Question ASRock B450M-HDV R4.0 HWMonitor temp questions. VRMs overheating? (Also, FYI on OCing Athlon 3000G in this board)

[VirtualLarry]

TMPIN6 and TMPIN8 around 120C and 111C. Max of 121C and 113C.

Are these the VRM temps, and should I be concerned?

APU is Athlon 3000G, 2x8GB Team Vulcan (Z?) DDR4-3200 @ XMP, APU is OCed to 3.900Ghz @ 1.3000V. Running CPU-Z stress-test.

Hmm, starting and stopping CPU-Z Stress-test, for a little while, and with a short delay in-between, TMPIN7 goes down to 31C, and goes up to 57C when I run the stress-test. Guess that's the VRMs?

What on a mobo would run 120C all the time? And if these were faulty / mis-read sensors, I wouldn't expect that they would change slightly, I would think that they would be pinned to 128 or some round number (binary).

 

[Iron Woode]

go into the BIOS and see if they list a VRM temp there.

If they don't then there might not be any software way to read vrm temps.

 

[VirtualLarry]

There's a temp for "motherboard", along with "CPU". With my 3900Mhz @ 1.300V OC, CPU was in the 40C range, and motherboard was 33C.

 

[VirtualLarry]

Also, I had a royal PITA trying to OC on this board, at least initially.

It shipped with the 3.70 BIOS/UEFI, which is the newest available on the web site.

Under the OC'ing menu, there's a setting for "OC Mode", "Auto" or "Manual". There is also a separate setting for "Frequency", "Auto", or "Manual", and Manual lets you dial in a freq and voltage.

For some reason, when I set OC Mode to Manual, and then set Frequency to Manual, even 3700 at the default of 1.300V, it wouldn't get past the POST screen. That's after setting XMP, on some Team Group Vulcan Z DDR4-3200 2x8GB RAM.

I couldn't get ANY OC to "stick", couldn't access BIOS again either, although I could sometimes reboot with C-A-D. Sometimes, the ATX Power Button wouldn't even force power-off. Very weird.

Then I decided to go a different route, after umpteen clear CMOS's, I left "OC Mode" on "Auto", and ONLY changed Frequency to Manual, 3600, left voltage at 1.300V.

Bingo! Booted Windows 10, ran CPU-Z stress-test without crashing for five minutes.

Went up, through 3600, 3700, 3800, even 3900, all at 1.300V Bingo!

Couldn't get 4000 (4.0Ghz) stable, even with 1.325V or 1.35V. Oh well, looks like we have a freq limit somewhere between 3900 and 4000Mhz on my chip.

So, set to Frequency, Manual, 3900, voltage 1.3000V, I then selected XMP Profile 1, rebooted, it booted Windows, ran CPU-Z stress-test, actually overnight.

Note that it gives a benchmark score while running, which is around 1100 in my case, but when I woke up in the morning, it was around 120-130. Very lowered. Nothing looked like it was temp-throttled, like I assumed the non-heatsinked VRMs would be throttling the CPU, but nothing stood out. I stopped the test, waited a few seconds, and then started it again, and back to 1100. Weird.

So, I don't know if it temp-throttles longer-term at this OC (CPU temp only 63C, with the little rinky-dink small heatsink that ships with the Athlon 3000G, probably 65W, maybe 95W, it's not a Wraith cooler), or if it's just some sort of software bug running the CPU stress-test longer-term.

I haven't looped Heaven benchmark, nor tested any games on this PC, but it should play Fortnite OK, according to what I saw on YT. (100-120FPS)