remember that your PSU should have a max wattage of at least 40 to 50% extra on top of what your PC will consume normally. But to me it could also be extreme CPU temps, or in some cases, low CPU vcore directly shutting the computer instead of sending you a BSOD
PSU might have some other fault than lacking power, MB could cause issue, but usually those have been freezing or blue screening, this is abrupt power cut off without a warning and very rare actually, so easiest thing to do would be getting another PSU, also it might be good to have a backup PSU just in case, so that would not be wasted effort, but yeah, don't need huge power with this system, just a little overkill to keep PSU fan quiet.
Have you RAN memtest 86, something like Asusrealbench, or some GPU benchmark? If you are getting a new PSU get a upgrade, not a downgrade like a Evga B1/W1 or a sidegrade like a Corsair CXM/CX. Get something like a Seasonic Focus, Evga G3, BitFenix Whisper M, or a Corsair RMx.
Yeah, quite lot of tests, for example Catzilla puts quite good stress on CPU and GPU, hours of pi and memtest, no problems with those, it's only BeamNG and usually hour or two after really pushing computer to limit for hours in BeamNG. All I get is relay in PSU says click and power disappears. This is how Vcore looks under heavy load, it is fluctuating a bit, but not too much I think? You might need to right click and choose view image to see it well: Here is link to my current PSU now that I remembered what it was https://www.anandtech.com/show/3962/seasonic-x460fl-460w-fanless upgrading from this is going to cost quite a bit I believe.
Its normal for voltage to fluctuate some. I thought it was based on the M12II/S12II platform due to their slientnight jar or whatever it was called, not the KM3 platform so even a CXM would be a downgrade. Yeah its not going to be cheap to replace. You are looking at around ~$80 to ~$100.
Updated Bios as there was finally new version available, Vcore behaves quite differently now, first part is under load, then it is less load and no load. There was mentioning about Vcore being fixed as well as HT being fixed as hotfix to one that fixed Vcore. Now it adjusts according to load. It can take few months for issue to show up again, it is kinda hard to tell if bios update has any effect.
Another thing you should check if you haven't already is that the CPU and 24pin power connectors aren't loose, both at the power supply and motherboard, you should also give the connectors at the power supply a wiggle, it isn't unheard of for the output sockets on modular power supplies to develop bad solder joints.
That's why I mentioned a bad solder joint here. Heck it could even be something off-the-wall like needing a BIOS update (forgot to mention that - and here you mentioned that). Could most certainly be something like that. Let's hope you don't have to throw money at it.
Have to knock wood, but so far I haven't been able to make shutdown issue happen again. How on earth it would be that BIOS update could fix such issue? Let's see if issue now stays away, but so far looking good.
Simple, the BIOS controls the VRM, your processor was getting 0 volts for a split-second. Usually, the part of the BIOS code that controls the VRM for the CPU also controls the VRM for the RAM, too. Cutting the power to the RAM for just a split microsecond causes a RESET. Cutting power to the CPU causes power off usually (among other things). Chances are, it was getting a condition that was 'crashing' that part of the bios microcode. Be it a certain power supply brand, local power conditions, board manufacture variations by chance (or by component changes on revisions), or something else, could be not 'getting along' with the board and making for non-optimum conditions (that's why I suggest power supply, but let's just leave it sit and test for now). So sit tight Mr. Guinea Pig (or Lab Rat), and test it out. Give it hell. Make sure to do it now while it's still all under warranty, heh. You know how that goes. So yeah, this is how it goes, it could be any number of things, let's just hope it's that. It's likely an odd-ball condition that didn't get noticed until the board was in the real world, with a certain combo of components, causing your issues. So that could have been fixed, or made less likely, or not been the problem at all (and it could still come up). I don't know the answer to if it fixed it 100% or not, but it certainly could have, it's 50/50. --Good Luck! I hope it's fixed for you bud. You're one of the good guys here. EDIT: Layman's terms here: VRM = Voltage Regulation Module, these drop the 12V down to 1.x V for your processor and about 1.2~1.35 for your RAM. There's usually 3~8 for the CPU (depending on how much motherboard you're willing to pay for) and 1 or 2 for the RAM itself. These are the little boxes and little flat square things (and capacitors too!) over by the Processor and the RAM on the top-end of most motherboards. The more motherboard you pay for, the more you get and the better quality they are. These are also the little bastards that can introduce COIL WHINE into the PC (that high pitched screech on 3d video use or even sometimes moving the mouse etc), usually happens via Nvidia Video cards, though can happen with anything that has them (power supplies, motherboards, video cards). This is why BOB has RADEON brand cards, because that screech is like 5x louder to me VS other people, and the RADEON's don't do it nearly as often as Nvidia Geforce does. It was so bad on my PC 10 years ago you could hear it 40 feet away AND upstairs! Custom GPU PCB's often do it LESS OFTEN, too, for what it's worth... *end rant*
I don't know if there really is 0V, because as long as I remember those voltages with all hardware I have owned has had those split second 0V readings, which I think are just flukes in code that pulls the data or something. It's only MB section of voltages that gets these flukes though, which are still happening. CPU section shows Vcore in more sensible range than this MB section. Coil whine is indeed annoying, did bit of research on that before ended up with Gigabyte, what I found out was that these versions from Gigabyte did have least of the coil whine. Seasonic PSU however tends to be bit of rodent in that regard, not too bad now with good power from the wall, but surely enough there is some buzz, especially after I have put computer running it's paces for a while.