Monitor sleep issue

[Default] On Thu, 28 Jan 2016 17:40:22 -0500, in
microsoft.public.windowsxp.general Paul wrote:

wrote:
On Thu, 28 Jan 2016 14:03:33 -0500, Paul wrote:

wrote:

The problem is it won't stay asleep but it does not completely wake up
when it is sleeping., It just flashes and goes right back to sleep.

If I do something to wake it up it comes up OK.

This is just frustrating when you are in the room and it winks at you.
I have not actually timed it but it might wink every 15 minutes (the
time out setting) Maybe the timer is still running, forgets the
monitor is already asleep and wakes it up, just to put it back to
sleep.
OK, then you should

1) Open Device Manager.
2) Locate the NIC.
3) Find the property that allows the NIC to
wake the computer. Untick the two boxes.
You want the NIC to stay powered at all times
when running the computer, as well as wanting
the NIC to *not* wake the computer when sleeping.
4) On an Intel NIC, there are two properties with
at least four Wake On LAN possibilities. One is
effectively "Wake On Carrier", and if you switch
your router off and on, the computer might wake up.
It's possible to leave WOL enabled, and refine the
conditions under which it will awaken. You can have
it wake on a "Magic Packet" and have it ignore the
carrier state.

The NIC, mouse, or keyboard can wake a sleeping computer.
For peace of mind, you can disable NIC and mouse waking,
and leave keyboard enabled. Keyboard is plenty of
functionality for purposeful waking.

I have a new Microsoft mouse, the optical sensor is awake
while the computer sleeps, and if I walk across the
floor, the mouse sensor will wake the computer. My
previous Logitech mouse isn't nearly that stupid.
On the Logitech, only the mouse buttons trigger
wake behavior, and the sensor does not run in sleep.
The new Microsoft mouse on the other hand, flashes
the sensor at regular intervals, as if it is trying
to save power for me or something. Just disabling
the mouse tick boxes, fixed that for me and
turned off the blue LED inside the mouse.

If a computer is in the process of entering the sleep
state, and a driver file in the process of shutting
down crashes, the computer will immediately "reappear"
on you. Your symptoms don't match such a shutdown-type
issue. But maybe, with a little luck, hammering a
couple things in Device Manager will fix it for you.

Paul

The problem with that theory is once awaken by the software, it would
stay awake for 15 minutes. This just looks like the adapter flashed a
video signal to the monitor and immediately went away.

Windows 7 might have a few more options for debugging.

All I see on my WinXP, is something like this (admin command prompt)

powercfg /devicequery wake_armed

Hey that's cool!

And it (properly) says only my keyboard can wake
the computer, because the "pesky" wake sources
have been disabled.

Does it actually list things that are disabled?
This is a Dell and I don't think it has Wake on Lan or Wake on Ring,
so maybe there's no way to list what couldn't be enabled anyhow.


Standard 101/102-Key...

It lists my keyboard and my mouse. I want my mouse to work because
iirc keys like cntl and alt and shift never worked, and I'm afraid to
type letters or numbers, for fear I won't find and remove them.

So I like it to work by moving my mouse, and if necessary, the right
mouse key. It seems to take both these days.

What other things might wake a computer?

My search engines couldn't find a Power-Troubleshooter
from MATS (that's the web page that had troubleshooters
for WinXP era).

HTH,
Paul

Micky wrote:

microsoft.public.windowsxp.general Paul wrote:

All I see on my WinXP, is something like this (admin command prompt)

powercfg /devicequery wake_armed

Hey that's cool!

Does it actually list things that are disabled?
This is a Dell and I don't think it has Wake on Lan or Wake on Ring,
so maybe there's no way to list what couldn't be enabled anyhow.

In Device Manager, do "Properties" on each interesting
device, and see how it is set up.

The BIOS has a section with some items like "wake on PME"
(Power Management Event). Enabling that might allow a NIC
plugged into a slot to do WOL. Many pieces of hardware
are logically wired-OR to PME.

The BIOS sometimes has a field where you can specify the
wake condition for a PS/2 keyboard. Some SuperI/O chips
support a "pattern" for waking, like pressing control-shift
or something. Many others look for any-old-keypress and
there is no user choice.

So for starters, you don't need a lot of fancy tools and
hours or research.

If you have a 15 year old NIC in a PCI slot, some
of those pre-date PME on the PCI bus, and a special three
wire cable goes from the NIC to a header on the motherboard.
So that's a consideration on that Celeron 300 you own.

On the mouse, with USB, all you can really be assured of,
is that everything wakes it. (A gross PME event.) I
don't know the details of refining the mouse response
so the LED on the optical mouse is turned off, versus
usage of the mouse buttons. I generally just tap the
keyboard for this sort of thing, and don't want to mess
with mice for waking. They're more trouble than they're
worth.

And for USB, eight year or older motherboards, can
have a USBPWR header, for selecting the power source
for each pair of USB ports. You can designate just
two ports on the back of the desktop, to be "armed"
for waking. If you use the USBPWR header jumper, and
select +5VSB, then those ports receive power while
the computer sleeps. If you move the USBPWR jumper to
select +5V, then the ports are only powered during
normal operation. (Gating the power in that way,
overrides both Device Manager and BIOS choices, neutering
them.) On a 2016 computer, all the ports run off +5VSB
hardwired, and there is no USBPWR header scheme to allow
user control. If the computer is modern enough, everything
will have the power it needs for waking, and you won't be
able to gate the power to stop stuff from happening. Then
you have to resort to the BIOS choices and Device Manager,
for "peace in the valley".

Paul