Constant X870e Carbon motherboard issues

M

ms159d02e8

New member
Joined
Dec 28, 2024
Messages
18
I've already submitted a support ticket on this, but figured I'd post here too incase anyone has any insight.

I just build my system today and installed Windows, everything was fine until I started some Windows updates and left the computer. I returned to see it had shut itself down.

When I restarted it, it was stuck on code 08. I couldn't turn it off or restart it. So I turned off the PSU and unplugged it. When I tried again it started booting, then stuck on 0D. The CPU light was red and the RAM was Yellow.

Now trying to boot it results in it sticking and shutting down.

I cannot even get as far as the BIOS.

CPU: 9950X
GPU: currently none
RAM: Kingston Fury 6000MT/s CL30 64GB (two sticks)
PSU: Enermax Revolution D.F.X 1650
SSDs: Crucial T705 1TB, Kingston Fury 4TB, Kingston Fury 2TB (not that these should matter)
AIO: Arctic Liquid Freezer III 420 (also shouldn't matter)
BIOS Version: E7E49AMSI.1A21
 
Last edited:
Sort by date Sort by votes
I should also mention I am on the latest BIOS, and I did try clearing the CMOS multiple times. I also tried one more time and finally it booted into Windows. It sat for around 5 minutes and then shut itself off again.

I tried another boot. It was also able to get into Windows again, and hasn't shut down for at least 40 minutes. Currently with default BIOS settings.
 
Last edited:
Upvote 0 Downvote
You see, it's not like we suspect those parts to be involved necessarily. But if every user omitted the parts that they deem not to matter, we would never discover that some of them have completely inadequate PSUs for example (in quality and/or age), that they might use a cooler with a known issue or that's insufficient, that they use an SSD model with a known compatibility issue on AM5, and so forth. So first we have to see all the hardware in the system, then we can collectively decide to ignore some of it, this is the better way.

Are you still in Windows right now? Can you download HWinfo64 and take a screenshot? Open "Sensors", then expand all sensors by clicking on the little <--> arrows on the bottom, also expand the columns of the sensors a bit so everything can be read. Make it three big columns of sensors (or four, if the screen resolution is high enough). In the end, it should be a screenshot with all the sensors visible at once, like this:

yes.png


Make sure the power plan in Windows is on "Balanced" and just leave it running for a bit with the sensors open, to see what all of them are reporting.
 
Upvote 0 Downvote
I couldn't fit everything in at once on my backup monitor, so I'm including two screenshots.
 

Attachments

  • HWInfo_Screenshot_01172025.jpg
    HWInfo_Screenshot_01172025.jpg
    888.8 KB · Views: 27
  • HWInfo_Screenshot_01172025_B.jpg
    HWInfo_Screenshot_01172025_B.jpg
    299.9 KB · Views: 24
Upvote 0 Downvote
Well, the readings couldn't be any more normal. Of course, the fans are still on a fixed setting instead of a fan curve, but that's to be expected when you haven't changed anything in the BIOS yet. All the temperatures and voltages seem fine. The Kingston SSD's ones even to the point of obvious underreporting (an SSD cannot be equal - let alone cooler - than room temperature if it has a certain power draw in idle).

Looks like some good old methodical troubleshooting is in order here. For example, even though it's a hassle, i would try to rule out the possibility of bent socket pins. For that you would take off the CPU cooler, take out the CPU, and check for bent pins inside the CPU socket, also see here and the following posts. While they are technically all bent, the pins all have to look completely uniform under light, the tips of the pins have to line up in a perfect pattern, with none visually sticking out from the rest. So if you have some pins that don't look like the others, then you really have some bent pins. Ideally take some photos of the socket, upload to an image hoster and link them here, even if you don't think there are bent pins (sometimes they are not that easy to spot for the untrained eye).
 
Upvote 0 Downvote
I did check the socket quite closely before building. However, I can look again with some stronger magnification this time.

Oddly the system has, so far, been stable since yesterday. The only issue today is not being able to install a particular Windows update (but that's likely irrelevant)
 
Upvote 0 Downvote
MSI support had be reseat the processor, but now after turning it on, the red and yellow LEDs were back on with code 15. After a minute it switched to just the yellow LED with code 0d. After waiting about 15 minutes, I hit the reset button. The PC started up into Windows again.
 
Last edited:
Upvote 0 Downvote
Ok. I would probably try with just one module in slot A2, see if that makes it boot reliably, check what happens when you run a basic memory test with Memtest86 Free: Launch the included ImageUSB program (in Windows) to prepare a USB drive with it, then boot from that drive (press F11 after turning on the PC for the boot menu). The row hammer tests at the end of the test can be skipped, they're not important. But there should be zero errors in the other tests. Repeat with just the other module in slot A2. Trying to rule out the RAM as best as possible.
 
Upvote 0 Downvote
Good. As soon as you get issues again, you can try with the single module in A2. Not necessarily the RAM test again (was it Memtest86?), because it already didn't show errors with both modules, but purely to see if it can help with the failure to POST sometimes.
 
Upvote 0 Downvote
Yes it was Memtest86. It looks like many AM5 MSI boards are having various booting issues based on the forums. I'll keep testing and hopefully there will be a stable BIOS release in the near future.
 
Upvote 0 Downvote
The computer has shutdown suddenly twice today while running idle. No errors were shown. Looking at the Event Viewer in Windows, and Hardware Info 64, there are no signs of any issues. I've disabled c-state in the BIOS to see if that will have any affect.

Edit: Looks like that did not help any. It shut down again. I cleared the CMOS. It shut down almost immediately. Powered on again, shut off even quicker. Cleared the CMOS again and it shut off instantly. Powered on once more and now it's stuck with the red and yellow LED on code 00. I can't get it past that now.

I also tried to do a BIOS flashback just in case, but still stuck at 00 with the red and yellow LED. Everything seems to be pointing to a bad board.

I should have taken notes, but I've tried everything short of swapping the CPU and PSU as I'd have to buy new ones to do so. However neither of those components seem to be the problem, considering the strange nature of my issues and that so many other AM5 boards from all manufacturers seem to have similar problems (except AsRock, but those have been out of stock for months).
 
Last edited:
Upvote 0 Downvote
Back
Top