• Now's your chance win big! Join our community and get entered to win a RTX 2060 GPU, plus more! Join here.

    Pi Cast Episode 3 streams live on Tuesday, August 4th at 2:30 pm ET (7:30 PM BST). Watch live right here!

    Catch Scharon on the Tom's Hardware Show live on Thursday, August 6th at 2:00 pm ET (7:00 PM BST). Click here!

[SOLVED] 3700x+X470 ASUS Hero Won't Boot

Anonymiter

Distinguished
Jan 5, 2013
40
3
18,535
0
My new system is no longer booting for me. It seems to POST without issue and I can access the BIOS. Occasionally, the BIOS will have graphical and menu errors, but seems to work. However, if I boot to a UEFI environment, I get what seem to be Secure Boot BSOD’s and errors (0xc0000428). If I boot to a non-UEFI (CSM/Legacy) environment, I get either a blank screen that is either black or the windows BSOD cyan.

What I’ve tried so far:
  • Reseat all cables
  • Removed all peripherals and unnecessary components
  • Each RAM slot separately with two different DIMMs
  • Swapped GPU
  • Moved GPU from 2nd/top slot (PCIEX16/X8_1) to 5th/middle slot (PCIEX8_4)
  • Flashing BIOS
    • Previously prepared USB with new release through BIOS
    • New USB with new release of prepared from a different system through BIOS FlashBack button
    • Previous release though FlashBack
  • Resetting CMOS
    • In BIOS system defaults
    • Clear CMOS button
    • Removing battery and power flushing
  • Reseat CPU
    • Pins looked good
    • No intrusion or paste in or around socket
I have tried booting into the following:
  • Windows 10 v2004 installed on M.2 NVME SSD
  • Windows 10 v2004
    • From a previously prepared USB drive
    • A new USB drive prepared from a different system
  • Windows 7 Install from USB
  • SATA SSD with Windows 10 installed.
  • Macrium Reflect Bootable USB
  • Parted Magic Recovery USB
  • Memtest
  • HBCD Win10 PE
  • Gandolf’s WIN10PE
I’ve tried these things in the BIOS:
  • Enabling/Disabling CSM
  • Secure Boot
    • Switching from Windows UEFI to Other OS
    • Clearing and Installing Key Management on Both
  • Enabling/Disabling SATA Controller
  • Enabling/Disabling Legacy USB
  • Enabling/Disabling XHCI Handoff
  • Enabling/Disabling Fast Boot
  • Running everything at Default/Auto
  • Enabling DOCP and running various RAM speeds
System Specs:
  • AMD Ryzen 7 3700X
  • ASUS ROG Crosshair VII Hero X470
  • CORSAIR Vengeance LPX 3600 16GB (2 x 8GB) CMK16GX4M2D3600C18
  • Nvidia RTX 2070 SUPER (Reference)
  • Sabrent 1TB Rocket NVMe PCIe M.2 (SB-ROCKET-1TB)
  • Noctua NH-D15 chromax.Black
  • OCZ ZT 750W
  • Fractal Define S
I assume this is a problem with the motherboard, CPU, or RAM. The RAM seems unlikely given the symptoms but I have ordered an identical kit to test. The motherboard took forever to ship so I’m outside the return window for the RAM and CPU, but I still have until the 10th to return the motherboard. Obviously, I can RMA all three considering they are recent purchases but without knowing which one I’d hate to start the process for each.

I was wondering if anyone had any experience with this problem or could give me some advice on how to proceed if the new RAM doesn’t solve the problem.

Additionally, to give some context, this is a new system upgrade. I replaced the CPU, RAM, Motherboard, GPU, and SSD. The system was completely stable with the 3770k, GTX 670s, and SATA SSD previously. The system seemed stable for around two weeks with the new upgrades.

About a week ago I got a BSOD (KERNEL_AUTO_BOOSt_INVALID_LOCK_RELEASE) accompanied by some terrible noise coming from the speakers. I figured it could be due to the RAM being set at the rated speed of 3600 so I set it back to AUTO in the BIOS. About a day later I started experiencing programs crashing/closing and occasionally followed by a BSOD (usually IRQL_NOT_LESS_OR EQUAL). The frequency grew from once a day to several times until I decided to try to troubleshoot the problem.

I first tried reinstalling the Chipset and GPU drivers. Eventually more crashes occurred. I saw there was a recently released BIOS update. I had already updated the BIOS using FlashBack when I first got the system to make certain the CPU was compatible, but the new release was several months newer so I installed it. There was no immediate change but the BSOD but the crashes and BSOD were increasing in frequency just like before. I decided to backup the system before reinstalling Windows, but was unable to boot into Macrium Reflect over USB. The same error occurred with my windows 10 Install USB so I tried running Memtest. It got about half way through a pass with no errors by the time I came back to check on it but seemed to be frozen. Since then I have been unable to boot into anything and experiencing the symptoms described above.

EDIT: Clarified what was installed to M.2 SSD.
 
Last edited:

Anonymiter

Distinguished
Jan 5, 2013
40
3
18,535
0
So after a bit more testing I was able to get the system to boot to Memtest but within a few seconds of the test starting it would lockup/freeze again.

I got the new RAM in today and the same thing was happening. After a few reboots Memtest progresses to test 8 and threw up these errors before freezing again. I've been able to reproduce this several times now though it is still inconsistent.



I also noticed the QCode changed from [Ad] to [0d]. From what I assume [Ad] is supposed to signify that the board is at a point for the OS to take over. It starts the Memtest with this active but once it freezes it changes to [0d] but I can't find a consistent answer on what the QCode signifies.



I thought it might be just a first core that is unstable but I don't have a way to disable individual cores in the BIOS. I'm just able set it to 2/4/8 cores and disable/enable SMT, but nothing changes.

I'm going to try getting in touch with ASUS tomorrow about an options for RMA for the board, but this seems really likely to be a defective CPU?


On a side note, I installed the stock AMD Wraith PRISM cooler that came with the 3700x just to make sure there was no unforseen issue with the NH-D15. While there wasn't any change, it is a hell of a lot easier to work around and seems like a really amazing cooler for something that comes stock. :D
 
Reset CMOS and make sure UEFI is selected.
Try each RAM in B2 slot, if it's still throwing errors, something is bad with them.
Forget about W7, it doesn't have drivers needed.
Make Windows 10 boot USB using Microsoft Media Creation tool but without solving memory problem it's not going to install.
 

Anonymiter

Distinguished
Jan 5, 2013
40
3
18,535
0
Thanks to both of you for your responses!

Reset CMOS and make sure UEFI is selected.
Try each RAM in B2 slot, if it's still throwing errors, something is bad with them.
This is listed under things I've tried. It's a good idea and I've done it several times. To reiterate, I'm not having trouble booting into my main OS, I'm struggling to get ANYTHING to boot fully.
Try each RAM in B2 slot, if it's still throwing errors, something is bad with them.
Another good suggestion but already attempted. in fact I purchased a second, identical, brand new kit and tried each of those DIMMs in each of the 4 slots separately (so 16 different configurations?) and nothing changes.
Forget about W7, it doesn't have drivers needed.
Make Windows 10 boot USB using Microsoft Media Creation tool but without solving memory problem it's not going to install.
I had Windows 10 installed on the M.2 NVME drive in UEFI mode. When it and several other boot media failed, I attempted whatever else I had on hand including a copy of Windows 7, as it doesn't boot in UEFI mode. I had prepared Windows 10 v2004 using Microsoft Media Creation tool prior to the system becoming unbootable and I tried preparing a new USB from a different computer after as well. Neither boot without BSOD (0xc0000428).


I think you are right the CPU seems most likely after you have already replaced memory.
I'll see if I can speak to someone from ASUS support today that is knowledgeable enough to stear me in one direction or the other. I usually wouldn't think to try replacing the CPU first as it's so rare to get one that is bad that isn't simply DOA. I wish I had access to a second CPU but this is the only thing from AMD since socket FM2 that I have around and I don't know of any shops that aren't going to charge me silly diagnostic fees that are equipped to test the board or CPU with different parts.
 

Flayed

Honorable
Thanks to both of you for your responses!



This is listed under things I've tried. It's a good idea and I've done it several times. To reiterate, I'm not having trouble booting into my main OS, I'm struggling to get ANYTHING to boot fully.


Another good suggestion but already attempted. in fact I purchased a second, identical, brand new kit and tried each of those DIMMs in each of the 4 slots separately (so 16 different configurations?) and nothing changes.


I had Windows 10 installed on the M.2 NVME drive in UEFI mode. When it and several other boot media failed, I attempted whatever else I had on hand including a copy of Windows 7, as it doesn't boot in UEFI mode. I had prepared Windows 10 v2004 using Microsoft Media Creation tool prior to the system becoming unbootable and I tried preparing a new USB from a different computer after as well. Neither boot without BSOD (0xc0000428).




I'll see if I can speak to someone from ASUS support today that is knowledgeable enough to stear me in one direction or the other. I usually wouldn't think to try replacing the CPU first as it's so rare to get one that is bad that isn't simply DOA. I wish I had access to a second CPU but this is the only thing from AMD since socket FM2 that I have around and I don't know of any shops that aren't going to charge me silly diagnostic fees that are equipped to test the board or CPU with different parts.
You could do with one of those "loaner" cpu kits AMD were sending out for B450 boards
 
Reactions: Anonymiter

Anonymiter

Distinguished
Jan 5, 2013
40
3
18,535
0
You could do with one of those "loaner" cpu kits AMD were sending out for B450 boards
Yeah, I remember that! I'm going to look into this. If ASUS doesn't have a clear answer I'm going to contact AMD support before shipping anything off. Maybe they'll be willing to send a new CPU out for testing. lol

Great suggestion!
 

Anonymiter

Distinguished
Jan 5, 2013
40
3
18,535
0
AMD loaner kit was a no go. They basically told me to use the FlashBack built into the motherboard or prove that I couldn't before they would send anything out. They didn't specifically say but it seemed pretty clear in their reply that it wasn't intended for my troubleshooting/warranty purposes.

However, I found a local repair shop that was willing to test out a 3600x in the system, and it booted several times without issue. I've started the RMA process with AMD. I've read good things from other people that have had to return CPUs, so let's hope this goes smoothly and doesn't take 30-90 days.

So damned odd to have it working for two weeks without issue and then within another week go so bad. ¯\_(ツ)_/¯
 
Last edited:
Reactions: Flayed

Anonymiter

Distinguished
Jan 5, 2013
40
3
18,535
0
The CPU always used to be the last thing I would consider faulty but these days it seems to be less of a rarity. Perhaps because of the smaller process tech or the increased complexity of modern CPU's
Agreed.

I mean with smartphones and laptops these days, I usually put most of the blame on the CPU as a failure point. The BGA chips with lead free solder are constantly going bad and creating massive amounts of e-waste. But I have no idea what to make of this. I don't imagine it's a bad solder joint in this case. You might be right that the smaller process could be a contributor but CPUs have always been going about this trend for awhile seemingly without an increase in unreliability. But with my sample size of 1 I'm not sure I could say either way.
 

Anonymiter

Distinguished
Jan 5, 2013
40
3
18,535
0
I got the replacement 3700x today and it is working great. Let's hope it continues to do so!

The RMA process overall took about 2 weeks I'd say. AMD sent me a prepaid label and all I had to send in was the CPU itself. The day they got it they sent me an email confirming receipt and that it was undamaged. About a day later they sent me another email confirming it had been tested and confirmed defective. I got an email stating that a new one was being sent but never got a tracking number. The new one arrived about a week after the last email.

And the new one came in a sealed, retail package so I now have two of the stock Wraith Prism coolers which is pretty cool I guess.
 
Reactions: Flayed

ASK THE COMMUNITY

TRENDING THREADS