    (and what is happening during that time, is reading a block of flash
    and putting it inside an internal buffer).

    Preferential positioning on an SSD is not necessary. Not even a little bit.

    *******

    If you have a lot of persistent shadows (VSS) set up on SSD, then defragmentation can compensate for slow COW (Copy On Write) performance.
    Since this is not normally the case (there is one software product
    that uses a ton of those, the rest hardly any at all ), defragmentation
    is not typically recommended.

    If you look at an SSD, you might see:

    04:02:08 Fragments Bytes Clusters Name

    04:02:10 195 1775366 434 C:\ProgramData\Microsoft\EdgeUpdate\Log\MicrosoftEdgeUpdate.log
    04:02:10 9 421888 103 C:\Windows\Logs\WindowsUpdate\WindowsUpdate.20261116.024004.647.1.etl
    04:02:10 3 40960 10 C:\Windows\ServiceProfiles\LocalService\AppData\Local\UsageAndQualityInsights\Logs\UQI.20261117_072938_634.etl
    04:02:10 2 4766 2 C:\Users\paul\AppData\Local\Microsoft\Edge\User Data\Breadcrumbs
    04:02:10 4 57344 14 C:\ProgramData\Microsoft\Windows Defender\Support\MpWppCoreTracing-20261116-013254-00000003-100000000.bin
    04:02:10 2 65536 16 C:\Users\paul\AppData\Local\NVIDIA\DXCache\fc52a9c95004c224.nvph
    04:02:10 5 73208 18 C:\ProgramData\Microsoft\Windows\Power Efficiency Diagnostics\energy-report-latest.xml
    04:02:10 3 40960 10 C:\Windows\ServiceProfiles\LocalService\AppData\Local\UsageAndQualityInsights\Logs\UQI.20261116_172957_601.etl
    04:02:10 2 16384 4 C:\Windows\Logs\waasmedic\waasmedic.20261117_120604_965.etl
    04:02:10 2 4676 2 C:\Windows\System32\Tasks\Microsoft\Windows\SoftwareProtectionPlatform\SvcRestartTask
    04:02:10 2 5046 2 C:\Users\paul\AppData\Local\Microsoft\Edge\User Data\Default\Code Cache\js\2971f42b8dd126b3_0

    Files that are fragmented, are split in two pieces. Things
    that are logs collected slowly, there are more opportunities
    for write collisions and fragmentation. While there are a *lot*
    of fragmented files, there aren't a lot of important files
    with 1000 fragments.

    [Picture] Yellow is fragmented, green is good, red is "unmoveable" [not a problem...]

    https://imgur.com/a/ud8xL9I

    https://postimg.cc/gwcffkpq
    https://i.postimg.cc/k5WqRBhZ/jkdefrag-ssd-c.gif

    *******

    Pathological fragmentation can be tested with Passmark Fragmenter.
    It does such a good job, my RAMDrive slowed to 80% normal speed.
    When each file has thousands of fragments, there are many many
    calls into the file system stack, and that eventually slows
    the thing down.

    And that is "pathological fragmentation". Normal usage isn't quite that bad. Hard drives tend to chop into more fragments, as the drive is slower
    and there can be more "overlaps of equal things".

    *******

    Also of note, there is a difference in slowdown on an SSD due to "mushy flash", versus a slowdown caused by excessive fragmentation. I bought some shitty SSDs on purpose, to observe them, and they really are dreadful. At first, I thought I had a bargain on my hands, nope, they're not a bargain. They're a liability as I have to back them up (as I don't know when the thing is going to show
    a CRC error). The shitty drive is slow, because each sector needs error correction by the ARM core inside the drive. A good SSD has three or four
    ARM cores, and most of the cores work on error correction. One core is used
    as a command interpreter and is the overall boss.

    I just re-paved one of the shitty SSDs, and it's back to 500MB/sec after
    the Macrium restore. But this won't last, and if I check that SSD drive
    three months from now, it'll be running at HDD speed due to all the error correction again.

    The shitty SSD is mostly annoying when you look at benchmark plots.
    If someone blindfolded you and put a random SSD in the machine,
    you might not be able to tell whether it was the shitty one or the
    good one. It's only when you benchmark it becomes painfully obvious.

    And sometimes, I have file copies running at weird speeds, and when
    i have to stop and determine what is going on, that's usually some
    sad ass thing slowing it down :-) This is what happens when you
    own a few whizzy things, they let you down just a little bit and
    you're all excited, whipped into a froth about it. If everything
    you own is slow, there is less to get excited about. If you get
    my meaning. For example, if I owned 14.9GB/sec NVMe and it
    started running at 14.8GB/sec, you could see my blood pressure
    rising :-) If it was a HDD doing 150MB/sec and then 140MB/sec,
    I would "hardly notice".

    Paul

    --- PyGate Linux v1.5.19
    * Origin: Dragon's Lair, PyGate NNTP<>Fido Gate (3:633/10)
  • From Carlos E.R.@3:633/10 to All on Wednesday, November 18, 2026 11:02:44
    On 2026-11-18 09:46, Daniel70 wrote:
    Does 'Defrag' achieve anything with a SSD disk?? Would it even shift all
    the remaining Files on the Disk towards the start of the SSD Disk??

    Not recommended. The improvement can not be measured, it is so small,
    even zero. And the disk suffers significant wear.

    --
    Cheers, Carlos.
    ES??, EU??;

    --- PyGate Linux v1.5.19
    * Origin: Dragon's Lair, PyGate NNTP<>Fido Gate (3:633/10)
  • From J. P. Gilliver@3:633/10 to All on Wednesday, November 18, 2026 12:49:17
    On 2026/11/18 9:57:57, Paul wrote:
    On Wed, 11/18/2026 3:46 AM, Daniel70 wrote:
    []
    Does 'Defrag' achieve anything with a SSD disk?? Would it even shift all
    the remaining Files on the Disk towards the start of the SSD Disk??

    Paul and Carlos have already answered this (Paul of course in much
    greater detail than I), but I'll have a go just for completeness.

    SSDs have "zero seek time", while on HDD it takes time for the head
    to go from one place to another. Even a head switch costs 1 millisecond.
    A seek could cost 10 milliseconds. Whereas an SSD is maybe 0.1 millisecond (and what is happening during that time, is reading a block of flash
    and putting it inside an internal buffer).

    (a) fragmentation _is_ a problem on a spinning disc - if a file is in
    lots of bits, then you have to wait for the head to move around as it
    reads the file. On an SSD, it takes the same time (or very close to it)
    to "move" (there's no physical movement, it's just reading different
    sectors of memory) from the beginning to the end as it does just moving
    between successive sectors. So whether a file is all in one place or
    scattered around the SSD, reading it will take much the same time.
    (b) though a lesser effect (I think), spinning discs - assuming they
    rotate at a constant angular velocity such as 7200 RPM - read more
    sectors per revolution on a track with more sectors in it; I think they
    tend to store more sectors in the outer tracks, usually considered to be
    the "start" of the disc, so moving everything to the start does speed up reading somewhat.

    Preferential positioning on an SSD is not necessary. Not even a little bit.

    And actually _not_ to be recommended, since any moving around involves
    read and writes, which is the major wear factor on an SSD (more so than
    on an HDD, anyway) - and there's nothing to be _gained_ by it, since an
    SSD reads at (or very close to) the same speed wherever on the "disc"
    it's reading from.
    []
    that uses a ton of those, the rest hardly any at all ), defragmentation
    is not typically recommended.
    []
    Actually counter-recommended, for the reason above. If it's an old
    defrag utility that does not know about SSDs, it will just do a lot of unnecessary (and degrading) read/writes. I'm not sure if a defrag
    utility that _does_ know about SSDs will actually do anything
    beneficial; I suspect it'll just not do /anything/ if it _knows_ it's an
    SSD it's being pointed at. (There may be _something_ to do with "trim" -
    I so far haven't been able to quite grasp what that's about.)
    --
    J. P. Gilliver. UMRA: 1960/<1985 MB++G()ALIS-Ch++(p)Ar++T+H+Sh0!:`)DNAf

    If you believe in telekinesis, raise my right hand

    --- PyGate Linux v1.5.19
    * Origin: Dragon's Lair, PyGate NNTP<>Fido Gate (3:633/10)
  • From J. P. Gilliver@3:633/10 to All on Wednesday, November 18, 2026 13:12:35
    On 2026/11/18 12:58:40, Daniel70 wrote:
    On 18/11/2026 11:25 pm, Daniel70 wrote:
    []
    Ah!! Well, there's another task of my "To Do" list.

    Thank you, Paul.

    s/task off my/task of my

    That's the second time I've seen it in the last day or two - not sure if
    it was you the first time, but anyway, IIRR, the syntax of s/// was:

    s/<string to be replaced>/replacement string>/

    , not the other way round. (And the final / _is_ needed.)
    --
    J. P. Gilliver. UMRA: 1960/<1985 MB++G()ALIS-Ch++(p)Ar++T+H+Sh0!:`)DNAf

    There will always be people who don't value music unless it's written
    for the concert hall - Debbie Wiseman (composer), RT 2025-5-24-30

    --- PyGate Linux v1.5.19
    * Origin: Dragon's Lair, PyGate NNTP<>Fido Gate (3:633/10)
  • From Frank Slootweg@3:633/10 to All on Wednesday, November 18, 2026 16:35:30
    J. P. Gilliver <G6JPG@255soft.uk> wrote:
    On 2026/11/18 12:58:40, Daniel70 wrote:
    On 18/11/2026 11:25 pm, Daniel70 wrote:
    []
    Ah!! Well, there's another task of my "To Do" list.

    Thank you, Paul.

    s/task off my/task of my

    That's the second time I've seen it in the last day or two - not sure if
    it was you the first time, but anyway, IIRR, the syntax of s/// was:

    s/<string to be replaced>/replacement string>/

    Of course.

    , not the other way round. (And the final / _is_ needed.)

    FWIW, the final slash is *not* needed in all the (mostly vi and vim)
    editors I have used over the last four decades. But I mostly *do* put it
    there anyway.

    --- PyGate Linux v1.5.19
    * Origin: Dragon's Lair, PyGate NNTP<>Fido Gate (3:633/10)
  • From Carlos E.R.@3:633/10 to All on Thursday, November 19, 2026 12:15:27
    On 2026-11-18 13:49, J. P. Gilliver wrote:
    beneficial; I suspect it'll just not do/anything/ if it_knows_ it's an
    SSD it's being pointed at. (There may be_something_ to do with "trim" -
    I so far haven't been able to quite grasp what that's about.)

    Just ask Google in AI mode, the answer is quite good. :-)

    --
    Cheers, Carlos.
    ES??, EU??;

    --- PyGate Linux v1.5.19
    * Origin: Dragon's Lair, PyGate NNTP<>Fido Gate (3:633/10)
  • From J. P. Gilliver@3:633/10 to All on Thursday, November 19, 2026 13:48:45
    On 2026/11/19 11:15:27, Carlos E.R. wrote:
    On 2026-11-18 13:49, J. P. Gilliver wrote:
    beneficial; I suspect it'll just not do/anything/ if it_knows_ it's an
    SSD it's being pointed at. (There may be_something_ to do with "trim" -
    I so far haven't been able to quite grasp what that's about.)

    Just ask Google in AI mode, the answer is quite good. :-)

    Interesting: I normally use Google starting at https://www.google.co.uk/advanced_search (though admit I haven't used
    other than the first box for ages), but I don't know how to select "AI
    mode"; normally I find that if what I put in is in the form of a
    question, or really anything that a simple search won't answer, I get
    the AI chattybot anyway.
    --
    J. P. Gilliver. UMRA: 1960/<1985 MB++G()ALIS-Ch++(p)Ar++T+H+Sh0!:`)DNAf

    "One of my dearest memories is playing the leader of a gang of gay
    Hell's Angels thundering across the Golden Gate bridge on a motorbike
    in fog, wearing full Nazi regalia with a young man in a purple dress on
    the pillion petrified we'd crash into the bay."
    Christopher Lee (1997). ["It was in _The Serial_."]

    --- PyGate Linux v1.5.19
    * Origin: Dragon's Lair, PyGate NNTP<>Fido Gate (3:633/10)
  • From Paul@3:633/10 to All on Thursday, November 19, 2026 09:48:35
    On Thu, 11/19/2026 6:15 AM, Carlos E.R. wrote:
    On 2026-11-18 13:49, J. P. Gilliver wrote:
    beneficial; I suspect it'll just not do/anything/ if it_knows_ it's an
    SSD it's being pointed at. (There may be_something_ to do with "trim" -
    I so far haven't been able to quite grasp what that's about.)

    Just ask Google in AI mode, the answer is quite good. :-)


    The classification of storage devices, for usage in the "Optimize"
    window, is done via "WinSAT formal" (a benchmark). Find the recipe for WinSAT formal,
    just for the HDD and SSDs in the box, run it, and the Optimize
    window can have the device types "corrected" if they were
    previously incorrect. I think I fixed that once, using
    that method.

    At least one classification is not measured by speed, it's a
    property of the thing. A thin provisioned storage container,
    receives a shabby defrag, as an example of a classification
    you would not want. And a "WinSAT Formal" run is unlikely
    to fix that for you.

    "Optimize" offers TRIM or defrag, according to the determined
    device type. And an SSD is not identified by probing the
    metadata for SSD parameters (as SMART differs between HDD
    and SSD). "WinSAT Formal" seems to play a part in it.
    One of my devices is mis-labeled as "fibre channel". The
    interface has a few different labels it can use.

    Paul


    --- PyGate Linux v1.5.19
    * Origin: Dragon's Lair, PyGate NNTP<>Fido Gate (3:633/10)
  • From Frank Slootweg@3:633/10 to All on Thursday, November 19, 2026 15:10:14
    J. P. Gilliver <G6JPG@255soft.uk> wrote:
    On 2026/11/19 11:15:27, Carlos E.R. wrote:
    On 2026-11-18 13:49, J. P. Gilliver wrote:
    beneficial; I suspect it'll just not do/anything/ if it_knows_ it's an
    SSD it's being pointed at. (There may be_something_ to do with "trim" -
    I so far haven't been able to quite grasp what that's about.)

    Just ask Google in AI mode, the answer is quite good. :-)

    Interesting: I normally use Google starting at https://www.google.co.uk/advanced_search (though admit I haven't used
    other than the first box for ages), but I don't know how to select "AI
    mode"; normally I find that if what I put in is in the form of a
    question, or really anything that a simple search won't answer, I get
    the AI chattybot anyway.

    Just start Google without the '/advanced_search' bit, so in your case <https://www.google.co.uk>

    The 'AI Mode' button is to the right of the search box, next to the microphone and Google Lens icons. (FWIW, I never use it that way, but
    there it is.)

    --- PyGate Linux v1.5.19
    * Origin: Dragon's Lair, PyGate NNTP<>Fido Gate (3:633/10)
  • From J. P. Gilliver@3:633/10 to All on Thursday, November 19, 2026 16:29:32
    On 2026/11/19 15:10:14, Frank Slootweg wrote:
    J. P. Gilliver <G6JPG@255soft.uk> wrote:
    On 2026/11/19 11:15:27, Carlos E.R. wrote:
    On 2026-11-18 13:49, J. P. Gilliver wrote:
    beneficial; I suspect it'll just not do/anything/ if it_knows_ it's an >>>> SSD it's being pointed at. (There may be_something_ to do with "trim" - >>>> I so far haven't been able to quite grasp what that's about.)

    Just ask Google in AI mode, the answer is quite good. :-)

    Interesting: I normally use Google starting at
    https://www.google.co.uk/advanced_search (though admit I haven't used
    other than the first box for ages), but I don't know how to select "AI
    mode"; normally I find that if what I put in is in the form of a
    question, or really anything that a simple search won't answer, I get
    the AI chattybot anyway.

    Just start Google without the '/advanced_search' bit, so in your case <https://www.google.co.uk>

    The 'AI Mode' button is to the right of the search box, next to the microphone and Google Lens icons. (FWIW, I never use it that way, but
    there it is.)

    Thanks - I see it. Though as I said I usually get the AIbot anyway.
    --
    J. P. Gilliver. UMRA: 1960/<1985 MB++G()ALIS-Ch++(p)Ar++T+H+Sh0!:`)DNAf

    What a strange illusion it is to suppose that beauty is goodness.
    -Leo Tolstoy, novelist and philosopher (1828-1910)

    --- PyGate Linux v1.5.19
    * Origin: Dragon's Lair, PyGate NNTP<>Fido Gate (3:633/10)
  • From Paul@3:633/10 to All on Thursday, November 19, 2026 13:48:01
    On Thu, 11/19/2026 10:10 AM, Frank Slootweg wrote:
    J. P. Gilliver <G6JPG@255soft.uk> wrote:
    On 2026/11/19 11:15:27, Carlos E.R. wrote:
    On 2026-11-18 13:49, J. P. Gilliver wrote:
    beneficial; I suspect it'll just not do/anything/ if it_knows_ it's an >>>> SSD it's being pointed at. (There may be_something_ to do with "trim" - >>>> I so far haven't been able to quite grasp what that's about.)

    Just ask Google in AI mode, the answer is quite good. :-)

    Interesting: I normally use Google starting at
    https://www.google.co.uk/advanced_search (though admit I haven't used
    other than the first box for ages), but I don't know how to select "AI
    mode"; normally I find that if what I put in is in the form of a
    question, or really anything that a simple search won't answer, I get
    the AI chattybot anyway.

    Just start Google without the '/advanced_search' bit, so in your case <https://www.google.co.uk>

    The 'AI Mode' button is to the right of the search box, next to the microphone and Google Lens icons. (FWIW, I never use it that way, but
    there it is.)


    That depends on which browser you use.

    I guess browser wars are still a thing, because even the
    MSEdge web browser, Google presented a "use Chrome" advert
    right in the middle of the page, which had to be dismissed
    before I took the screen shot. Such weirdness. Such hopeful
    thinking. Like "I wanna pony" over and over again. MSEdge
    *is* Chrome, only with someone elses shit in it.

    [Picture] Google-web-page-diffs.gif

    https://imgur.com/a/wqbLxfi

    https://postimg.cc/56nMh3jb
    https://i.postimg.cc/6pN5GbX2/Google-web-page-diffs.gif

    As for the junk injected on the end of each URL, I did not type
    those URLs like that. It was a plain "www.google.com" entry and
    it comes back decorated for tracking in the case of MSEdge.

    A "modern Firefox", looks the same as MSEdge, only *without* the
    advert for Chrome :-) Hahaha. What ? Now you don't wanna pony ?
    It also lacks the inline tracking on the end of the URL. Maybe
    that's part of Manifest V3 or something (the Firefox didn't have it).

    It's good that the topic came up, as otherwise I wouldn't have
    done another check for differences. Maybe I will also find that
    the shit I get in one browser for search results, will differ
    in another browser.

    Paul

    --- PyGate Linux v1.5.19
    * Origin: Dragon's Lair, PyGate NNTP<>Fido Gate (3:633/10)
  • From Frank Slootweg@3:633/10 to All on Thursday, November 19, 2026 19:36:15
    Paul <nospam@needed.invalid> wrote:
    On Thu, 11/19/2026 10:10 AM, Frank Slootweg wrote:
    J. P. Gilliver <G6JPG@255soft.uk> wrote:
    On 2026/11/19 11:15:27, Carlos E.R. wrote:
    On 2026-11-18 13:49, J. P. Gilliver wrote:
    beneficial; I suspect it'll just not do/anything/ if it_knows_ it's an >>>> SSD it's being pointed at. (There may be_something_ to do with "trim" - >>>> I so far haven't been able to quite grasp what that's about.)

    Just ask Google in AI mode, the answer is quite good. :-)

    Interesting: I normally use Google starting at
    https://www.google.co.uk/advanced_search (though admit I haven't used
    other than the first box for ages), but I don't know how to select "AI
    mode"; normally I find that if what I put in is in the form of a
    question, or really anything that a simple search won't answer, I get
    the AI chattybot anyway.

    Just start Google without the '/advanced_search' bit, so in your case <https://www.google.co.uk>

    The 'AI Mode' button is to the right of the search box, next to the microphone and Google Lens icons. (FWIW, I never use it that way, but
    there it is.)

    That depends on which browser you use.

    Possibly. John reported that he sees what I wrote. I don't know which
    browser he uses (Firefox)?

    You showed that Edge shows what I wrote and obviously so does Chrome.

    You also showed an odd-one-out, Seamonkey, which doesn't show the
    three buttons, but shows an 'Advanced search' button instead. That's
    probably a Seamonkey-specific thing.

    [Left for reference:]

    I guess browser wars are still a thing, because even the
    MSEdge web browser, Google presented a "use Chrome" advert
    right in the middle of the page, which had to be dismissed
    before I took the screen shot. Such weirdness. Such hopeful
    thinking. Like "I wanna pony" over and over again. MSEdge
    *is* Chrome, only with someone elses shit in it.

    [Picture] Google-web-page-diffs.gif

    https://imgur.com/a/wqbLxfi

    https://postimg.cc/56nMh3jb
    https://i.postimg.cc/6pN5GbX2/Google-web-page-diffs.gif

    As for the junk injected on the end of each URL, I did not type
    those URLs like that. It was a plain "www.google.com" entry and
    it comes back decorated for tracking in the case of MSEdge.

    A "modern Firefox", looks the same as MSEdge, only *without* the
    advert for Chrome :-) Hahaha. What ? Now you don't wanna pony ?
    It also lacks the inline tracking on the end of the URL. Maybe
    that's part of Manifest V3 or something (the Firefox didn't have it).

    It's good that the topic came up, as otherwise I wouldn't have
    done another check for differences. Maybe I will also find that
    the shit I get in one browser for search results, will differ
    in another browser.

    --- PyGate Linux v1.5.19
    * Origin: Dragon's Lair, PyGate NNTP<>Fido Gate (3:633/10)