X670E Taichi crashes / reboots

7700x seems to be working (Task Manager) OK with this BIOS. Tried with Userbenchmark, no problems with CPU performance.

sunl1te wrote: For those setups which has all 4 RAM slots populated, I would recommend install only 2 sticks and check if it is stable. Because having 4 sticks of RAM installed have higher (extremely higher) requirement on the RAM timing match. The only reliable way that make 4 sticks RAM running smooth is to buy the 4 stick package, never expect that 2 sets of 2 sticks package can run without issue, and reduce the clock speed or lower the timing setting or increase the voltage doesn't help here. Regarding "auto reboot without blue screen", this is indeed a blue screen error, check your windows "advanced system setting" ( press Win+Pause ), open "Startup and Recovery" setting page, uncheck "Automatically restart", so you can see the error code when BSOD happens again. In order to test the RAM stability, it's better to use memtest86+, test it for full one pass is enough, but if any error are detected it's RAM compatibility/timing issue(this doesn't necessarily means the RAM module has faulty bits but just the timing compatibility caused the instability). You can either remove one stick and re-test, or have the memory module RMA. But remember: if you are using 4 sticks of memory, always buy the 4 stick package. And also note the official supported memory speed: 2x1R DDR5-5200 2x2R DDR5-5200 4x1R DDR5-3600 4x2R DDR5-3600 https://www.amd.com/en/products/cpu/amd-ryzen-9-7950x I have seen so many cases that the blue screen, so called auto reboot ( this is actually a blue screen ), application BSOD are 90% related to the RAM. Hope this can help some people.

Well, I for one would like to thank you for bringing this to attention - I've had a couple of unexpected reboots since finishing my new build and testing it, and as many do, I went to BIOS-updates firstly, switching over to the latest in the beta branch just a few minutes ago. Since i'm running 2x32gb RAM at 5600mhz i'm surmising it's likely I'll end up having to turn them down to 5200mhz to stay within the CPU spec for the 7950 and attain complete stability.. I was sure I had checked everything beforehand, but it's really easy to miss important details like these, so again thank you!

Here's the build:
X670E Taichi
Ryzen 9 7950x
G.Skill D5 5600 64gb (F5-5600J3636D32GX2-RS5K - should be on QVL for the board)
Samsung 990 Pro 1tb
Samsung 870 EVO 4tb
Straight Power 11 1000w PSU
Dark Rock Pro 4 CPU cooler
Asus TUF OC 4080
A couple of Light Wings for chassis airflow
All in the good old Panteks Enthoo Pro full tower for plenty of space

The issues i've experienced so far:

One reboot at the windows login screen - the screen itself was stable for at least 5 minutes, moving the mouse around yielded nothing on-screen, at the first keypress to (maybe) engage Windows Hello, the system rebooted unexpectedly instead. This was a cold boot in the early morning.

One reboot during multitask stresstesting, running a GPU benchmark on one monitor, and "power-browsing like an idiot" on the other to monitor to check for crosschatter issues. Without warning, both screens went into inverted colours for about 2 seconds and froze, then the system rebooted. This was after about 4 or so hours of running other stresstests and benchmarks with no issues.

Both logged as classic ID 41 Kernel Power failure.

During both, the MB was running the latest non-beta BIOS, 1.11. I've now gone into the latest in the beta-branch, 1.11AS06, though as mentioned above, I'm kind of expecting to have another occurrence or two of unexpected reboots and then downclocking my RAM to 5200 in that eventuality. Updates may or may not follow, no promises, but also no threats

I've been following this thread because I've been having a similar issue. Upon booting from a cold start, the pc enters win 11 and runs for about 5 minutes before it shuts down and restarts. Occasionally upon restart the pc fails to boot into windows. Once I've shut down and restart, the system's rock solid with zero errors. This is a clean install of win 11, I've run both memtest and windows memory diagnostic with zero errors. I've also stress tested the cpu with zero errors. I'm guessing it's something in the bios. I'm running 1.11 latest stable...I tried the betas before but they were nothing but blue screens for me. I'll give as06 another shot though...possibly c-state related? Or could it be windows related? I don't have any other equipment to test the components on.

Yes, this is the error from event log that almost always shows up:

Faulting application name: dwm.exe, version: 10.0.22621.608, time stamp: 0x8783b131
Faulting module name: dwmcore.dll, version: 10.0.22621.900, time stamp: 0xb419c772
Exception code: 0x8898008d
Fault offset: 0x0000000000122e3e
Faulting process id: 0x0x6C78
Faulting application start time: 0x0x1D93B31AFA05048
Faulting application path: C:\Windows\System32\dwm.exe
Faulting module path: C:\Windows\System32\dwmcore.dll
Report Id: c576ec08-7b3d-496f-a7ec-7b27033a6c48
Faulting package full name:
Faulting package-relative application ID:

Again, this is from a clean install and I've tried repairing the installation but no errors were found. After rebooting from a cold start I get no errors in event log at all.

Yes, I never disabled the onboard gpu. I'll check to see if it's disabled in bios. If it is, I'll disable it in windows, leave my pc off while I sleep and see what happens.