The problem is, the computer will "go to sleep" but then wake up
because the USB Hub wants a drink of water, or someone to tell it a
story, or because there's a monster under the bed. What that means in
turn is that when the computer monitors "click on" in the middle of
the night, they wake me up.

What I want to know is how do I make the computer actually stay asleep
till morning or I wake it up?

The work around I've found is to have a Custom Power Management set to
shut off monitors after 5 minutes and "sleep" to 10.

Related Question: How many of the 5 "HID Generic Keyboards" do I
really need?

--
pyotr filipivich
"We have now sunk to a depth at which restatement of the obvious
is the first duty of intelligent men." George Orwell

On 7/14/2022 7:15 PM, pyotr filipivich wrote:
>
> The problem is, the computer will "go to sleep" but then wake up
> because the USB Hub wants a drink of water, or someone to tell it a
> story, or because there's a monster under the bed. What that means in
> turn is that when the computer monitors "click on" in the middle of
> the night, they wake me up.
>
> What I want to know is how do I make the computer actually stay asleep
> till morning or I wake it up?
>
> The work around I've found is to have a Custom Power Management set to
> shut off monitors after 5 minutes and "sleep" to 10.
>
> Related Question: How many of the 5 "HID Generic Keyboards" do I
> really need?

Do you really need the computer to be instantly awake when you get up in
the morning? Why not merely turn it off just before you go to sleep?

When I leave my PC for more than 2-3 minutes, I disable its Internet
connection. If I leave the house or go to bed, I shut it down.

--
David E. Ross
"A Message to Those Who Are Not Vaccinated"
See my <http://www.rossde.com/index.html#vaccine>.

Il giorno Fri 15 Jul 2022 05:41:58a, *David E. Ross* ha inviato su
alt.windows7.general il messaggio news:taqnm7$11v0$1@gioia.aioe.org.
Vediamo cosa ha scritto:

> Do you really need the computer to be instantly awake when you get up in
> the morning? Why not merely turn it off just before you go to sleep?
>

yes, some BIOS have a tab to schedule automatic power on (and off) at a
precise hour:minutes, even different for weekdays (i.e. excluding Saturday
and Sunday)

--
/-\ /\/\ /\/\ /-\ /\/\ /\/\ /-\ T /-\
-=- -=- -=- -=- -=- -=- -=- -=- - -=-
............ [ al lavoro ] ...........

"pyotr filipivich" <phamp@mindspring.com> wrote

| What I want to know is how do I make the computer actually stay asleep
| till morning or I wake it up?
| You're lucky. Mine will sleep but then often refuses
to wake up. Then I have to unplug it so I can reboot.
I've never figured it out. Tried a couple of things, but
I mostly only use my Win7 box for streaming movies
to a TV, so it's not a big deal.

Have you checked the settings in Device Manager?
Under the Power Management tab, make sure only the
keyboard is set to allow wake up. If you allow the mouse
then any little vibration will wake it up. I've never heard of
any connection to anything else being able to wake up
a computer, though I do have an option grayed out on
my USB hubs on XP. In any case, just disable that except
for the keyboard....
Or turn off your monitor. Or better yet, turn off the
computer.

"David E. Ross" <nobody@notme.invalid> on Thu, 14 Jul 2022 20:41:58
-0700 typed in alt.windows7.general the following:
>On 7/14/2022 7:15 PM, pyotr filipivich wrote:
>>
>> The problem is, the computer will "go to sleep" but then wake up
>> because the USB Hub wants a drink of water, or someone to tell it a
>> story, or because there's a monster under the bed. What that means in
>> turn is that when the computer monitors "click on" in the middle of
>> the night, they wake me up.
>>
>> What I want to know is how do I make the computer actually stay asleep
>> till morning or I wake it up?
>>
>> The work around I've found is to have a Custom Power Management set to
>> shut off monitors after 5 minutes and "sleep" to 10.
>>
>> Related Question: How many of the 5 "HID Generic Keyboards" do I
>> really need?
>
>Do you really need the computer to be instantly awake when you get up in
>the morning? Why not merely turn it off just before you go to sleep?

Ah yes. There seems to be a second issue: when booting up in the
morning, I can not only get a cup of coffee, but grind and brew a pot
of coffee. Except for those days when it hangs at "Starting Windows"
until I cycle the power.
>
>When I leave my PC for more than 2-3 minutes, I disable its Internet
>connection. If I leave the house or go to bed, I shut it down.
--
pyotr filipivich
This Week's Panel: Us & Them - Eliminating Them.
Next Month's Panel: Having eliminated the old Them(tm)
Selecting who insufficiently Woke(tm) as to serve as the new Them(tm)

Ammammata <ammammata@tiscalinet.it> on Fri, 15 Jul 2022 09:53:32 -0000
(UTC) typed in alt.windows7.general the following:
>Il giorno Fri 15 Jul 2022 05:41:58a, *David E. Ross* ha inviato su
>alt.windows7.general il messaggio news:taqnm7$11v0$1@gioia.aioe.org.
>Vediamo cosa ha scritto:
>
>> Do you really need the computer to be instantly awake when you get up in
>> the morning? Why not merely turn it off just before you go to sleep?
>>
>
>yes, some BIOS have a tab to schedule automatic power on (and off) at a
>precise hour:minutes, even different for weekdays (i.e. excluding Saturday
>and Sunday)

Which is great, but I do not want it to shut down completely, as
to go to sleep and stay there. No waking up to ask for a cup of
water, because there's a monster under the bed, because the cat walked
through the room, because the moon is too bright, or any other excuse.
--
pyotr filipivich
This Week's Panel: Us & Them - Eliminating Them.
Next Month's Panel: Having eliminated the old Them(tm)
Selecting who insufficiently Woke(tm) as to serve as the new Them(tm)

On Thu, 14 Jul 2022 19:15:29 -0700, pyotr filipivich wrote:
> The problem is, the computer will "go to sleep" but then wake up
> because the USB Hub wants a drink of water, or someone to tell it a
> story, or because there's a monster under the bed. What that means in
> turn is that when the computer monitors "click on" in the middle of
> the night, they wake me up.
>
> What I want to know is how do I make the computer actually stay asleep
> till morning or I wake it up?
>
> The work around I've found is to have a Custom Power Management set to
> shut off monitors after 5 minutes and "sleep" to 10.
>
> Related Question: How many of the 5 "HID Generic Keyboards" do I
> really need?

Try unplugging the mouse before putting it to sleep, because there are some
faulty mice which report mouse movement to the system even though they don't
actually cause the mouse cursor position to change.

On 7/15/2022 8:16 AM, Mayayana wrote:
> "pyotr filipivich" <phamp@mindspring.com> wrote
>
> | What I want to know is how do I make the computer actually stay asleep
> | till morning or I wake it up?
> |
> You're lucky. Mine will sleep but then often refuses
> to wake up. Then I have to unplug it so I can reboot.
> I've never figured it out. Tried a couple of things, but
> I mostly only use my Win7 box for streaming movies
> to a TV, so it's not a big deal.
>
> Have you checked the settings in Device Manager?
> Under the Power Management tab, make sure only the
> keyboard is set to allow wake up. If you allow the mouse
> then any little vibration will wake it up. I've never heard of
> any connection to anything else being able to wake up
> a computer, though I do have an option grayed out on
> my USB hubs on XP. In any case, just disable that except
> for the keyboard....
> Or turn off your monitor. Or better yet, turn off the
> computer.

Anything which generates a PME event in hardware,
can wake a computer.

To control PME events, some of them can be controlled
from Device Manager. Presumably the information at the
OS level, is being passed via ACPI to the BIOS.
"Allow this device to wake the computer" means that
the PME block for that device has been left enabled
and powered when the computer changes states.

Mouse (wake on activity)
Keyboard (wake on any keypress, or specific keypress sequence)
LAN (wake on carrier, wake on LAN magic packet, two others...)
SB timer ??? (depends on who owns BIOS timer, that lives inside SB)
(nothing in BIOS for this, with exceptions)

The BIOS is the last thing to be alive when a computer
changes states. On an ATX computer with Soft Off,
it is the BIOS which runs the last code on the computer.
On an AT computer, it is the OS which prints "it is safe
to turn off your computer" and the BIOS, who knows what
has happened to it.

If the OS were to run the RESET instruction in code,
sure, that usurps BIOS control, but it also causes a
restart, not a shutdown.

If the OS executes a HALT instruction, the next clock tick
wakes the OS scheduler, so this isn't useful as a means to
halt a modern machine. Halt does not mean what it used to.

No matter what, the Power button on the front panel
generates a PME. This is the button that SHOULD wake
the computer in your case. If it does not, it suggests
the computer has crashed, or the Southbridge status
bits that track machine state are corrupt.

If you hibernate a computer, or hybrid-sleep a computer,
there may be something in the SB that says "I am hibernated"
or "I am asleep". The BIOS response at startup, of finding
the RAM controller already running and in autorefresh,
it uses the SB "hints" left at the end of the last session
as a guide.

If you turn the power switch off at the back of the
computer, wait >60 seconds, turn the computer back
on at the back, the "hints" are lost. The last
state is effectively "I am soft off". Which means
to the BIOS, that everything needs to be initialized.

It is the OS in that case, which analyzes the C: on the
drive, notices the header on the hiberfil.sys is "valid",
and the OS loads the hiberfil.sys, rather than loading
individual files and the kernel file ("booting"). Even
if the SB hints are lost, a hibernated computer can
recover the session via hiberfil.sys analysis. It is
the OS early code, which makes the decision as to
what to do.

If a Linux CD is booting, it too can analyze the C: partition,
see the hiberfil.sys and "prevent user from mounting C: partition".
Because it knows files can be open on the C: partition at
that point. Linux wants Windows to clean up that partition,
the next time that Windows runs (if ever). Windows is
not the only thing sniffing a hiberfil.sys.

In the case of plain S3 sleep, switching off at the back
causes session loss and potentially file system damage
that a journaled file system can clean up. Fragments left
by open files are removed. If a computer crashes in sleep,
it is roughly equivalent to a power cycling scenario.

Summary: Try power button on front, a quick tap, not
a four second application. If the computer does
not wake, then it is not really in S3 sleep at
the moment, and is "messed up" as you surmised.

There are probably several signals on the SB,
indicating what is going on. These would run off
+5VSB, the rail that must be running to keep RAM
contents autorefreshed during sleep. Nobody ties
LEDs to these, so we can see what is going on.

Paul

"Paul" <nospam@needed.invalid> wrote

| Summary: Try power button on front, a quick tap, not
| a four second application. If the computer does
| not wake, then it is not really in S3 sleep at
| the moment, and is "messed up" as you surmised.
|

I've tried the power button in the past, with no effect
at all. This is a Dell and has no reset button, so I have
to unplug it to start over. So you seem to be saying
there's a hardware problem. But it doesn't always happen.
Since I don't use it regularly I haven't been systematic
about looking for patterns.

On 7/14/2022 10:15 PM, pyotr filipivich wrote:
>
> The problem is, the computer will "go to sleep" but then wake up
> because the USB Hub wants a drink of water, or someone to tell it a
> story, or because there's a monster under the bed. What that means in
> turn is that when the computer monitors "click on" in the middle of
> the night, they wake me up.
>
> What I want to know is how do I make the computer actually stay asleep
> till morning or I wake it up?
>
> The work around I've found is to have a Custom Power Management set to
> shut off monitors after 5 minutes and "sleep" to 10.
>
> Related Question: How many of the 5 "HID Generic Keyboards" do I
> really need?

As administrator, powercfg /lastwake

This might record, that the computer was asleep, and was awakened
by a particular piece of hardware.

In Device Manager, mouse, keyboard, LAN, have "wake this computer"
tick boxes. On the Intel Pro LAN, there is a pulldown menu selecting
the wake type, and it should be set to Off rather than Wake On Carrier.

In Power Options control panel, in Advanced, would be

Sleep
Sleep after
Allow wake timers
Setting: Disable <=== set to disable if you don't want
the OS to be using the timer. Disable
is the wrong setting for a "TV Recording" PC.

These are the things I'd look at.

But a broken piece of hardware could tug on PME, and
I don't know what happens then. Even PCI Express slots,
I believe there is a PME signal on them. But a semi-busted
PME would likely wake immediately and sleep would never work.
PME is Power Management Event.

Keyboards with five identities in Device Manager, I have
one of those rubbish keyboards here. One of the device
entries may have to do with LED control (backlight animation).
But the duplicate HID entries, I'm not sure what's up with that.
You have the option of using Disable on any device in Device Manager.
Just don't do something stupid. Disable disables the whole
device, while the "Allow this device to wake the computer"
tick box is only related to the PME part of the device.

Paul

On Thu, 14 Jul 2022 19:15:29 -0700, pyotr filipivich
<phamp@mindspring.com> wrote:

>
>The problem is, the computer will "go to sleep" but then wake up
>because the USB Hub wants a drink of water, or someone to tell it a
>story, or because there's a monster under the bed. What that means in
>turn is that when the computer monitors "click on" in the middle of
>the night, they wake me up.
>
>What I want to know is how do I make the computer actually stay asleep
>till morning or I wake it up?
>
>The work around I've found is to have a Custom Power Management set to
>shut off monitors after 5 minutes and "sleep" to 10.

In addition to all the other suggestions, don't forget to check Task
Scheduler, since scheduled tasks can wake your computer from sleep
too. Many are set for the wee hours of the morning since it is assumed
- usually correctly - that the user won't be doing anything of import
at the time and thus its a perfect time for the PC to do some
self-maintenance (backing up files, taking out the trash, applying
updates, even defragging the drive if you still have spinning-rust).

>Related Question: How many of the 5 "HID Generic Keyboards" do I
>really need?

Well, optimally, just the one ;-)

Still, many USB devices self-identify as a keyboard if they pass-along
keyboard-like button presses (for instance, some mice with
more-than-the-usual number of buttons do this). Or if you use a dongle
for wireless mouse, it often has the capability to also connect a
wireless keyboard, so it identifies as both.

On 15/07/2022 17:24, Paul wrote:
>
> Keyboards with five identities in Device Manager, I have
> one of those rubbish keyboards here.

Sometimes it happens that when the same keyboard is plugged into
different USB sockets, it leaves a 'ghost' of itself as a hidden device
for every USB sockets that it's ever been used on, the only live device
being the one corresponding to the USB socket in use at the moment.

Perhaps both you and Pyotr have ...
View/Show hidden devices
.... ticked (it's possible to have this set permanently, as I have), in
which case the ghosts will appear, but not appear to be connected to
anything, shown by their icon having a faint greyed out appearance.
These ghosts can be discriminated more easily by choosing ...
View/Show devices by connection
.... in which case they will be hanging off the root of the tree rather
than a USB port under the 'ACPI [something or other] PC' branch.

However in what follows be aware that while all 'ghost' devices will be
hanging off the root, not all devices hanging off the root are 'ghosts',
most will *NOT* be. The 'ghosts' can be distinguished from the fact
that you'd normally expect such a device to be hanging somewhere off the
'ACPI [something or other] PC' branch.

Such 'ghost' devices appear for any hardware that is not currently
connected, and can be deleted safely, as long as you don't mind it
having to be reinstalled next time it is connected. In effect, this
means that it's probably best to leave 'ghost' devices for things like
scanners, printers, etc, but phantom HID mice and keyboards can safely
be removed.

Returning to Pyotr's problem, I'm fairly sure that it shouldn't be
possible for a ghost device to wake the computer from sleep. For that
problem I second the suggestions of others about selecting which devices
are allowed to waken the PC. Pyotr may also care to think about setting
the computer to hibernate rather than just sleep, as that is a deeper
and more energy saving alternative, and *may* (I can't remember for
sure, but perhaps someone may care to run a test) only be wakened by
using the power button.

--

Fake news kills!

I may be contacted via the contact address given on my website:
www.macfh.co.uk

On 7/14/22 21:15, pyotr filipivich wrote:
>
> The problem is, the computer will "go to sleep" but then wake up
> because the USB Hub wants a drink of water, or someone to tell it a
> story, or because there's a monster under the bed. What that means in
> turn is that when the computer monitors "click on" in the middle of
> the night, they wake me up.

You could physically turn the monitors off when you go to sleep. That's
what I do. Monitors are never on unless I want to use the computer.

[snip]

--
Mark Lloyd
http://notstupid.us/

"Man is a credulous animal, and must believe something; in the absence
of good grounds for belief, he will be satisfied with bad ones."
[Bertrand Russell]

On 7/15/2022 12:15 PM, Mayayana wrote:
> "Paul" <nospam@needed.invalid> wrote
>
> | Summary: Try power button on front, a quick tap, not
> | a four second application. If the computer does
> | not wake, then it is not really in S3 sleep at
> | the moment, and is "messed up" as you surmised.
> |
>
> I've tried the power button in the past, with no effect
> at all. This is a Dell and has no reset button, so I have
> to unplug it to start over. So you seem to be saying
> there's a hardware problem. But it doesn't always happen.
> Since I don't use it regularly I haven't been systematic
> about looking for patterns.

The control logic on the average motherboard is
not all that systematic. There is backfeed protection,
to try to ensure that when the logic needs to "recycle"
and prepare for your button press, that this is not
defeated by currents flowing into the motherboard
from other places.

Nobody just xeroxes the Intel reference schematic and
runs off and has a beer. It's always custom-redesigned,
and if you had the schematic, you might even be able
to tell who designed your board.

One example of a foulup, is one poster discovered
their system would restart properly, if they unplugged
the VGA cable. Then pressed the button on the front. Then
plugged the VGA cable back in. As pin 9 is missing on
a lot of VGA designs, you would think this path is
no longer there. You can see the article here, is
"of two minds".

https://en.wikipedia.org/wiki/VGA_connector

There can also be temperature marginalities. Such
as a motherboard not recovering, when room temp is
below 50F or so. The paths in question, are mixed
analog-digital, have MOSFETs and so on. Spaghetti.
I can't even follow the logic through that section,
when I have a schematic in front of me.

On some chipsets, this was actually a chipset bug,
where the chipset did not work properly at 50F.
And they knew that when they shipped it. If you
had a choice, between re-spinning a chip and needing
3-6 months to do it (completely missing the sales
window), and just shipping broken shit, which
would you do :-/ Isn't it amazing that all those
chipsets are always ready on time ? What wizardry.

Paul

On 7/15/2022 1:10 PM, Java Jive wrote:
> On 15/07/2022 17:24, Paul wrote:
>>
>> Keyboards with five identities in Device Manager, I have
>> one of those rubbish keyboards here.
>
> Sometimes it happens that when the same keyboard is plugged into different USB sockets,
> it leaves a 'ghost' of itself as a hidden device for every USB sockets that it's ever
> been used on, the only live device being the one corresponding to the USB socket in
> use at the moment.
>

I can see this rubbish in the dmesg boot log when booting
a Linux LiveDVD.

I could use USBTreeView and spot a USB composite device
with the rubbish hiding behind it.

I think you can do a workup and figure it out if you want.

But just seeing the entries flash by in the dmesg,
is good for a first guess. They're real and not
some artifact (in an OS that does not have a registry).

https://www.uwe-sieber.de/files/UsbTreeView_x64.zip

======================== USB Device ========================

+++++++++++++++++ Device Information ++++++++++++++++++
Device Description : USB Composite Device
Hardware IDs : USB\VID_125F&PID_9318&REV_0102

Child Device 1 : USB Input Device
Child Device 1 : HID Keyboard Device
Child Device 2 : USB Input Device
Child Device 1 : HID-compliant mouse
Child Device 2 : HID-compliant system controller
Child Device 3 : HID-compliant consumer control device
Child Device 4 : HID-compliant vendor-defined device
Child Device 5 : HID Keyboard Device

That's my newest keyboard, and no other keyboard here
is as complicated as that one. That was selected at the
computer store, during COVID times, I wasn't allowed to
touch or examine the keyboards and my "assistant" put one
in the basket. You were escorted by one of their sales people.

These could have been pinned off, using a serial flash chip,
but they save a buck by allowing those to flap in the breeze
like that. The OS seems to know which rubbish is non-functional.

Paul

On 16/07/2022 00:49, Paul wrote:
>
> On 7/15/2022 1:10 PM, Java Jive wrote:
>>
>> On 15/07/2022 17:24, Paul wrote:
>>>
>>> Keyboards with five identities in Device Manager, I have
>>> one of those rubbish keyboards here.
>>
>> Sometimes it happens that when the same keyboard is plugged into
>> different USB sockets, it leaves a 'ghost' of itself as a hidden
>> device for every USB sockets that it's ever been used on, the
>> only live device being the one corresponding to the USB socket in
>> use at the moment.
>
> I can see this rubbish in the dmesg boot log when booting
> a Linux LiveDVD.
>
[snip]
>
>       ======================== USB Device ========================
>
>         +++++++++++++++++ Device Information ++++++++++++++++++
> Device Description       : USB Composite Device
> Hardware IDs             : USB\VID_125F&PID_9318&REV_0102
>
>  Child Device 1          : USB Input Device
>    Child Device 1        : HID Keyboard Device
>  Child Device 2          : USB Input Device
>    Child Device 1        : HID-compliant mouse
>    Child Device 2        : HID-compliant system controller
>    Child Device 3        : HID-compliant consumer control device
>    Child Device 4        : HID-compliant vendor-defined device
>    Child Device 5        : HID Keyboard Device
>
> That's my newest keyboard, and no other keyboard here
> is as complicated as that one.

Interesting, JOOI, do you have USB emulation switched on in the BIOS of
that PC, and if so, does it make any difference to the above listing if
it is switched off?

--

Fake news kills!

I may be contacted via the contact address given on my website:
www.macfh.co.uk

On 7/15/2022 8:06 PM, Java Jive wrote:
> On 16/07/2022 00:49, Paul wrote:
>>
>> On 7/15/2022 1:10 PM, Java Jive wrote:
>>>
>>> On 15/07/2022 17:24, Paul wrote:
>>>>
>>>> Keyboards with five identities in Device Manager, I have
>>>> one of those rubbish keyboards here.
>>>
>>> Sometimes it happens that when the same keyboard is plugged into different USB sockets, it leaves a 'ghost' of itself as a hidden
>>> device for every USB sockets that it's ever been used on, the
>>> only live device being the one corresponding to the USB socket in
>>> use at the moment.
>>
>> I can see this rubbish in the dmesg boot log when booting
>> a Linux LiveDVD.
>>
> [snip]
>>
>>        ======================== USB Device ========================
>>
>>          +++++++++++++++++ Device Information ++++++++++++++++++
>> Device Description       : USB Composite Device
>> Hardware IDs             : USB\VID_125F&PID_9318&REV_0102
>>
>>   Child Device 1          : USB Input Device
>>     Child Device 1        : HID Keyboard Device
>>   Child Device 2          : USB Input Device
>>     Child Device 1        : HID-compliant mouse
>>     Child Device 2        : HID-compliant system controller
>>     Child Device 3        : HID-compliant consumer control device
>>     Child Device 4        : HID-compliant vendor-defined device
>>     Child Device 5        : HID Keyboard Device
>>
>> That's my newest keyboard, and no other keyboard here
>> is as complicated as that one.
>
> Interesting, JOOI, do you have USB emulation switched on in the BIOS of that PC, and if so, does it make any difference to the above listing if it is switched off?
>

Not that I know of.

Paul

"David E. Ross" <nobody@notme.invalid> wrote in message
news:taqnm7$11v0$1@gioia.aioe.org...
> On 7/14/2022 7:15 PM, pyotr filipivich wrote:
>>
>> The problem is, the computer will "go to sleep" but then wake up
>> because the USB Hub wants a drink of water, or someone to tell it a
>> story, or because there's a monster under the bed. What that means in
>> turn is that when the computer monitors "click on" in the middle of
>> the night, they wake me up.
>>
>> What I want to know is how do I make the computer actually stay asleep
>> till morning or I wake it up?
>>
>> The work around I've found is to have a Custom Power Management set to
>> shut off monitors after 5 minutes and "sleep" to 10.
>>
>> Related Question: How many of the 5 "HID Generic Keyboards" do I
>> really need?
>
> Do you really need the computer to be instantly awake when you get up in
> the morning? Why not merely turn it off just before you go to sleep?

It depends how long it takes the PC to boot up and for all the background
processes to become fully functional.

My Windows 7 PC takes a shade under 10 minutes until everything is running:
Skype and Dropbox icons in the system tray take ages to appear and to show
"functional" (ie connected to their remote servers). Until that time, any
other app that is started (Firefox, Excel, etc) is very slow to start -
sometimes 1-2 minutes to display the window and its contents (web page,
spreadsheet, etc). CPU usage is not excessive (averaging around 50% on each
of the four CPU cores, with fairly short peaks on some cores to 100%) but
disk activity is almost continuous.

OK, I could copy the system disk on an SSD and boot from that. But it's
easier just to put the PC to sleep each night: 2.5 W is a lot less power
than about 80 W (PC and monitor). Even starting from sleep, I occasionally
notice that response is a bit sluggish which suggests that the delay is due
to re-establishing network connections rather than just reloading the saved
memory image from disk.

When I used to use that PC for logging weather station data, the software
logged a new data point every 10 minutes. It was a race to see if I could I
could reboot the PC, starting immediately after a point had been logged,
before the next one was due 10 minutes later.

No doubt if I reinstalled everything from scratch and then customised it as
before, I could improve the boot time. But that would be a major exercise,
making sure everything was installed and configured as it was before.

On Mon, 18 Jul 2022 11:55:02 +0100, "NY" <me@privacy.invalid> wrote:

>"David E. Ross" <nobody@notme.invalid> wrote in message
>news:taqnm7$11v0$1@gioia.aioe.org...
>> On 7/14/2022 7:15 PM, pyotr filipivich wrote:
>>>
>>> The problem is, the computer will "go to sleep" but then wake up
>>> because the USB Hub wants a drink of water, or someone to tell it a
>>> story, or because there's a monster under the bed. What that means in
>>> turn is that when the computer monitors "click on" in the middle of
>>> the night, they wake me up.
>>>
>>> What I want to know is how do I make the computer actually stay asleep
>>> till morning or I wake it up?
>>>
>>> The work around I've found is to have a Custom Power Management set to
>>> shut off monitors after 5 minutes and "sleep" to 10.
>>>
>>> Related Question: How many of the 5 "HID Generic Keyboards" do I
>>> really need?
>>
>> Do you really need the computer to be instantly awake when you get up in
>> the morning? Why not merely turn it off just before you go to sleep?
>
>It depends how long it takes the PC to boot up and for all the background
>processes to become fully functional.
>
>My Windows 7 PC takes a shade under 10 minutes until everything is running:

That's an abnormally long time. Undoubtedly you have a problem that
could be fixed.

But when I get up in the morning, I power on the computer and go into
the kitchen to get my coffee. When I come back, it's finished booting.
I don't know how long it took and I don't care.

"NY" <me@privacy.invalid> on Mon, 18 Jul 2022 11:55:02 +0100 typed in
alt.windows7.general the following:
>"David E. Ross" <nobody@notme.invalid> wrote in message
>news:taqnm7$11v0$1@gioia.aioe.org...
>> On 7/14/2022 7:15 PM, pyotr filipivich wrote:
>>>
>>> The problem is, the computer will "go to sleep" but then wake up
>>> because the USB Hub wants a drink of water, or someone to tell it a
>>> story, or because there's a monster under the bed. What that means in
>>> turn is that when the computer monitors "click on" in the middle of
>>> the night, they wake me up.
>>>
>>> What I want to know is how do I make the computer actually stay asleep
>>> till morning or I wake it up?
>>>
>>> The work around I've found is to have a Custom Power Management set to
>>> shut off monitors after 5 minutes and "sleep" to 10.
>>>
>>> Related Question: How many of the 5 "HID Generic Keyboards" do I
>>> really need?
>>
>> Do you really need the computer to be instantly awake when you get up in
>> the morning? Why not merely turn it off just before you go to sleep?
>
>It depends how long it takes the PC to boot up and for all the background
>processes to become fully functional.
>
>My Windows 7 PC takes a shade under 10 minutes until everything is running:
>Skype and Dropbox icons in the system tray take ages to appear and to show
>"functional" (ie connected to their remote servers). Until that time, any
>other app that is started (Firefox, Excel, etc) is very slow to start -
>sometimes 1-2 minutes to display the window and its contents (web page,
>spreadsheet, etc). CPU usage is not excessive (averaging around 50% on each
>of the four CPU cores, with fairly short peaks on some cores to 100%) but
>disk activity is almost continuous.

Exactly. I used to get up, turn on the computer, feed the cat,
start the espresso machine, do my morning prayers, fix my morning
espresso, and the computer would be ready to go when I got back.

The delay would not be that much of an issue, save that restarting
the computer requires certain "default" configurations to be reset:
e.g., the blinking back lights on the keyboard.
--
pyotr filipivich
This Week's Panel: Us & Them - Eliminating Them.
Next Month's Panel: Having eliminated the old Them(tm)
Selecting who insufficiently Woke(tm) as to serve as the new Them(tm)

Ken Blake <Ken@invalid.news.com> on Mon, 18 Jul 2022 07:19:46 -0700
typed in alt.windows7.general the following:
>On Mon, 18 Jul 2022 11:55:02 +0100, "NY" <me@privacy.invalid> wrote:
>
>>"David E. Ross" <nobody@notme.invalid> wrote in message
>>news:taqnm7$11v0$1@gioia.aioe.org...
>>> On 7/14/2022 7:15 PM, pyotr filipivich wrote:
>>>>
>>>> The problem is, the computer will "go to sleep" but then wake up
>>>> because the USB Hub wants a drink of water, or someone to tell it a
>>>> story, or because there's a monster under the bed. What that means in
>>>> turn is that when the computer monitors "click on" in the middle of
>>>> the night, they wake me up.
>>>>
>>>> What I want to know is how do I make the computer actually stay asleep
>>>> till morning or I wake it up?
>>>>
>>>> The work around I've found is to have a Custom Power Management set to
>>>> shut off monitors after 5 minutes and "sleep" to 10.
>>>>
>>>> Related Question: How many of the 5 "HID Generic Keyboards" do I
>>>> really need?
>>>
>>> Do you really need the computer to be instantly awake when you get up in
>>> the morning? Why not merely turn it off just before you go to sleep?
>>
>>It depends how long it takes the PC to boot up and for all the background
>>processes to become fully functional.
>>
>>My Windows 7 PC takes a shade under 10 minutes until everything is running:
>
>That's an abnormally long time. Undoubtedly you have a problem that
>could be fixed.
>
>But when I get up in the morning, I power on the computer and go into
>the kitchen to get my coffee. When I come back, it's finished booting.
>I don't know how long it took and I don't care.

I don't either. However, the computer in question is my wife's.
She is not "techn0-friendly" and ... I do what I can to make it easier
for her.
--
pyotr filipivich
This Week's Panel: Us & Them - Eliminating Them.
Next Month's Panel: Having eliminated the old Them(tm)
Selecting who insufficiently Woke(tm) as to serve as the new Them(tm)

On 7/18/2022 6:55 AM, NY wrote:
> "David E. Ross" <nobody@notme.invalid> wrote in message news:taqnm7$11v0$1@gioia.aioe.org...
>> On 7/14/2022 7:15 PM, pyotr filipivich wrote:
>>>
>>> The problem is, the computer will "go to sleep" but then wake up
>>> because the USB Hub wants a drink of water, or someone to tell it a
>>> story, or because there's a monster under the bed.  What that means in
>>> turn is that when the computer monitors "click on" in the middle of
>>> the night, they wake me up.
>>>
>>> What I want to know is how do I make the computer actually stay asleep
>>> till morning or I wake it up?
>>>
>>> The work around I've found is to have a Custom Power Management set to
>>> shut off monitors after 5 minutes and "sleep" to 10.
>>>
>>> Related Question: How many of the 5 "HID Generic Keyboards" do I
>>> really need?
>>
>> Do you really need the computer to be instantly awake when you get up in
>> the morning?  Why not merely turn it off just before you go to sleep?
>
> It depends how long it takes the PC to boot up and for all the background processes to become fully functional.
>
> My Windows 7 PC takes a shade under 10 minutes until everything is running: Skype and Dropbox icons in the system tray take ages to appear and to show "functional" (ie connected to their remote servers). Until that time, any other app that is started (Firefox, Excel, etc) is very slow to start - sometimes 1-2 minutes to display the window and its contents (web page, spreadsheet, etc). CPU usage is not excessive (averaging around 50% on each of the four CPU cores, with fairly short peaks on some cores to 100%) but disk activity is almost continuous.
>
> OK, I could copy the system disk on an SSD and boot from that. But it's easier just to put the PC to sleep each night: 2.5 W is a lot less power than about 80 W (PC and monitor). Even starting from sleep, I occasionally notice that response is a bit sluggish which suggests that the delay is due to re-establishing network connections rather than just reloading the saved memory image from disk.
>
> When I used to use that PC for logging weather station data, the software logged a new data point every 10 minutes. It was a race to see if I could I could reboot the PC, starting immediately after a point had been logged, before the next one was due 10 minutes later.
>
> No doubt if I reinstalled everything from scratch and then customised it as before, I could improve the boot time. But that would be a major exercise, making sure everything was installed and configured as it was before.

The machine always has a story.

https://support.microsoft.com/en-us/topic/the-windows-7-startup-process-is-slow-when-you-create-many-restore-points-3cfbf9c4-d461-8e76-7cc2-2686aa041d6b

You can clear all the Restore Points, using sysdm.cpl and the protection tab.

https://superuser.com/questions/369660/is-there-a-way-to-reset-superfetch

Run "net stop fileinfo"

delete AgGlFaultHistory.db 166KB
AgGlFgAppHistory.db 149KB
AgGlGlobalHistory.db 1205KB
(all under C:\Windows\Prefetch) 149 files 50.8MB

Run "SC Start SysMain" will get it going again.

I find on my W7 machine, that Seamonkey is slow to start, after boot.

Flushing the RAM with testlimit64 ( https://live.sysinternals.com/ ) seems to help.
The only thing I can think of, is some interaction with SysMain/SuperFetch (or similar).

1) Boot machine.
2) Run testlimit64 two or three times, for the fun of it.

testlimit64.exe -d 1 -c 8500 # Use up all the "free" RAM

3) Now, start Seamonkey and see how many seconds.

Cleaning the multi-purpose RAM should not really be
making this difference. Some "Microsoft Optimization" is doing this.

All that using an SSD does in cases like this, is obfuscate what was
doing it originally. Which is fine, if you have money for the drives.

Paul