NF7-S: How to stop waking up from hibernate?
*Vanguard*
I put the machine into hibernate mode but it will wakeup. All BIOS
settings for Wakeup and timer are disabled. The WakeOnLan property for
the onboard NIC is disabled. The machine goes into hibernate mode okay
but the next morning I find it has powered on. It is on a 2kVA
sinusoidal UPS so it is not a power spike turning it on, plus it only
turns on by itself it was put into hibernate mode, not if the machine
was shutdown (and turned off). I have gone through Device Manager
looking at the properties of devices to check their Advanced Power
property, if it has that panel, to ensure that none are configured to
resume out of standby mode (but that would actually mean there was still
power but in low-power mode).
I have the system password enabled in BIOS which requires entering a
password to continue the boot process so I find the machine in the
morning sitting at the BIOS prompt to enter a password. So what else
might be triggering the machine to power back up but only when it was
powered down for hibernate mode?
--
____________________________________________________________
*** Post replies to newsgroup. Share with others.
*** Email domain = ".com" *AND* append "=NEWS=" to Subject.
____________________________________________________________
Iain
> Abit NF7-S v2.0 motherboard
> I put the machine into hibernate mode but it will wakeup. All BIOS
> settings for Wakeup and timer are disabled. The WakeOnLan property for
> the onboard NIC is disabled. The machine goes into hibernate mode okay
> but the next morning I find it has powered on. It is on a 2kVA
> sinusoidal UPS so it is not a power spike turning it on, plus it only
> turns on by itself it was put into hibernate mode, not if the machine
> was shutdown (and turned off). I have gone through Device Manager
> looking at the properties of devices to check their Advanced Power
> property, if it has that panel, to ensure that none are configured to
> resume out of standby mode (but that would actually mean there was still
> power but in low-power mode).
> I have the system password enabled in BIOS which requires entering a
> password to continue the boot process so I find the machine in the
> morning sitting at the BIOS prompt to enter a password. So what else
> might be triggering the machine to power back up but only when it was
> powered down for hibernate mode?
You could be looking for something (anything) timed; ACPI allows the OS
to do stuff like "suspend till 4am"... In no particular order:
* waking up to renew a DHCP lease, or otherwise refresh some network
information; disable the network connection then hibernate to test.
* task scheduler waking up to do some scheduled task; stop the task
scheduler and at services to test.
* disk indexing; stop the Indexing Service, if you can find it to test.
* auto-update, either windows (disable the background intelligent
transfer service, and automatic updates services to test), or something
else; if you have automatic updates enabled for your AV or firewall
software, stop or temp. disable them each in turn till you find the culprit.
Of course, AFAIK, ACPI allows the OS to override selections made in the
BIOS setup tool as to what will cause a wakeup event; it may be worth
disabling devices like serial ports/internal modems/ USB controllers
etc. (actually, if you have any USB devices connected, they could be
sending wakeup events, or the controller could be triggering a wakeup
event based on a port-state-change event; if you have devices that
disconnect from the USB bus for any reason, that might do it.(hint: a
USB hub not connected to the UPS will generate a new connection event
when the power comes back). Try going into hibernate and disconnecting
USB devices...)
Check if it'll resume from hibernate on keypress/mouse action. If you
have a cordless mouse/keyboard, be aware that some may send events
indicating battery/link status, test with conventional corded devices.
Some ethercards will wake-on-link-status-change; normally this is
configured in the EEPROM tool, or the advanced properties of the adaptor
in question. Go hibernate, and disconnect/reconnect the ethernet cable
to test (or power-cycle whatever's at the other end).
ACPI does actually provide a couple things to potentially help; there's
a setting specifying what devices/events will cause a wakeup, and then
there's another specifying what caused the last wakeup event... Not that
there's any guarantee this will be either helpful or accurate, or
there's any way to get at it...
The last think I can think of, is checking for BIOS updates; the BIOS
gets to parse the ACPI control flags, and program the power management
component of the southbridge accordingly...
Good luck tracking this down... You'll need it...
--
Regards, Iain.
*Vanguard*
Thanks for the suggestions but most only apply if the computer goes into
Standby mode (which still has power and still has the OS loaded in
memory). Hibernate mode copies the memory to a file (hiberfil.sys) and
powers down the computer completely (so there is nothing in memory
anymore).
> You could be looking for something (anything) timed; ACPI allows the
> OS
> to do stuff like "suspend till 4am"... In no particular order:
> * waking up to renew a DHCP lease, or otherwise refresh some network
> information; disable the network connection then hibernate to test.
There is no network connection in hibernate mode because there is no
power. You're thinking of standby mode.
> * task scheduler waking up to do some scheduled task; stop the task
> scheduler and at services to test.
Task scheduler isn't running because the OS isn't running because I am
NOT in standby mode.
> * disk indexing; stop the Indexing Service, if you can find it to
> test.
Again only relevant in standby mode, not when powered off completely in
hibernate mode.
> * auto-update, either windows (disable the background intelligent
> transfer service, and automatic updates services to test), or
> something else; if you have automatic updates enabled for your AV or
> firewall software, stop or temp. disable them each in turn till you
> find the culprit.
Only applicable in standby mode, not when powered off in hibernate mode.
Hibernate mode copies all the memory into a file and completely powers
off the computer. There are NO programs, processes, threads, or
anything else running on the computer when in hibernate mode. Hibernate
mode is no different than shutting down the OS and powering off your
computer (for its hardware state). Hibernate just saves the memory into
a file and triggers on reloading that memory image on power up.
Of course, AFAIK, ACPI allows the OS to override
> selections made in the BIOS setup tool as to what will cause a wakeup
> event; it may be worth disabling devices like serial ports/internal
> modems/ USB controllers
I have looked in the BIOS and checked that all wakeup events are
disabled. The timer (used to wakeup the system) is also disabled.
Although the OS is not running when the computer is powered off (for
hibernate mode), some hardware settings can be configured from within
the OS. For example, the onboard NIC has a WakeOnLAN setting that will
affect the hardware setup so let the NIC wakeup the system. However,
that is disabled. I have gone through all devices listed in Device
Manager and checked that they have no Advanced power configuration
settings to wakeup on an event.
> etc. (actually, if you have any USB devices connected, they could be
> sending wakeup events, or the controller could be triggering a wakeup
> event based on a port-state-change event; if you have devices that
> disconnect from the USB bus for any reason, that might do it.(hint: a
> USB hub not connected to the UPS will generate a new connection event
> when the power comes back). Try going into hibernate and disconnecting
> USB devices...)
The only USB "function" (i.e., device) is a printer. No USB hubs are
employed. Nothing USB is getting connected or disconnected during the
time when the computer is powered down (as the result of using
hibernate). I power down, go to sleep, and when I wake up is when I
find the computer powered up at some time during the night. The printer
is an inkjet printer and doesn't consume much power even when printing
so it is also connected to the same 2kVA UPS to which the computer is
connected.
> Check if it'll resume from hibernate on keypress/mouse action. If you
> have a cordless mouse/keyboard, be aware that some may send events
> indicating battery/link status, test with conventional corded devices.
At one time, I did have a cordless mouse (but a corded keyboard). I had
the problem then of the computer powering back up after powered down
when using hibernate. However, I later went back to a tethered mouse
because the extra weight of the batteries in the cordless mouse were
overly fatiguing my pinkie that has to squeeze the mouse when repeatedly
having to lift and move it. I also tired of replacing the batteries
every 2 to 3 weeks (far shorter than the manufacturer's claim although I
had them replace the cordless mouse three times and tried two other
brands). This power-up-after-hibernate-power-down problem has persisted
when changing from a cordless to a corded mouse.
> Some ethercards will wake-on-link-status-change; normally this is
> configured in the EEPROM tool, or the advanced properties of the
> adaptor
> in question. Go hibernate, and disconnect/reconnect the ethernet cable
> to test (or power-cycle whatever's at the other end).
There is WakeUp for the NIC in BIOS. I did disable the WakeUpLan
property for the onboard NIC in Windows' Device Manager. However, on
the next test of hibernate, I'll just yank the CAT5 cable, go to sleep,
and check if the computer turned on or not during the night.
> ACPI does actually provide a couple things to potentially help;
> there's
> a setting specifying what devices/events will cause a wakeup, and then
> there's another specifying what caused the last wakeup event... Not
> that there's any guarantee this will be either helpful or accurate, or
> there's any way to get at it...
Those are all disabled in the BIOS, along with the timer (RTC alarm)
being disabled which can wakeup you system at a preset time.
> The last think I can think of, is checking for BIOS updates; the BIOS
> gets to parse the ACPI control flags, and program the power management
> component of the southbridge accordingly...
This problem has happened under BIOS revisions 17, 19, and 23 (I didn't
use revision 20 long enough to check this problem). Revision 24 is the
latest available but I haven't had the time to install it yet. However,
if this problem has existed across 3 revisions, and because the
description of the changes for revision 24 deal with a wheeled PS/2
mouse not being responsive on odd-numbered Windows restarts and with USB
keyboards, it doesn't look like the latest revision will fix or even
touches upon this problem.
> Good luck tracking this down... You'll need it...
Definitely. Thanks for the suggestions, though.
TomG
--
Thomas Geery
Network+ certified
ftp://geerynet.d2g.com
ftp://68.98.180.8 Abit Mirror <----- Cable modem IP
This IP is dynamic so it *could* change!...
over 130,000 FTP users served!
^^^^^^^
"*Vanguard*" <reply-to-...@to-email.use-Reply.obey-signature.invalid>
wrote in message news:gmCzc.25492$Hg2.22554@attbi_s04...
*Vanguard*
> any Scheduled Tasks set to wake the system on the advanced properties
> page?
Scheduled tasks don't run when the computer is powered down because
there is no operating system running to be checking the timing and
events configured in Task Scheduler. The "Wake the computer to run this
task" option (under the Settings panel, not an Advanced panel) applies
only if the computer is still powered on in Standby mode.
Standby mode reduces power but leaves the power on. The CPU continues
to cycle away and the memory contents remain intact (and can get updated
to reflect changes by running processes). Hibernate copies the memory
to a file and powers off the system. There is no difference in the
state of your computer when powered off by a shutdown and when powered
off by a hibernate. The only difference exhibits itself when you power
back up in which case the file gets copied into memory (which is usually
faster) rather than having to reload and execute all the files to
startup Windows.
Description of the Different Advanced Power Management States
http://support.microsoft.com/?id=308535
For the Ready, Standby, and Suspend modes, the computer is still powered
on. For the Hibernate mode, the computer gets turned off completely.
Thanks for the suggestion, however. The Task Scheduler (for other
problems) sometimes gets missed as a possible source of the problems,
like when trying to find how spyware or malware is getting loaded.
Iain
> Iain said in news:40cefe4e$0$4592$db0f...@news.zen.co.uk:
> Thanks for the suggestions but most only apply if the computer goes into
> Standby mode (which still has power and still has the OS loaded in
> memory). Hibernate mode copies the memory to a file (hiberfil.sys) and
> powers down the computer completely (so there is nothing in memory
> anymore).
Not quite... Yes, the OS saves state to disk, and yes the OS isn't
running after it's finished the hibernate process. But ACPI allows stuff
like "turn the power off, but turn it back on in 15 mins". And "turn the
power off, but turn it back on if <some wake up event> happens, even if
the user has disabled that in the BIOS util".
Now, I'm almost positive that task scheduler offers that functionality
(see TomGs post), and I suspect quite strongly that other OS components
use this, although deciding what/how is tricky...
>>You could be looking for something (anything) timed; ACPI allows the
>>OS
>>to do stuff like "suspend till 4am"... In no particular order:
>>* waking up to renew a DHCP lease, or otherwise refresh some network
>>information; disable the network connection then hibernate to test.
> There is no network connection in hibernate mode because there is no
> power. You're thinking of standby mode.
There may be no working network connection, but if you have a DHCP
lease, that has an expiry date. Can you say catagorically that windows
won't schedule a resume-from-hibernate in order to refresh this?
>>* task scheduler waking up to do some scheduled task; stop the task
>>scheduler and at services to test.
> Task scheduler isn't running because the OS isn't running because I am
> NOT in standby mode.
Again, doesn't need to be; it'll just use the ACPI (and now I think of
it, ye olde APM also offered this) call to request a wake-up just before
the time of whatever the next task is. TomG pointed to somewhere to
specifically disable this.
>>* disk indexing; stop the Indexing Service, if you can find it to
>>test.
> Again only relevant in standby mode, not when powered off completely in
> hibernate mode.
>>* auto-update, either windows (disable the background intelligent
>>transfer service, and automatic updates services to test), or
>>something else; if you have automatic updates enabled for your AV or
>>firewall software, stop or temp. disable them each in turn till you
>>find the culprit.
> Only applicable in standby mode, not when powered off in hibernate mode.
> Hibernate mode copies all the memory into a file and completely powers
> off the computer. There are NO programs, processes, threads, or
> anything else running on the computer when in hibernate mode. Hibernate
> mode is no different than shutting down the OS and powering off your
> computer (for its hardware state). Hibernate just saves the memory into
> a file and triggers on reloading that memory image on power up.
As I said, you can't discount these things so easily. Unless you know
for a certain fact (i.e. have read the MS source code in question, or
have gone through it with an MS engineer or sommat in some depth, I
doubt the docs will be terribly useful here) that these components don't
use the ACPI scheduled power-up (which, BTW, overrides any setting you
have set in wake on timer in your BIOS util).
>>etc. (actually, if you have any USB devices connected, they could be
>>sending wakeup events, or the controller could be triggering a wakeup
>>event based on a port-state-change event; if you have devices that
>>disconnect from the USB bus for any reason, that might do it.(hint: a
>>USB hub not connected to the UPS will generate a new connection event
>>when the power comes back). Try going into hibernate and disconnecting
>>USB devices...)
> The only USB "function" (i.e., device) is a printer. No USB hubs are
> employed. Nothing USB is getting connected or disconnected during the
> time when the computer is powered down (as the result of using
> hibernate). I power down, go to sleep, and when I wake up is when I
> find the computer powered up at some time during the night. The printer
> is an inkjet printer and doesn't consume much power even when printing
> so it is also connected to the same 2kVA UPS to which the computer is
> connected.
Hmm. If it's a canon, it might do an auto-off trick, after a timeout.
You might just check this. Go hibernade, and disconnect/reconnect it,
and power-cycle it and see if anything happens. If nothing happens, it's
probably safe to discount that.
Glad you mentioned the UPS again; got the monitoring cable connected?
Could windows have asked for a wakeup on serial activity? Remember, this
may well happen even if you set to ignore COMx in the BIOS util (but it
can't hurt to check the BIOS util, normally under IRQ monitoring in the
power management section)...
>>Check if it'll resume from hibernate on keypress/mouse action. If you
>>have a cordless mouse/keyboard, be aware that some may send events
>>indicating battery/link status, test with conventional corded devices.
> At one time, I did have a cordless mouse (but a corded keyboard). I had
> the problem then of the computer powering back up after powered down
> when using hibernate. However, I later went back to a tethered mouse
> because the extra weight of the batteries in the cordless mouse were
> overly fatiguing my pinkie that has to squeeze the mouse when repeatedly
> having to lift and move it. I also tired of replacing the batteries
> every 2 to 3 weeks (far shorter than the manufacturer's claim although I
> had them replace the cordless mouse three times and tried two other
> brands). This power-up-after-hibernate-power-down problem has persisted
> when changing from a cordless to a corded mouse.
Purely as a side-note, some boards will do wake-on-mouse-move. While not
a particularly useful feature, it can't hurt to just make absolutely
certain it won't wake the thing from hibernate.
>>Some ethercards will wake-on-link-status-change; normally this is
>>configured in the EEPROM tool, or the advanced properties of the
>>adaptor
>>in question. Go hibernate, and disconnect/reconnect the ethernet cable
>>to test (or power-cycle whatever's at the other end).
> There is WakeUp for the NIC in BIOS. I did disable the WakeUpLan
> property for the onboard NIC in Windows' Device Manager. However, on
> the next test of hibernate, I'll just yank the CAT5 cable, go to sleep,
> and check if the computer turned on or not during the night.
Well, if you disconnect the cat5 after it's gone to hibernate, it's
probably safe to discount it...
>>ACPI does actually provide a couple things to potentially help;
>>there's
>>a setting specifying what devices/events will cause a wakeup, and then
>>there's another specifying what caused the last wakeup event... Not
>>that there's any guarantee this will be either helpful or accurate, or
>>there's any way to get at it...
> Those are all disabled in the BIOS, along with the timer (RTC alarm)
> being disabled which can wakeup you system at a preset time.
As I warned, these settings are ignored if the OS requests a wakeup...
(from what is it? S4 for hibernate, S5 for soft-off?)
> This problem has happened under BIOS revisions 17, 19, and 23 (I didn't
> use revision 20 long enough to check this problem). Revision 24 is the
> latest available but I haven't had the time to install it yet. However,
> if this problem has existed across 3 revisions, and because the
> description of the changes for revision 24 deal with a wheeled PS/2
> mouse not being responsive on odd-numbered Windows restarts and with USB
> keyboards, it doesn't look like the latest revision will fix or even
> touches upon this problem.
Fair enough.
>>Good luck tracking this down... You'll need it...
> Definitely. Thanks for the suggestions, though.
You'll also need a fairly open mind while attacking this; ACPI allows
all sorts of strange stuff to happen...
If you're despirate for a quick-fix, you could try searching the MS
knowledgebase for the directions on changing to the non-ACPI HAL. You
should be prepared for any of the following: massive drop in
performance, IRQ/PCI configuration problems, the need to reinstall
windows. http://support.microsoft.com/default.aspx?scid=kb;en-us;309283
makes it clear that you would have to reinstall XP to switch back to the
ACPI HAL. Really not recommended...
Have a read through
http://www.microsoft.com/whdc/archive/winpowmgmt.mspx . Note that S4 (==
Hibernate) isn't S5 (== Off).
Nearest docs in MSDN
http://msdn.microsoft.com/library/default.asp?url=/library/en-us/power/base/pbt_apmresumeautomatic.asp
documenting the Windows Message PBT_APMRESUMEAUTOMATIC which indicates
the system resumed to handle some event, but the user isn't present.
There's a load of stuff at http://www.acpi.info related to the various
tricks ACPI can do.
--
Regards, Iain.
Iain
> Standby mode reduces power but leaves the power on. The CPU continues
> to cycle away and the memory contents remain intact (and can get updated
> to reflect changes by running processes). Hibernate copies the memory
> to a file and powers off the system. There is no difference in the
> state of your computer when powered off by a shutdown and when powered
> off by a hibernate. The only difference exhibits itself when you power
> back up in which case the file gets copied into memory (which is usually
> faster) rather than having to reload and execute all the files to
> startup Windows.
> Description of the Different Advanced Power Management States
> http://support.microsoft.com/?id=308535
> For the Ready, Standby, and Suspend modes, the computer is still powered
> on. For the Hibernate mode, the computer gets turned off completely.
OK, please go to Device Manager. Look at the "Computer" or "PC" entry
(whatever it's called).
If it says "Standard PC", the rest of us are barking up the wrong tree.
If it says anything like "ACPI PC", "Advanced Configuration and Power
Interface UniProcessor PC", XP will use ACPI states to manage power, and
will NOT use APM (Advanced Power Management).
Under ACPI, SoftOff is system sleep state S5. Hibernate is S4. Under XP,
it happens to be OS-Controlled S4, but it's still S4, and you cannot use
previously learned information about how APM states work to predict how
ACPI states work.
There are the ACPI specs at www.acpi.info, but be warned, they're about
500 pages long, and in any case, are geared to developers writing an
ACPI-compliant BIOS, or an ACPI-compliant OS. Still, they're the
definitive reference when it comes to G-states, S-states, D-states, or
C-states... (probably more, I haven't exactly studied the documents
there too carefully...)
If you can get the system to transition into S3, then the CPU is
actually stopped and powered off. The RAM contents are maintained,
though. Technically, the OS stops running, and the only way to cause the
CPU to start again is by some thing causing a wakeup event, which will
cause a transition to S0 (probably) to happen. Same system will cause
transitions from S4 to happen.
One could probably argue, however, that the fact your BIOS asks for a
password at this point is actually a bug; the system isn't starting a
normal boot from S5/SoftOff, it's starting a transition from S4/STD to
S0. After all, it doesn't ask for a password when transitioning from
S3/STR to S0, does it?
--
Regards, Iain.
TomG
explanation anyway. it was just a thought. I have a laptop that DOES wake
itself from hibernate for a scheduled task, each and every day. maybe there
is some third party software involved but I didn't think so...
--
Thomas Geery
Network+ certified
ftp://geerynet.d2g.com
ftp://68.98.180.8 Abit Mirror <----- Cable modem IP
This IP is dynamic so it *could* change!...
over 130,000 FTP users served!
^^^^^^^
"*Vanguard*" <reply-to-...@to-email.use-Reply.obey-signature.invalid>
wrote in message news:uhIzc.110192$Ly.36747@attbi_s01...
*Vanguard*
Thanks guys. I'll try disabling the "Wake up computer" option on any
scheduled events. When I click the "?" titlebar button and then click
on this option, it says, "... even if the computer is in Sleep mode
...". So then I had to wonder what was "Sleep mode." Well, apparently
sleep mode is hibernate mode (ACPI S4 sleep state). I had incorrectly
associated sleep mode with standby mode, but then I didn't realize "Wake
the computer" even included a wake from sleep mode. When I read the
ABit manual, it says:
"Two options are available: S1 (PowerOn-Suspend) --> S3
(Suspend-to-RAM). The default setting is S1 (PowerOn-Suspend). POS is
"Power On Suspend", and STR is "Suspecnd To RAM". Generally, ACPI has
six states: System S0 state, S1 state, S2 state, S3 state, S4 state, S5
state. S1 and S3 are described below:"
So my understanding was that the NF7-S v2.0 motherboard only supported
the S1 and S3 states. Maybe it is just that S1 and S3 are the only ones
you can select within the BIOS but the other states are still usable.
I had figured the "Wake the computer" option for a scheduled task would
take it out of Standby mode but not power it up from a hibernate. I
will disable this option to see if the computer stays powered down.
However, when the computer is in Standby mode and with this option
disabled, doesn't that mean that scheduled tasks will not run?
Scott
Could something externally be causing the system to power up? IE: Cat, dog,
etc...
Just a thought....
Scott
"TomG" <tgeery-...@cox.net> wrote in message
news:9nJzc.134$cj3.57@lakeread01...
Bill Drake
page is locking the machine into S0 on a return-from-sleep condition.
It is perfectly normal (and the WXP default) for a return-from-sleep on
the monitor to prompt-for-password if you have the machine set that
way in the Screen Saver tab on the Display Preferences.
It is *also* perfectly normal (and the WXP default) for a return-from-sleep
on the Standby or Hibernate functions to prompt-for-password.
WXP is *supposed* to leave the screen blank and only bring back the
prompt-for-password screen when you wiggle the mouse or do some
other action which confirms user activity. However, there's that infernal
variation between theory and reality again -- so you need to confirm
what WXP is *actually* doing -- and ignore the theoretical machinations
of pie-in-the-sky rose-coloured-glasses marketdroids.
Things to check:
1. Temporarily disable your BIOS power-on password for testing
purposes.
1. Disable the password prompt for both the return-from-sleep for
the screen-saver and the return-from-sleep for the
suspend/hibernate condition. Observe what the machine does
when configured in this mode.
2. If the problem goes away, then one or the other of the
return-from-sleep conditions is triggering a password-prompt.
This is locking the machine into S0 while waiting for your input.
3. Selectively enable one or the other of the password prompts
for the return-from-sleep conditions. Confirm whether either or
both of the return-from-sleep conditions trigger the
password-prompt-lock-to-S0 condition.
4. If neither of the above changes solve your problem, then the
issue has nothing to do with the password-prompt-lock-to-S0
problem and that issue can be discarded as a possible source
of your problem.
5. At that point, you need to experiment by disconnecting each of
the peripherals that can trip the return-from-sleep condition to
isolate *which* of the peripherals is triggering the problem.
6. If none of the hardware is responsible for the false-return, then
you need to investigate the software for possible interactions
there.
Caveats, wrinkles and "gotchas":
1. Mouse drivers are a *known* contributor to return-from-sleep
issues. Please remember that mouse drivers interact with the
Video Card (both software drivers *and* hardware-cursor) -- so
there is a *lot* more going on behind-the-scenes with the mouse
driver than is immediately obvious.
2. It is quite common for older mouse drivers to be notoriously buggy
when it comes to power-management issues. If you are running
any version of the Logitech mouse drivers or keyboard drivers,
then IMO it is *mandatory* to upgrade to the latest-and-greatest.
If you are running both a Logitech Keyboard and Mouse set, then
IMO it is mandatory to upgrade *both* the keyboard and mouse
driver sets.
3. Microsoft mouse users with non-tilt-wheel mice should be running
the Microsoft IntelliPoint 4.12 driver set. This can be obtained by
going to the Microsoft Hardware website and downloading the
mouse drivers for Windows 98.
Note: The driver set that Microsoft offer when you select the
option for W98 is the *global* package for IntelliPoint 4.12.
This package also works for W2K and WXP.
Do *not* download and install the IntelliPoint 5 package
unless you are absolutely married to your tilt-wheel function.
There are serious deficiencies in the IntelliPoint 5 package
which are going to be addressed in the IntelliPoint 6 package
slated for release over the next few months.
4. Microsoft keyboard users should be running the Microsoft IntelliType
Pro Version 2.22 driver set. This can be obtained in the same
manner as the IntelliPoint 4.12 mouse driver set -- and again this
is a *global* package which also works for W2K and WXP.
Note: Do *not* download and install the IntelliType 5 package
unless you are married to a tilt-wheel Microsoft mouse.
The IntelliType Pro Version 2.22 driver set is the matching
keyboard driver to the IntelliPoint 4.12 mouse driver.
5. GART drivers are known to interact with ACPI power-saving logic.
And again, there are differences between theory and reality here
that mandate a thorough investigation of your chipset driver and
BIOS revision levels to confirm you are running the system with
the latest bugfixes in place.
6. In another post in this thread, you have confirmed a variation in
your machine having to do with mouse drivers for a removed
device. It is possible that if you can't properly remove all traces
of the old mouse driver version, the only way you're going to fix
your problem is to do a complete reinstall-from-scratch.
Best I can do for now. <tm>
Bill
*Vanguard*
NOT powered up this morning. So it looks promising that disabling the
"Wake the computer" option on my scheduled tasks was the cause. I'll
have to monitor my scheduled tasks to see if they continue to run when
the computer goes automatically into Standby after the configured 2-hour
interval. A couple more nights of checking hibernate mode will help to
confirm that the reason the computer was getting powered up was because
of the "Wake the computer" option. Thanks again for your help.