Hello folks; I was sent this way from Reddit, with the suggestion that you smart people might be able to help me troubleshoot my new PC somewhat. So... here's some backstory for you first:

I recently purchased myself a new machine, Ryzen 1700, X370 Taichi, G.Skill DDR [16gb rated @ 3200 CL14], Silverstone 1000w PSU and a Vega64.

Whenever it would spin up any intense gpu-usage situation I get some awesome (-ly bad) coil whine, and more recently I've been getting an increasing number of hard crashes.

I assumed it was a bad Vega, so I did the logical things; driver updates, driver downgrades, underclocked, stock clocked, etc... nothing. Eventually I retrieved my old R9 290 and swapped them out. It seemed better, but after a while it proved to be unsuccessful; the crashes & coil whine returned.

I was then suggested to use a tool called OCCT to 'test the power supply'; the outcome of that is a rather interesting discovery:

-- my +12v rail is seeing severe voltage drops of 7+ volts. My +5v rail was also seeing similar behaviour ("only" about 2.6 volts there though)

At this point I assumed, "nasty broken psu!" - disassembled my machine, swapped in my older one, and .... I'm back to crashing... no coil whine though! And, to top it all off, I am still seeing the +12v rail dropping to 5v whenever any real load is put on the machine.

I'm using HWiNFO64 to read all this data, but various other tools are pretty much giving me the same output [ie: whatever the motherboard tells them]. (In case it matters, I'm using Win10 x64 latest updates etc)

So... question now is; Are both my PSUs broken? Or is it both my GPUs? All four devices!?... Motherboard? ... something else ? -- I work from home, so down time without my machine is extremely not good, thus I'd like to find some sort of logic to what I should RMA if I'm forced to go that route. Or, if at all possible, some "go press X in bios and fix everything" suggestion (though obviously I'm not expecting that!)