New Box no sleep

Drew wrote:
On 8/20/2017 4:22 AM, VanguardLH wrote:
Paul wrote:

Char Jackson wrote:
On Sat, 19 Aug 2017 17:30:50 -0600, Drew
wrote:

New computer not sleeping?, running win 10 64 pro. When I set it to
sleep, The monitor shuts off, the keyboard and mouse do the same.
Computer itself (desktop) stays on and fans are turning. Router also
shows that the pc is sleeping After waiting a few minutes for the
pc to
go dark and the fans to stop (it does not) I have tried to move or
click
the mouse and or keyboard and nothing happens without a hard restart.
Cannot seem to figure out why. I have tried for hrs digging through
different settings and to no avail. I have tried power cfg -h on
and -h
off and still. Tried power options, usb settings and cannot seem to
find
what is needed. Any help would be appreciated greatly.

Specs are
Asus Rog Strix 270g motherboard
Intel core i7 7700k
32gigs ram
Nvidia turbo 1070 graphics card

Is this a homebuilt computer? Are the fans plugged in where they should
be? That looks like a gaming motherboard. Is there anything in the BIOS
that affects sleep and/or fans?


There was a time when "S1, S3, S1&S3" was a popular thing to
adjust in the legacy BIOS. If that happened to be set incorrectly,
a person would need to use "dumppo" to do an "override".

Dumppo no longer works in Win10. Which means you won't have
to learn how to use it.

I noticed powercfg has an "override" option, but I haven't
checked out what it can override.

And I don't think that setting is in a UEFI BIOS any more.
As the OS provides all the adjustments you could ever need.
Both Linux or Windows can hammer the settings needed. So
we should gradually wave goodbye to dumppo, in the
rear view mirror.

Paul

Since the hardware is responsible for regulating CPU temperature
*before* any OS can be loaded, and especially if no OS is loaded (drive
could be missing, dead, unreachable or no OS yet installed), there must
be controls in the BIOS/UEFI to have it control fan speed based on
temperature sensor(s).

http://dlcdnet.asus.com/pub/ASUS/mb/...B_20170406.pdf


Notice it has fan control. While the OS can issue commands to the BIOS
to regulate fan speed, the BIOS gets to override any such requests. If
the BIOS is configured to always run the fans at full speed then it
doesn't matter what the OS requests.

The manual also mentions differentiation between PWM (pulse width
modulation) and DC mode for controlling fan speed. PWM is used for
4-pin CPU fans and DC (voltage) for 3-pin CPU fans. With just 3 pins
(12VDC, ground, RPM), there is no direct control of the fan's speed.
Software (e.g., Speedfan) can effect PWM with some controllers by
changing the duty cycle of the voltage to the fan; i.e., some percentage
voltage on, the rest of the cycle with voltage off. I know that
Speedfan can help effect PWM via duty cycle of power cycling but only
with some controllers. Duty cycle is not linear to fan speed. That is,
a 50-50 duty cycle (on half the cycle, off the other half) does not
equate to half speed for the fan. The effect to average the voltage to
reduce it is akin to using a series resistor to slow the fan.

PWM does not alter voltage to the fan. Voltage is constant to the fan.
The PWM signal tells the fan the duty cycle but the fan changes its
speed, not because voltage to it was averaged to less voltage.

A 3-pin DC fan will have +12 VDC line, ground line, and sense (RPM)
output input. A 4-pin PWM fan will have +12 VDC and ground lines, a
sense (RPM) output, and a control (PWM) input. The PWM signal works
similar to duty cycling the input power but has the fan changing its
speed with a constant 12 VDC line.

The problem with DC fans is that duty cycle could be lowered so far that
the fan was incapable of spinning, and some BIOSes would trigger a
shutdown if they sensed zero RPM on the CPU fan. A good program using
power duty cycle to reduce voltage should set the fan to full speed upon
exit of the fan control software as a safety measure, like when
unloading Speedfan (but not if you kill it via taskkill.exe or using
Task Manager since the process is not shutdown to run its exit routine
but instead immediately killed).

While the OS can issue the same commands to the controller as can
Speedfan, it has no means of making a 3-pin fan into a 4-pin fan or
compensating for the wrong type of control configured in the BIOS.
Could be, for a home-built, the OP installed cheaper 3-pin DC fans, the
BIOS come pre-configured with PWM control, and the OS is trying to
control the fan using the wrong method. Could be the BIOS is configured
to always run the fan at full speed so the request from the OS gets
ignored.

"New computer" doesn't say if the OP bought a pre-built or bought the
components and built his own. Could be the pre-built has the wrong BIOS
settings for the type of fan(s) used. Could be the OP connected 4-pins
to 3-pin headers or used 3-pin fans (on 3- or 4-pin headers) and didn't
make the BIOS settings match.

Because the OP gave the make and model of the motherboard makes me
suspect that he jobbed the build himself but he used the wrong type of
fan or might've even got is misaligned on the header pins or he needs to
change the BIOS to match the type of fans that he actually installed.


VanguardLH, The computer was built by Xidax and I have not changed any
setting that I know of. Maybe just maybe they have a setting wrong.

Paul, I was up half the night using the powercfg -energy tool and there
was about 25 different settings and or problems that needed addressing
and a follow up test. Still no luck. I also noticed the ONLY way to shut
the computer off is by way of power switch on the computer. Any attempt
at a "software" shutdown does not work.

The one I could find here, only sleep was affected.

https://www.tenforums.com/software-a...onfigured.html

*******

In this one, Internet Connection Sharing seems to be hosing things.
With ICS, you can have two LAN connections, and the second connection
is for your 192.168.1.x subnet. It allows running multiple PCs (in
a serial string of machines) without a router.

https://www.tenforums.com/general-su...t-how-fix.html

*******

And there was nothing in the board forum.

http://vip.asus.com/forum/topic.aspx... nguage=en-us

*******

I can't find one that blocks shutdown. If a driver crashed,
I would think the PC would reboot.

Paul

On 20/08/2017 17:47, Drew wrote:
On 8/20/2017 10:15 AM, David B. wrote:
On 20/08/2017 15:40, Drew wrote:
On 8/20/2017 8:16 AM, David B. wrote:
Have you tried contacting 'support' (in green box) here?

https://www.xidax.com/lifetimewarranty

Yes I have done that. Long time customer. They build great machines
and you cannot beat their warranty.

So ..... how are they going to help you, Drew?

It is Sunday and I am waiting for a real person to answer. I am sure
they will but as I said it is Sunday.

OK - I understand! :-)

--
“Men occasionally stumble over the truth, but most of them pick
themselves up and hurry off as if nothing ever happened.� (Winston S.
Churchill)

On 8/20/2017 11:01 AM, Paul wrote:
Drew wrote:
On 8/20/2017 4:22 AM, VanguardLH wrote:
Paul wrote:

Char Jackson wrote:
On Sat, 19 Aug 2017 17:30:50 -0600, Drew
wrote:

New computer not sleeping?, running win 10 64 pro. When I set it to
sleep, The monitor shuts off, the keyboard and mouse do the same.
Computer itself (desktop) stays on and fans are turning. Router also
shows that the pc is sleepingÂ* After waiting a few minutes for the
pc to
go dark and the fans to stop (it does not) I have tried to move or
click
the mouse and or keyboard and nothing happens without a hard restart.
Cannot seem to figure out why. I have tried for hrs digging through
different settings and to no avail. I have tried power cfg -h on
and -h
off and still. Tried power options, usb settings and cannot seem
to find
what is needed. Any help would be appreciated greatly.

Specs are
Asus Rog Strix 270g motherboard
Intel core i7 7700k
32gigs ram
Nvidia turbo 1070 graphics card

Is this a homebuilt computer? Are the fans plugged in where they
should
be? That looks like a gaming motherboard. Is there anything in the
BIOS
that affects sleep and/or fans?


There was a time when "S1, S3, S1&S3" was a popular thing to
adjust in the legacy BIOS. If that happened to be set incorrectly,
a person would need to use "dumppo" to do an "override".

Dumppo no longer works in Win10. Which means you won't have
to learn how to use it.

I noticed powercfg has an "override" option, but I haven't
checked out what it can override.

And I don't think that setting is in a UEFI BIOS any more.
As the OS provides all the adjustments you could ever need.
Both Linux or Windows can hammer the settings needed. So
we should gradually wave goodbye to dumppo, in the
rear view mirror.

Â*Â*Â*Â* Paul

Since the hardware is responsible for regulating CPU temperature
*before* any OS can be loaded, and especially if no OS is loaded (drive
could be missing, dead, unreachable or no OS yet installed), there must
be controls in the BIOS/UEFI to have it control fan speed based on
temperature sensor(s).

http://dlcdnet.asus.com/pub/ASUS/mb/...B_20170406.pdf


Notice it has fan control.Â* While the OS can issue commands to the BIOS
to regulate fan speed, the BIOS gets to override any such requests.Â* If
the BIOS is configured to always run the fans at full speed then it
doesn't matter what the OS requests.

The manual also mentions differentiation between PWM (pulse width
modulation) and DC mode for controlling fan speed.Â* PWM is used for
4-pin CPU fans and DC (voltage) for 3-pin CPU fans.Â* With just 3 pins
(12VDC, ground, RPM), there is no direct control of the fan's speed.
Software (e.g., Speedfan) can effect PWM with some controllers by
changing the duty cycle of the voltage to the fan; i.e., some percentage
voltage on, the rest of the cycle with voltage off.Â* I know that
Speedfan can help effect PWM via duty cycle of power cycling but only
with some controllers.Â* Duty cycle is not linear to fan speed.Â* That is,
a 50-50 duty cycle (on half the cycle, off the other half) does not
equate to half speed for the fan.Â* The effect to average the voltage to
reduce it is akin to using a series resistor to slow the fan.

PWM does not alter voltage to the fan.Â* Voltage is constant to the fan.
The PWM signal tells the fan the duty cycle but the fan changes its
speed, not because voltage to it was averaged to less voltage.

A 3-pin DC fan will have +12 VDC line, ground line, and sense (RPM)
output input.Â* A 4-pin PWM fan will have +12 VDC and ground lines, a
sense (RPM) output, and a control (PWM) input.Â* The PWM signal works
similar to duty cycling the input power but has the fan changing its
speed with a constant 12 VDC line.

The problem with DC fans is that duty cycle could be lowered so far that
the fan was incapable of spinning, and some BIOSes would trigger a
shutdown if they sensed zero RPM on the CPU fan.Â* A good program using
power duty cycle to reduce voltage should set the fan to full speed upon
exit of the fan control software as a safety measure, like when
unloading Speedfan (but not if you kill it via taskkill.exe or using
Task Manager since the process is not shutdown to run its exit routine
but instead immediately killed).

While the OS can issue the same commands to the controller as can
Speedfan, it has no means of making a 3-pin fan into a 4-pin fan or
compensating for the wrong type of control configured in the BIOS.
Could be, for a home-built, the OP installed cheaper 3-pin DC fans, the
BIOS come pre-configured with PWM control, and the OS is trying to
control the fan using the wrong method.Â* Could be the BIOS is configured
to always run the fan at full speed so the request from the OS gets
ignored.

"New computer" doesn't say if the OP bought a pre-built or bought the
components and built his own.Â* Could be the pre-built has the wrong BIOS
settings for the type of fan(s) used.Â* Could be the OP connected 4-pins
to 3-pin headers or used 3-pin fans (on 3- or 4-pin headers) and didn't
make the BIOS settings match.

Because the OP gave the make and model of the motherboard makes me
suspect that he jobbed the build himself but he used the wrong type of
fan or might've even got is misaligned on the header pins or he needs to
change the BIOS to match the type of fans that he actually installed.


VanguardLH, The computer was built by Xidax and I have not changed any
setting that I know of. Maybe just maybe they have a setting wrong.

Paul, I was up half the night using the powercfg -energy tool and
there was about 25 different settings and or problems that needed
addressing and a follow up test. Still no luck. I also noticed the
ONLY way to shut the computer off is by way of power switch on the
computer. Any attempt at a "software" shutdown does not work.

The one I could find here, only sleep was affected.

https://www.tenforums.com/software-a...onfigured.html


*******

In this one, Internet Connection Sharing seems to be hosing things.
With ICS, you can have two LAN connections, and the second connection
is for your 192.168.1.x subnet. It allows running multiple PCs (in
a serial string of machines) without a router.

https://www.tenforums.com/general-su...t-how-fix.html


*******

And there was nothing in the board forum.

http://vip.asus.com/forum/topic.aspx... nguage=en-us


*******

I can't find one that blocks shutdown. If a driver crashed,
I would think the PC would reboot.

Â*Â* Paul

As always if I need help, this is the place to go.Tried every trick
mentioned and I have been at this for hrs now and still no luck. Will
wait for tech support I guess. It has to be a setting somewhere or
assembled incorrectly. I guess since tech support is in the same
building as the guy who built it they will put there heads together and
figure it out. I have never had a problem with them before. (3 pc's) as
well as I know friends who have purchased as well.

"Drew" wrote

| I have never had a problem with them before. (3 pc's) as
| well as I know friends who have purchased as well.

I've been building my own for many years and have
found that if I add a fan directly connected to power
it won't turn off with sleep. In other words, the CPU
fan and case fan plugged into the board turn off, but
at times I've added an extra that I've attached directly
to a power plug from the power supply. Those don't
turn off.

I don't know whether that may be your issue or not.
And I don't know whether there's any configuration
that can be used to override that behavior.

On 8/20/2017 4:44 PM, Mayayana wrote:
"Drew" wrote

| I have never had a problem with them before. (3 pc's) as
| well as I know friends who have purchased as well.

I've been building my own for many years and have
found that if I add a fan directly connected to power
it won't turn off with sleep. In other words, the CPU
fan and case fan plugged into the board turn off, but
at times I've added an extra that I've attached directly
to a power plug from the power supply. Those don't
turn off.

I don't know whether that may be your issue or not.
And I don't know whether there's any configuration
that can be used to override that behavior.


I can't tell you exactly where, but somewhere in the logs
you can find the wake events. Whether what they say
can tell you what to fix is...maybe.

I had one computer that was awakened by the lan port.
I found no way to fix it. Ended up disabling the
internal lan port in the BIOS and using a plug-in lan card.
Worked fine since.

On 8/20/2017 5:44 PM, Mayayana wrote:
"Drew" wrote

| I have never had a problem with them before. (3 pc's) as
| well as I know friends who have purchased as well.

I've been building my own for many years and have
found that if I add a fan directly connected to power
it won't turn off with sleep. In other words, the CPU
fan and case fan plugged into the board turn off, but
at times I've added an extra that I've attached directly
to a power plug from the power supply. Those don't
turn off.

I don't know whether that may be your issue or not.
And I don't know whether there's any configuration
that can be used to override that behavior.


Through trial and error I have found that when you use the (start menu)
power button to
(a) put the computer to sleep= monitor,keyboard,mouse= shutdown but pc
stays on. cannot wake it by keyboard or mouse

(b) restart= works as it should.

(c)choose shutdown and all seems to be the same as sleep but pc remains
on

(d) using the powerbutton on pc then all shutdown including pc
(lights,fans etc)

I do not think it is hibernate as I have disabling and re enabling that
several times.

Drew wrote:
On 8/20/2017 5:44 PM, Mayayana wrote:
"Drew" wrote

| I have never had a problem with them before. (3 pc's) as
| well as I know friends who have purchased as well.

I've been building my own for many years and have
found that if I add a fan directly connected to power
it won't turn off with sleep. In other words, the CPU
fan and case fan plugged into the board turn off, but
at times I've added an extra that I've attached directly
to a power plug from the power supply. Those don't
turn off.

I don't know whether that may be your issue or not.
And I don't know whether there's any configuration
that can be used to override that behavior.


Through trial and error I have found that when you use the (start menu)
power button to
(a) put the computer to sleep= monitor,keyboard,mouse= shutdown but pc
stays on. cannot wake it by keyboard or mouse

(b) restart= works as it should.

(c)choose shutdown and all seems to be the same as sleep but pc remains
on

(d) using the powerbutton on pc then all shutdown including pc
(lights,fans etc)

I do not think it is hibernate as I have disabling and re enabling that
several times.

Boot a Linux LiveCD and test the shutdown capability from there ?

That's to eliminate a Windows issue.

*******

Even their shutdowns aren't bulletproof. So I kinda hesitate to
suggest that, because of all the different symptoms I've
seen there.

Maybe I should test this :-)

OK, for the task, I'll suggest something which is non-mainstream.
Burn this to a DVD, even though it's a small file. It boots
a bit faster, because DVDs read faster. This is a kind of
Puppy Linux distro, only it works with modern hardware.

377,487,360 bytes MD5=50720d2edc71a3336aa0935a0151b4fd

http://distro.ibiblio.org/fatdog/iso/Fatdog64-710.iso

The top item in the boot menu on that works fine. You can
take your hands off keyboard, and it should boot on its own
after a ten second delay.

You'll see "five rows of dots" while it boots.
It's a "TORAM" distro and copies itself into RAM, which
is what takes so long. And that's why the speed of the
DVD, helps, as the boot process is mostly sequential.

When the GUI desktop eventually appears (with the FatDog
logo in the center), push the eject button on your optical
drive tray. It might take ten or fifteen seconds to eject,
but it should eject at that point.

Now, go to the menu in the lower left corner, and it should
have a shutdown option. It will put up two dialogs containing
"stupid questions". No, you don't want to save the session,
and yes, you're really sure. That sort of thing.

Then, when the actual shutdown sequence happens, you'll see
text, you'll see mention of "disks spinning down". And it
should actually execute shutdown without any additional
keyboard input from the user.

HTH,
Paul