MSI B450 Tomahawk PCIe 16x slot glitch

[vileness999]

Hi, I've been having "Live kernel error 141" problems with my setup for unknown reasons.

 

TLDR:

Worked without fault for about 1 month, then started randomly crashing without blue screen, display signal completely lost, OS unresponsive, required forced shut down. Windows reported “Livekernelerror 141”. Tried uninstalling and updating everything (drivers, Windows, BIOS), problem became worse, became consistent crashing every time booting the computer unless power cycled quickly. If computer were off for more than 10 seconds, powering on results in normal POST but Windows fails to load, blank screen, “no signal” on monitors. Required forced shutdown and if powered back on in less than 5 seconds, Windows boots normally. Reinstalling Windows resolved issue for 1 week, then started again. Tested different graphics card (GTX 960), same problem. Switched back to using RTX 2070, used the PCIe 4x slot, problem stopped.

 

MORE DETAILS:

For more information, this is my complete setup:

Motherboard: MSI B450 Tomahawk ( version 1, NON-MAX)
CPU: Ryzen 3600, with Coolermaster 212 Cooler

RAM: GSkill Ripjaws V 16GB 3200MHz, SK Hynix C-DIE, used default XMP settings

GPU: MSI RTX 2070 Gaming Z

Storage: Samsung 860 EVO 1TB

PSU: EVGA supernova G+ 650W

OS: Windows 10 64bit

Case: Silverstone RL06

Monitors: dual monitors, ACER KG271UA connected via displayport and secondary old 1080P Asus monitor connected via HDMI adapter to DVI.
Manual Overclock: NEVER, not even once.

 

I was an early adopter, bought everything the day after Ryzen 2's release brand new. The retailer updated the BIOS so the computer was working perfectly the moment it was assembled. The first time it glitched out, I was not gaming, just streaming video/browsing memes and suddenly the audio glitched out and displays went black. After happening a few more times, I updated the BIOS and (Windows automatically updated) Nvidia drivers were up to date but I used DDU software to try and do clean installs. I updated to the 1C beta BIOS on the MSI support page, then the crashes stopped being random, they caused Windows to fail to load every time unless forced to shutdown followed by rapidly powering it on within 5s. I rolled back BIOS to the 1B version and it still happened - then it recovered by some miracle and worked without fault for about 2 days, only to continue failing afterwards. Switching out to another reliable GPU didn't resolve the problem. Reinstalling Windows fixed the problem for 1 week only. The only reliable solution was to not use the PCIe 16x slot and switch to the 4X slot, which is terrible for gaming because I'm getting a 70% or so performance drop. I am fairly certain this is not the memory, MemTest86 confirms the sticks are fine. There's a small possibility it could be a PSU or CPU or SSD problem, but considering switching PCIe slots fixed the problem, the probability is low.

 

I have submitted an RMA for the motherboard but who knows how long that will take. I am going to purchase a B450 Tomahawk MAX board and if they replace/repair my old non MAX board, I'll keep it in case the new board breaks as well. Money is no barrier, my PC is my life, I wouldn't skimp here. My question is, what do you suspect could have caused this? Could I run some more accurate tests to verify if the mobo PCIe 16x slot in particular is defective? Is my GPU too heavy? Just in case there's the odd chance I could have done something wrong, I don't want to break my new board, so any suggestions here? MSI online forums didn't offer any explanations. If any experts here have any advice, I give my thanks in advance. Also if other people have experienced this problem, maybe it could be more widespread regarding these MSI boards.