ASUS Crosshair VI Hero Ryzen 1700x no longer runs Windows 10

[cletus87408]

Here's the situation.

The system has been running stable for many months. I have been running a Ryzen 7 @ 3.8GHz / 3200 MHz memory (about 11% overclocked) with a stock cooler for the entire time with no issues. I started a backup to my USB drive and walked away. When I returned, the system was at the boot prompt indicating that it had rebooted due to a CPU overheat. I have been monitoring the system for a long time - it has never shown any indications of overheating even when gaming or fully stressed - certainly not while performing an unattended backup. This is the first time anything like this has occurred.

Since then, I cannot run Windows except in safe mode. I get a selection of blue screen events - often, IRQL_NOT_LESS_OR_EQUAL, but several others too. The system boots fine, I can log in, but within 2 minutes something fails and the system crashes. If I boot in safe mode, the system runs without error. So clearly it's a bad driver, right? I finally bit the bullet and decided just to reinstall windows since I was too limited in what I could accomplish in safe mode, but here's the rub. I cannot get the windows installer to run either. It boots, discovers my hard disks, but before I get to the installation phase, it hangs.

I have replaced the video card with a spare, removed my sound card, uplugged every USB device save keyboard and mouse. I have reset the BIOS to optimized defaults for now, with no overclocking and the memory bus running at 2133 MHz. Windows memory diagnostics pass without issue. I can run CPU-Z from safe mode and execute a CPU stress test, but this I expect isn't a very telling diagnostic. I have some reason to suspect it might be related to the USB chipset given some of the failure modes, but nothing concrete.

So I'm left thinking it's one of two things - something non-critical failed catastrophically on the motherboard (hence the unexpected overheat event) or Windows has made a recent kernel change, delivered through windows update and in the most recent online boot image that is not compatible in some way with my hardware. Either way, I'm kind of stuck - I can't run the system, I can't reinstall the system, and I don't really want to shotgun a motherboard and processor in search of the cause. Does anyone have any experience like this?

 

[cletus87408]

Crossposted from Windows section, just covering my bases.


Here's the situation.

The system has been running stable for many months. I have been running a Ryzen 7 @ 3.8GHz / 3200 MHz memory (about 11% overclocked) with a stock cooler for the entire time with no issues. I started a backup to my USB drive and walked away. When I returned, the system was at the boot prompt indicating that it had rebooted due to a CPU overheat. I have been monitoring the system for a long time - it has never shown any indications of overheating even when gaming or fully stressed - certainly not while performing an unattended backup. This is the first time anything like this has occurred.

Since then, I cannot run Windows except in safe mode. I get a selection of blue screen events - often, IRQL_NOT_LESS_OR_EQUAL, but several others too. The system boots fine, I can log in, but within 2 minutes something fails and the system crashes. If I boot in safe mode, the system runs without error. So clearly it's a bad driver, right? I finally bit the bullet and decided just to reinstall windows since I was too limited in what I could accomplish in safe mode, but here's the rub. I cannot get the windows installer to run either. It boots, discovers my hard disks, but before I get to the installation phase, it hangs.

I have replaced the video card with a spare, removed my sound card, uplugged every USB device save keyboard and mouse. Installed BIOS is 1602, the latest at this time. I have reset the BIOS to optimized defaults for now, with no overclocking and the memory bus running at 2133 MHz. Windows memory diagnostics pass without issue. I can run CPU-Z from safe mode and execute a CPU stress test, but this I expect isn't a very telling diagnostic. I have some reason to suspect it might be related to the USB chipset given some of the failure modes, but nothing concrete.

So I'm left thinking it's one of two things - something non-critical failed catastrophically on the motherboard (hence the unexpected overheat event) or Windows has made a recent kernel change, delivered through windows update and in the most recent online boot image that is not compatible in some way with my hardware. Either way, I'm kind of stuck - I can't run the system, I can't reinstall the system, and I don't really want to shotgun a motherboard and processor in search of the cause. Does anyone have any experience like this?

 

[Robert17]

Your other thread (http://www.overclockers.com/forums/...VI-Hero-Ryzen-1700x-no-longer-runs-Windows-10) should be enough to git 'r dun. Mods please merge.

 

[Robert17]

Welcome to OCF!

First, please list all your system components in a signature (Forum Actions tab) so that we can better able understand all that's involved. You stated 'stock cooler' which usually means the one that came with the CPU? If so, you are probably running on the low end of cooling options, especially when you mention 3.8GHz on AMD. Check the temps using HWInfo for starters and let's see what that yields.

Also, when installing an OS you should unplug all devices but the basics: HDD, graphics card, keyboard, mouse. Once you've got a stable image, add the other peripherals; you'll be able to narrow down the suspects of your consternation better that way.

Latest BIOS?

 

[||Console||]

Stock cooler ?
I didnt think any Ryzen with the X came with a HSF .
am I wrong ?

 

[cletus87408]

Ok, for purposes of this discussion, the only system specs that matter are -

AMD Ryzen 1700x CPU
ASUS Crosshair VI Hero motherboard, BIOS 1602 (which was current until today - I see a 1701 version was released as we speak)
8 Gb Patriot Viper PVE416G320C6KGY DRAM (when only a single stick is installed). No, this is not on the vendor QVL
Zalman CNPS 10X Optima sharkfin cooler
Radeon 6850 medieval graphics card, for testing
Lepa MaxBron B1000M power supply
Everything else including all hard disks have been disconnected.

This system has been running stable for about 6 months. As of today, running Windows setup from a USB stick locks up every time after a couple of minutes. Before I tore the system down, it would run in safe mode to a dual Samsung EVO 850 SSD Raid 0 array, but for troubleshooting this has been disconnected. Behavior is the same for BIOS default settings and DOCP profile.

Not sure what changed from a completely reliable and stable gaming system to one that will no longer boot Windows.

 

[cletus87408]

And for the record, BIOS 1701 made no difference.

This has now become a case of determining what should be replaced - motherboard, CPU, RAM?

 

[||Console||]

Looks like you have tried most things , does that MB have a dual bios ?
What Voltage have you been using for 3.8 I didnt see it listed .
Can you make a New Windows USB install? or test on another system to see if it is borked .
Have you remounted the HSF ?
Im just tossing ideas here .


Do you know any one that has a ryzen that you can swap things around to test ?
If not your going to have to take it to a shop that will be able to test .
Or start rma on one part @ a time .

 

[Johan45]

Have you tried a different SSD/HDD ? Or even different data cables

 

[cletus87408]

Looks like you have tried most things , does that MB have a dual bios ?
What Voltage have you been using for 3.8 I didnt see it listed .

I'm not sure. I never wrote them down after letting the ASUS auto-tuner run. I'm not into aggressive overclocking, especially since I don't yet have a water cooler, but I figured I could let the motherboard manufacturer do the heavy lifting for a 10% gain.

Can you make a New Windows USB install? or test on another system to see if it is borked .

I tested it as far as verifying that it doesn't lock up on another machine.

Have you remounted the HSF ?

Yup, regreased and remounted.

Im just tossing ideas here .

Yeah, I know the feeling. I did bite the bullet and order 16GB of G.Skill from the QVL. Not that I have any real confidence that there is a memory problem, but it "feels" like a memory issue. If that fails, then I guess the MB gets shipped to ASUS.

Thanks for the input.

- - - Updated - - -

Have you tried a different SSD/HDD ? Or even different data cables

The USB installer dies with no hard disk attached at all. The only components I have installed are motherboard, memory, graphics card, and power supply.

It boots properly, I can punch through all of the menus, but if I let it sit for 60 seconds at the choose a partition screen, everything locks up - with or without an attached drive. If I reattach the system disk, boot to windows, reboot to the memory diagnostic, all of the memory tests pass with no issues. If I run in safe mode, no crash. It seems to be only a full Windows load, either my current installation or the installer triggers the crash. The crash is always delayed by a couple of minutes from boot time.

 

[Sentential]

Can you get to the recovery console? I had my laptop grenade itself and it was kernel damage, no ideas as to the source but I was able to fix it. You just have to run chkdsk on the disk itself or the drive. ie: chkdsk c: /f /r /b /x

 

[t1nm4n]

Let me get this right, your trying to install an OS onto nothing? Or did you mean to say there is no HDD but you are using a SSD. I saw you said you had 2 SSDs in raid, but when asked about whether or not your drives are good or if you have spares to test you said you had no HDD (I'm really not trying to be nit-picky but it does matter) which means your installing the OS onto what? Don't feel bad I tried to install windows before I plugged my drives in, installed and cables run, looked plugged in from a few feet away, took me a bit to figure out they were unplugged when I was wondering why nothing was working right.

 

[Mr.Scott]

Ram issue.
Wait for the new sticks and try again.

 

[cletus87408]

Let me get this right, your trying to install an OS onto nothing?

Exactly. As part of troubleshooting to see if one of my SSD drives was going bad, I tried the experiment of running the Windows installer with no drives attached. Obviously, I can't complete the installation, since there is no drive to put the OS onto, but I can still run the installer up to the point where it asks for a target partition.

And even with no drive in the system, it still hangs at the same point.

 

[cletus87408]

Ram issue.
Wait for the new sticks and try again.

Nope.

Turns out it's either the motherboard or the CPU. Not sure which.

I was in contact with ASUS to see if I should set up for an RMA. The last thing he suggested was to try the old CPU in a known good motherboard, or vice versa.

Who has $600 worth of spare hardware just lying about unused? Well, I do now. I bought replacement parts for both over the weekend since it became clear that whatever the problem, I was going to be out a computer for some time. So I decided to just replace the old components and get them fixed since both are still under warranty. Put the system back together again, and it runs like a champ, no changes to the OS.

Now the rock and the hard place - am I willing to drop a possibly bad CPU into a known good motherboard? Or probably even worse, put a known good CPU into a possibly defective motherboard? And risk the brand new hardware that I just purchased to replace it? I'm an E.E. - I know the likelihood that the old hardware will trash the new is not highly likely, especially since it DOES run in limp mode, but it's also not impossible, especially if a component on the motherboard has gone south. It's my money on the line. So we're in negotiation as to how I should proceed.

Thanks all for the input. MS is off the hook on this one.

 

[cletus87408]

Old CPU in new motherboard works as expected. ASUS tech agrees that I may start an RMA for repair of the motherboard. So at the end of the day I'll have two.