Z170 Extreme7+ Shut Down Problem

One missing bit of important information... the OS and version you are using.

I'm guessing it is Windows 10?

I have the same mother board as you do, and I use G.SKILL memory running at 3200, and I can OC it to 3600. But I don't have the issue that you are experiencing.

Your symptoms are different than we usually see, with a system running on normally immediately after a shutdown. This has become an issue that I won't call common, but has been seen enough to be known to us. I'll mention a few things that could be the cause. Please read all of them before deciding what to do.

One cause of this is the Windows Fast Startup feature, introduced in Win 8.

If you'd like to try this, it can be disabled by first opening the Windows Power Options, normally used to choose or modify a Power Plan. On the upper left of this screen find and select the "Choose what the power buttons do" option.

You'll likely need to click on the Change settings that are currently unavailable before you can disable the Fast Startup option. Then uncheck this option under Shutdown settings. You'll need to shutdown normally (if possible) before this feature is disabled. You can enable it again by simply reversing the process.

The only time I had a similar issue is when I used a program to clone an OS installation from one drive to another. That was using Macrium Reflect, although I really do not know if it was the cause. The only fix I found was a new OS installation.

A simple potential fix is clearing the UEFI/BIOS. You could do a full clear by moving the jumper to the clear position, and removing the board's battery. Just removing the battery will not provide the same results as doing both. Don't forget to save your UEFI/BIOS settings in a profile before doing this, if it is difficult to reset them from your memory. One potential issue with a saved profile is...

A setting in the UEFI might be the cause. The Deep Sleep option in Chipset Configuration is best left at its default setting. Did you happen to change the Restore on AC/Power Loss option? Just curious as a potential cause of this issue in general.

Do you have the Intel network driver installed? Win 10 (if you use it) does not get along with the Intel networking driver in some ways, and still does after several releases. Disabling the various Wake on LAN options in Device Manager might help.

I suggest checking the wiring of the power button on the PC case to the system header on the board. While the polarity of this switch may seem irrelevant, I leave nothing to chance. Do you think the wiring and switch in the PC case is in good condition?

Edited by AMGala - 13 Apr 2016 at 10:45pm

AMGala wrote: parsec wrote: One missing bit of important information... the OS and version you are using. I'm guessing it is Windows 10? I have the same mother board as you do, and I use G.SKILL memory running at 3200, and I can OC it to 3600. But I don't have the issue that you are experiencing. Your symptoms are different than we usually see, with a system running on normally immediately after a shutdown. This has become an issue that I won't call common, but has been seen enough to be known to us. I'll mention a few things that could be the cause. Please read all of them before deciding what to do. One cause of this is the Windows Fast Startup feature, introduced in Win 8.  If you'd like to try this, it can be disabled by first opening the Windows Power Options, normally used to choose or modify a Power Plan. On the upper left of this screen find and select the "Choose what the power buttons do" option. You'll likely need to click on the Change settings that are currently unavailable before you can disable the Fast Startup option. Then uncheck this option under Shutdown settings. You'll need to shutdown normally (if possible) before this feature is disabled. You can enable it again by simply reversing the process. The only time I had a similar issue is when I used a program to clone an OS installation from one drive to another. That was using Macrium Reflect, although I really do not know if it was the cause. The only fix I found was a new OS installation. A simple potential fix is clearing the UEFI/BIOS. You could do a full clear by moving the jumper to the clear position, and removing the board's battery. Just removing the battery will not provide the same results as doing both. Don't forget to save your UEFI/BIOS settings in a profile before doing this, if it is difficult to reset them from your memory. One potential issue with a saved profile is... A setting in the UEFI might be the cause. The Deep Sleep option in Chipset Configuration is best left at its default setting. Did you happen to change the Restore on AC/Power Loss option? Just curious as a potential cause of this issue in general. Do you have the Intel network driver installed? Win 10 (if you use it) does not get along with the Intel networking driver in some ways, and still does after several releases. Disabling the various Wake on LAN options in Device Manager might help. I suggest checking the wiring of the power button on the PC case to the system header on the board. While the polarity of this switch may seem irrelevant, I leave nothing to chance. Do you think the wiring and switch in the PC case is in good condition? Thank you for your reply.  I am running Win10 Enterprise public (not insider), fully updated.  I did a fresh install via DVD using an external DVD drive. I did see the thread on Fast Startup causing issues.  I do not have Hibernate enabled, and therefore I don't have any option to enable/disable Fast Startup. I do have the Intel network driver installed.  I'll disable WOL and see if it solves the problem before tackling the BIOS suggestions. I'll also double check the power switch polarity, but it is otherwise functioning correctly to power up the system and also to wake the system from sleep.  I am using a brand new case, but of course manufacturing defects can and do happen. I'll report back soon.

parsec wrote: The 00 Dr Debug code is just the first test that occurs in the POST process, a test of the CPU. If the CPU was broken or one was not inserted in the board, the resulting code displayed would be 00. In your case, the 00 code shown during shutdown is not related to anything besides the display being on (for some reason), and the first code is shown while the display is waiting for the POST process to begin. Whatever controls this display cannot know if the PC is shutting down or restarting, so the first code is displayed, 00. At least that is my analysis of how and why this display is acting. I do not believe it has any significance. I don't understand what you mean about being unable to disable the Windows Fast startup feature because Hibernate is not enabled. For example, this is my configuration of that screen on another PC: You must click Change settings that are currently unavailable, but it can be done. Is this option still enabled? Your memory causing the inability to shutdown seems really strange, but there is so much going on behind the scenes with PCs, I would never say nothing is impossible with PCs. Speaking of your memory, is this it: http://www.newegg.com/Product/Product.aspx?Item=N82E16820233863 If so, the XMP profile may be optimized for X99 systems, rather than Z170. That means manual setting of the VCCSA and VCCIO voltages might be required. Unless your memory is a double sided (DS) design, when using two DIMMs you should use the A2 and B2 slots.

Great, great, great, sooo glad you figured this out!

DDR4 memory is becoming a nightmare, the crazy problems it seems to cause! I have never heard of such things with DDR3 memory. I still think Win 10 is related to this somehow, but hardware (DDR4) is a contributing factor.

About the Dr. Debug codes... let's remember what those are. They are POST codes, Power On Self Test, a set of hardware verification tests that ONLY occurs when a PC starts after a shutdown or restart, before the OS booting process begins. Each test that occurs during POST has a corresponding code. That code is displayed on the Dr. Debug display while it is running.

The codes are NOT error codes in the usual sense. The appearance of a POST code does NOT indicate a problem UNLESS a POST sub-process/test fails, then the code for that sub-process/test remains displayed to give us a clue about what failed. POST then stops, and the PC will not attempt to boot since POST did not complete successfully. A POST code displayed under any other circumstance is an undefined situation.

Seeing a POST code, and then another, and another, etc, only means the POST sub-process identified by each code completed successfully, and moved on to the next. The codes do not indicate that a POST process failed every time we see one, or that some type of error occurred.

When the non-shutdown, no POST or OS boot situation you had occurs, the PC never went through the POST process, or the boot process, right? It never shutdown, and just goes right back to a booted OS desktop.

Therefore I cannot take anything shown on the Dr. Debug display seriously in the non-shutdown, no POST or OS boot, back to the desktop situation. Code 00 is a CPU failure of some kind. If the CPU failed, how can the OS be running? That code happens to correspond to the first test done by POST, so the display might simply have it ready for a POST process to occur.

How the Dr. Debug display functions during a non POST process situation, I don't understand. But as long as the POST process is not happening, anything displayed on Dr. Debug when the POST process is clearly not happening is meaningless. Plus the OS kept running, so nothing went wrong besides the shutdown failure, which is not related to POST at all.

I don't know what is shown on the Dr. Debug display on my PC when it shuts down. Frankly I'm not watching it, since there is no reason to do so. Since POST was not happening, and my PC shutdown correctly, any spurious, random POST codes that appear on the display are meaningless.

If you go into the UEFI/BIOS, what do you see on the display? A4, A6, or Ax? Those are the last POST codes displayed

Yes, my board has the Dehumidifier function, which I don't use, and is disabled. It's in the Tools screen.

I see what happened to your Hibernate option now, you removed it via non-conventional means (meaning outside of normal Windows operations) with the Samsung Magician software. That seems to do a Windows Registry edit that causes the option to be completely removed.

While both Hibernate and Fast Startup depend upon a file written to the OS drive to work, the Fast Startup file is smaller and different than the Hibernate file. The only time a Hibernate file is created is if you specifically select Hibernate instead of Sleep or Shutdown.

I'm long past the "wear out the SSD with too many writes" paranoia, and don't bother to do any of those kind of SSD write reduction tweaks anymore. If a drive of any kind can't be used without babying it, it's not worth using. The irony of "knowing" when NAND memory will wear out (any SSD ever endurance tested surpassed its NAND specs and Total Bytes Written specs by such a large margin it was amazing) created the concern for its lifetime. Meanwhile HDDs never provide a specification similar to NAND for the magnetic media it uses. So while HDD media lifetime is unknown, we worry about SSD NAND lifetime?

The Magician software and its various options for configuring the SSD environment are kinda funny, some of them contradict what the others say. Each option plays to the audience that chose it.