I have not tested manual clocks in BIOS. Curve Optimizer tuning is more in line with what I want to achieve, since all manual results in lower clockspeeds than the maximum possible single core turbo speed achievable. CTR of course aims to remedy this by switching off the all core overclock when low CPU utilization is detected (similar to Hybrid OC on the Asus Dark Hero, but with different detection methods), but another aim of Curve Optimizer is that it can potentially reduce power consumption even after hitting max turbo clocks. CTR will only support Curve Optimizer in version 2.1.

Could be anything. Which USB ports are you using (USB3.1 Gen2/Gen1/USB 2.0), motherboard model, BIOS version?

There's also a separate LAN issue on some B550 motherboards, but IIRC it's due to Intel 2.5Gbit/s LAN as similar reports are also out for Z490.

As you can see in the post, small sample of people... usually affected those power users who max loaded the USB ports, like me is a racing sim rig and an audio interface. Many cases also due VR. There are cases of people running multuple external HDD also kena. And also streamers who uses multiple cameras.

A quick google and it’s actually quite a huge sample size and had been happening for quite a long while.

Finally have the time to compare the two systems.

About that CPU PPT wattage:

5950x
[attachmentid=10798468]

3950x
[attachmentid=10798467]
Why in the world CPU SOC on 5950x X570 uses around 10 watts more than 3950x B550 at idle?  :confused:

Not to mention 3950x seems to idle at 3400mhz and 5950x idle at 3600mhz?

Edit: The temp doesn't lie, 3950x is indeed more cooler than 5950x. More or less same coolers and in the same room. PBO + OC had been disabled, set to default.

X570i meanwhile is very known to melt... .

https://www.reddit.com/r/Amd/comments/cd6ze...t_temperatures/
https://forums.tomshardware.com/threads/saf...rature.3531213/
https://hardforum.com/threads/safe-x570-chi...oncern.2000597/

basically it will always be above 50C even on ATX in cold ountries. on ITX, easily 70-90C at load depending on airflow... sandwich one more gen4 nvme under the shared heatsink then 100C . That's why x570i only got gigabyte, asrock and asus which are all not being recommended. MSI didnt even bother to make an ITX version.. and ASUS created the DTX form factor just to beef up the cooling even more.

You rarely see it in the itx community because majority knows of this problem and got the X470i/B450i instead for ITX builds... and now use B550...especially some B550i model have even beefier vrm than the x570.

I think you're taking the word "melting" too literally 
Bonchi is probably using "melting" as a figure of speech signifying too damn hot only 

Are you using the NVME slots linked to the chipset?

Melting is just an expression of it running really hot.. and by means anything above 60C is considered hot, 70-80 is really hot. Melting does not mean breaking down but literally means higher than comfort.

It's the same as to any electric components as we all want to keep it lower than 60C vrms included.

Then, Max Boost CPU Override to 200Mhz.

If you find the CO first, then later put override, it may work but it wont be 100% stable.

The voltage requirements gets higher as frequency increases in a non-linear way. Even if the CO adjusts this v-f curve non-linearly (as it is supposed to be, which is why CO is better than the classic vcore offset), using a CO setting which is stable at 0MHz offset may still not provide enough voltage for a 200Mhz offset.

Yes that WHEA ID 18 event log is useful, but you will still have a risk of data corruption since there may be some read-write operations which is happening when the system crashes with that WHEA error.

I use the corecycler script to check my individual core CO stability. Would take days, if not a couple of weeks to get fully stable, but CO stability testing is less time consuming than fully stability testing RAM OC which is a whole different game IMHO (at least for my case with a 5600x 6 cores only)