Taichi x399. Only 00 error code, won't shutdown

System:
1950x
Taichi X399
RX Vega 64
G.Skill Flare X 3200MHz 4x8GB
Cooler Master V1000 Power Supply

CPU has always been run at stock settings. Only settings I changed were to enable XMP.

Issue:
My Taichi X399 has had all sorts of odd issues pop up recently. Firstly the board only displays code 00 on Dr. Debug. It doesn't matter the state of the system. From the moment you hit power on it just shows 00 even while in windows.

Second issue is that the board doesn't shut down from anywhere without a hard shutdown being required (holding the power button on the case or board.) This happens in BIOS, Windows, live USBs, etc.

Thirdly the board sometimes won't even power on. When I moved cases the board would have only a blue LED behind the gear and no power. All cables were fine. I breadboarded the system (took out of case and put on anti-static open air surface), and it voila it worked again. So I put it back in the old case and the issue arised. I breadboarded again, followed by putting into the new case and it worked fine. I took the board out for system maintenance, and when I put it back in it stopped powering on yet again. I took it out of the case for the last time and it wouldn't power on. I removed everything from the board (including CPU) let it sit overnight and put the CPU back in and it booted. It's currently booting on the open air table.

Things I have done to fix the issue:
Updated bios
Used a different power supply
Removed all but 1 RAM stick (and ran Memtest86 to confirm my RAM was fine)
Tried multiple different drives with windows
Tried different graphics cards
Load optimized defaults

The board ran fine for months till the shutdown issue started about a week and a half ago. I thought it was maybe just a drive error at first and went about my business moving it to the new case when all these other issues started popping up and I noticed the 00 debug code.

Something I will say about my board is the CPU mounting system has always been a bit of a headache. I'm not sure if this is true with all of x399, but you have to really use some force on screws 2 and 3 to get them to thread so they close. I've tried different levels of pressure on screw 1 as well before attempting 2 and 3 but it didn't change the outcome.

Anyone have any ideas of what's going on? Should I send the board back for RMA? I don't have any way to test a different CPU in the system to see if its causing the issue but I highly doubt its the CPU. I checked for bent pins and the such but no visible bent pins.

Thanks for the help!


Edited by JimJames1978 - 17 Feb 2018 at 3:03pm

MisterJ wrote: JimJames1978, weird!  There are two different sockets for the X399 boards.  One has your (and my) screw problem.  My first X399 board had another vendor's socket and did not have the problem.  What version of Windows are you running - also BIOS/UEFI?  Right now my Dr.Debug is off (blank).  I will check it when booting in a minute.  Please post all your drives.  You should only run the chip set drivers from here: https://support.amd.com/en-us/download I think you should open an ASRock support ticket and start a discussion.  I suspect an RMA is in your future.  Enjoy, John. EDIT: After power up lots of different hex codes, finally d2, then blank.

I tried an update to the latest bios using the built in updater on the UEFI bios screen and it didn't change anything. Running the latest version of W10.

I have tried multiple drives. I can't say for sure, that this is the case but this mobo may have killed my SSD or it may have been a coincidence that it died at the same time. I was getting crashes and BSODs on the fresh windows install I had on it when I would put it under load (caused by netwtw4.sys). So I decided to clone my windows that wasn't having issues from my HDD to it. It got to the bios screen and sat there for about 5 minutes, then got to windows and I got an immediate BSOD "something is unplugged". After that my SSD was no longer detected anywhere and having it plugged it to the machine causes that massive lag on the bios splash page when booting. It's a 4 year old SSD so maybe it just finally kicked the bucket, but it sure was a coincidental timing. The funny part is, I enabled SATA hotplug to check and see if the drive was still being read on plug in, and it is when you are in windows. It clearly detects something plugged in, but it's not displayed anywhere including on the disk management screen, diskpart, bios sata options, etc.


Edited by JimJames1978 - 18 Feb 2018 at 2:01am

MisterJ wrote: JimJames1978, please answer both these questions.  What version of W10 and UEFI are you running?  I am running Version 1709 (OS Build 16299.248) of W10 x64 and UEFI 2.00.  Only use Instant Flash from the UEFI to update and always do a F9 and F10 before.  I would also suggest never running a cloned system, only fresh installed.  Enjoy, John.

Same bios. Windows Version 1709 (OS Build 16299.192)

It was a clone of a windows install that had worked for 2 months on the system. The fresh install was having the problem.


Edited by JimJames1978 - 18 Feb 2018 at 5:47am

MisterJ wrote: JimJames1978, is your W10 32 or 64 bit?  Enjoy, John.

It's 64 bit of course.

I did some messing around in the bios and realized that there was an option for Dr. Debug. It was set to disabled on my bios. I enabled, but nothing different happened. I also took a peak at the HW tab and it was showing obviously wrong readings on the temps there. It was showing -1c cpu and -5c mobo temps. At this point I'm wondering if there's a broken sensor causing all my issues.