Changes in Sleep behavior

I'm in the habit of putting my desktop into Sleep mode when I'm not using
it. Its behavior has changed in the past couple of weeks and, naturally,
I cannot pin the change on anything I did. (Must be an update! ... )
Until recently, the machine would turn off the fans when it entered sleep
mode and the power-on led would blink slowly but there was nothing else
going on. Now, I still get the slow blink but (some of?) the fans remain
running. There is only one BIOS choice regarding power management; it is
simply enabled or not. If disabled, Windows does not offer Sleep as an
option. Hibernation is not enabled.

Any ideas?

On 9/24/2015 7:19 PM, Jason wrote:
I'm in the habit of putting my desktop into Sleep mode when I'm not using
it. Its behavior has changed in the past couple of weeks and, naturally,
I cannot pin the change on anything I did. (Must be an update! ... )
Until recently, the machine would turn off the fans when it entered sleep
mode and the power-on led would blink slowly but there was nothing else
going on. Now, I still get the slow blink but (some of?) the fans remain
running. There is only one BIOS choice regarding power management; it is
simply enabled or not. If disabled, Windows does not offer Sleep as an
option. Hibernation is not enabled.

Any ideas?

I agree it might have been caused by an update, but not specifically a
Microsoft update.

Reason I say that is that my system has an Asus motherboard that has an
Asus program specifically to control the fan speeds. The program can be
set by the use to respond in various ways to the system.

With that in mind you might want to verify if your motherboard
manufacturer offers a similar program which might have been updated
recently or just needs some tweaking to keep up with the changes
Microsoft might have made to the Windows OS.

---
This email has been checked for viruses by Avast antivirus software.
https://www.avast.com/antivirus

On 9/24/2015 11:00 PM, GlowingBlueMist wrote:
On 9/24/2015 7:19 PM, Jason wrote:
I'm in the habit of putting my desktop into Sleep mode when I'm not using
it. Its behavior has changed in the past couple of weeks and, naturally,
I cannot pin the change on anything I did. (Must be an update! ... )
Until recently, the machine would turn off the fans when it entered sleep
mode and the power-on led would blink slowly but there was nothing else
going on. Now, I still get the slow blink but (some of?) the fans remain
running. There is only one BIOS choice regarding power management; it is
simply enabled or not. If disabled, Windows does not offer Sleep as an
option. Hibernation is not enabled.

Any ideas?

I agree it might have been caused by an update, but not specifically a
Microsoft update.

Reason I say that is that my system has an Asus motherboard that has an
Asus program specifically to control the fan speeds. The program can be
set by the use to respond in various ways to the system.

With that in mind you might want to verify if your motherboard
manufacturer offers a similar program which might have been updated
recently or just needs some tweaking to keep up with the changes
Microsoft might have made to the Windows OS.

---
This email has been checked for viruses by Avast antivirus software.
https://www.avast.com/antivirus

Hopefully Avast's crappy advertisement should be gone from my system
now. I was not aware it was automatically configured by them to do this.

Jason wrote:
I'm in the habit of putting my desktop into Sleep mode when I'm not using
it. Its behavior has changed in the past couple of weeks and, naturally,
I cannot pin the change on anything I did. (Must be an update! ... )
Until recently, the machine would turn off the fans when it entered sleep
mode and the power-on led would blink slowly but there was nothing else
going on. Now, I still get the slow blink but (some of?) the fans remain
running. There is only one BIOS choice regarding power management; it is
simply enabled or not. If disabled, Windows does not offer Sleep as an
option. Hibernation is not enabled.

Any ideas?


So you've lost S3 and it is using S1 instead.

dumppo can give some information on sleep, but it is an
ancient utility, and doesn't work on OSes like Win10. It's
possible it still works on Win7.

It can do an administrative override on available sleep
states. But only if you "fix" whatever is blocking Sleep S3.

http://forums.pcper.com/showthread.p...31#post1825058

The tool "dumppo" is here. Note that some browsers cannot
connect to this. If you get stuck, try command line FTP to get it.

ftp://ftp.microsoft.com/products/Oem...cpi/dumppo.exe

In Command Prompt

cd /d C:\downloads\dumppo --- define a safe place for the download

ftp ftp.microsoft.com
username anonymous --- this server supports anonymous access
password --- the password takes the form of an
email address

cd products/Oemtest/v1.1/WOSTest/Tools/Acpi
binary
get dumppo.exe
quit

With tool in hand, use the pcper article if you need
to restrict or examine ACPI states.

There used to be a tutorial page on a French site,
which claimed that dumppo would even tell you
what driver was blocking a sleep state. As a driver
may know it is not compatible with particular S states.
However, I've never seen this information in practice,
ever. It never seems to tell you about the root cause.

So I can only recommend the tool, for administrative
override. Without override capability, if the OS decides
to block a particular S state due to a bad BIOS setting,
you can't get the OS to "switch back" to normal later.

And even elevated, dumppo doesn't seem to work in Win10.
So while the utility remained compatible for a long time,
the fun has finally ended. But you can try it in Win7
and see what it says.

The ACPI information (which sleep states to accept and so on),
may be stored in as many as 200 registry entries. Which is why
we need tools like this to mess around with, rather than
modifying the registry by hand.

Maybe GPEDIT has some options for this, but I've never bothered
to look for the issue there.

Paul

On Fri, 25 Sep 2015 12:10:33 +0000 "Stormin' Norman"
wrote in article




1. Have you power cycled the system, power off for at least 30 seconds?
Yes. That's my first line of attack for most problems...recalcitrant
routers included! Didn't make a difference
I've noticed over the years that a system restart, versus power off/on,
can give different results; I believe a restart does NOT reset (some?)
devices the way a power cycle does.


2. Have you cleaned the interior of the system thoroughly with compressed air to remove all dust from the
mainboard, power supply, hard drives and system memory?
I did that recently when I installed an SSD. The power behavior did NOT
change when I did that - the change is more recent.

3. Have you reset the bios of the system? Pull the battery and jump or short the CMOS reset contacts. After
doing this, be sure to enter the bios and reconfigure the system per your desires.
Yes. When I added the SSD I replaced the battery.

On Thu, 24 Sep 2015 20:19:44 -0400 "Jason" wrote
in article

Any ideas?

Thanks Paul et al! dumppo disclosed that S3 was not available, only S1.
Changing that to enable S3 did the trick.

Jason

On Fri, 25 Sep 2015 11:06:21 -0400 "Jason" wrote

Thanks Paul et al! dumppo disclosed that S3 was not available, only S1.
Changing that to enable S3 did the trick.

Pauk's remark that there are probably GP settings got me hunting around
and, indeed, Computer Configuration/Administratitve
Templates/System/Power Management reveals them.

Stormin' Norman wrote:
On Fri, 25 Sep 2015 11:06:21 -0400, Jason wrote:

On Thu, 24 Sep 2015 20:19:44 -0400 "Jason" wrote
in article
Any ideas?
Thanks Paul et al! dumppo disclosed that S3 was not available, only S1.
Changing that to enable S3 did the trick.

Jason


I wonder what caused this policy override, all of a sudden?

Typically it's caused by a hardware event, that prevents
the state from being entered.

In "retaliation" the OS removes the S state from the
list of available/usable S states.

If the user can then restore/correct the problem,
the OS will refuse to put things back the way they
were. So while the OS has a "defense" against hardware
that prevents a state from being entered, it has no
proactive policy to notice everything is working
again, and put it back. That's where dumppo comes in.

Typically, dumppo is used after a user enters the BIOS and
changes the S1 versus S1/S3 setting. Sometimes, it gets flipped
by a BIOS battery failure or replacement.

While I'm told a bad driver can cause S3 problems, that's
typically more of an Apple thing than a Windows thing.
Windows is supposed to have the same potential problem, but
I just don't run into it as a root cause.

Paul

On Fri, 25 Sep 2015 14:19:10 -0400 "Paul" wrote in
article
Typically, dumppo is used after a user enters the BIOS and
changes the S1 versus S1/S3 setting. Sometimes, it gets flipped
by a BIOS battery failure or replacement.


The BIOS setting for the machine in question has only two choices re
sleep behavior: permitted or not. There's no finer distinction to be
made.

In article ,
lid says...
I wonder what caused this policy override, all of a sudden?

I sure don't know. I joked earlier about a [damned] update and another
poster suggested it might not necessarily be a Win update, that it could
be a driver. I did visit Dell recently and a couple drivers were flagged
as out of date so I replaced them. I don't recall if that's when things
got weird. It's hard to believe that such an update would make a change
like that.

On Sat, 26 Sep 2015 12:10:28 +0000 "Stormin' Norman"
wrote in article

Was one an update for the BIOS?

No. There have been no updates for this machine for a year or two.

In article ,
says...

In article ,
lid says...
I wonder what caused this policy override, all of a sudden?

I sure don't know.
....
The system sometimes reverts to the fan-on problem that started this
thread. Today when I started it S3 was among the missing again. I
repaired that with dumppo and it was ok again. What I'd read about
dumppo said a re-boot was necessary after a change. I've found that not
to be the case.

Jason wrote:
In article ,
says...
In article ,
lid says...
I wonder what caused this policy override, all of a sudden?

I sure don't know.
...
The system sometimes reverts to the fan-on problem that started this
thread. Today when I started it S3 was among the missing again. I
repaired that with dumppo and it was ok again. What I'd read about
dumppo said a re-boot was necessary after a change. I've found that not
to be the case.

The fan-on problem probably means it crashed at BIOS level.
If a CPU crashes, there is nothing to program the fan
control in an intelligent fashion. The fan control is
designed to run at 100%, until loaded with proper settings.

Depending on the motherboard maker, one of the recovery
techniques is to reset the BIOS settings (on the assumption
the crash is caused by a bad setting). And if the BIOS
default values after clearing CMOS aren't "good" ones, you're
forever going to be cleaning up after it.

In addition to a BIOS crash, doing a forced power off
is also interpreted as a BIOS crash. Since the BIOS doesn't
get to do a controlled shutdown.

Different brands have different practices. On Asus, if it
crashes in the BIOS, the settings are reset. Only on modern
boards, is a subset of settings reset (clock related only). Leaving
other innocent settings alone. Whereas older BIOS designs, everything
is set back to default settings (like turning off the Promise
Controller chip on my P4C800-E grrr). On an Asrock, pushing the reset button
three times in a row (with a couple seconds delay between each press),
causes the BIOS settings to return to defaults. Which means a persistent
crash event is needed to eventually cause the BIOS to reset the CMOS
values.

If it were not for options like this, a user would use the
CLEAR_CMOS header to achieve the same results. That's what you'd
do on other brands, to recover from a BIOS crashing problem (assuming
the crash is not caused by a hardware failure like bad RAM).

Paul

On Sun, 27 Sep 2015 21:16:25 +0000 "Stormin' Norman"
wrote in article
Earlier you said you replaced the battery when you installed an SSD. Have you recently tried resetting the
bios by shorting out the CMOS reset contacts on the mainboard?

I did that when I changed the battery.

You could also reapply the most recent BIOS from Dell.
That's easy and worth a try

Lastly, you could create a startup / logon script which uses Dumppo to re-enable S3 each time you boot the
machine. DUMPPO.EXE admin /ac minsleep=s3
I have written one already! I hate to do things to mask problems, but not
always


Oh, did this problem start at the same point in time you installed the SSD?
Nope. SSD went in 3 months ago. This behavior just began in the past
week.

Thanks.