Does your motherboard have a little display? It might give an error code when the freeze happens. Seems like a hardware issue

Unplug the HDDs, remove the mobo from the case, place it on a non-conductive surface and try Linux again.

I have an update: I can boot the computer if there aren't any USB plugged into. If I plug anything, the peripheral works for a couple of seconds then the whole computer freezes. I've managed to get to my desktop by typing my password and quickly unplugging the keyboard. I also managed to open the Peripherals window, which told me there were several unrecognized peripherals plugged into my computer. Using the "Check for new Hardware" function or trying to update the drivers failed.
Current plan is to wait for my SSD and install a fresh Windows on it.

It does not afaik

Any news?

Hi,
My issue was solved very randomly, I found a couple of flashes for my BIOS, applied them, and it worked. I have no idea what caused it in the first place, when I was done with the flashes it was still giving me the same freeze. It randomly started working the next day, witchcraft basically.
Sorry for the late notice

Hi again,
When I thought my problems were actually fixed, it turns out they weren't. I no longer have the freeze issue, but my computer restarts at random times (the screen just goes black, and then returns to the bios screen).
I tried to find a pattern for when this actually happens, but couldn't. Some of the moments that are very likely to make my computer restart are listed below:
-opening a new program (although steam & mumble are 2 of the less likely programs to crash it)
-closing a program
-exiting a full screened video (this one is very odd but very consistent, if I leave a video in full screen for more than a minute my PC crashes the moment I press escape or return to desktop button)
All of the vents are spinning, CPU temps is around 45 degrees on average, Motherboard 30 degrees. I tried to take down the vent and re-apply thermal paste, didn't work.
I also tried to upgrade from Windows 8.1 to Windows 10, and the installation crashes during the final restart, while giving me the 0xC1900101 - 0x20017 "Install failed during the SAFE_OS phase with an error on the BOOT operator".
Looking for suggestions on how to fix this once and for all, any help is appreciated.
Thanks!

Have you tried only using one stick of RAM? Swap sticks if the problem persists and see if any stick of RAM works individually. Probably won't work, but worth a try.

Double check your PSU cables are plugged in correctly at both ends (PSU side and components). Also, if you didn't plug the AC wall cable in properly at the back it can cause problems on some PSUs from my experience. If you can use a known good power supply to test

Double check you plugged things into the motherboard correctly (fans, front usb, power/reset etc)

Double check your graphics card to see if it is plugged in all the way into the PCI-e slot and the power is plugged in properly too.

Just tried it, didn't work

Gonna properly double check everything

pls send help I want to scrim

Also upgrading to W10 usually breaks things so I'm not surprised.

First step is to figure out whether or not this is a hardware problem. See the quote above.

To be honest I think windows is just fucked. Boot manager is fucked and now it's even more confused with the new microcode.

Just for the record the Windows I use now is a non activated 8.1 version installed on the SSD I ordered after I figured something might have been up with my HDDs. i checked, everything is plugged correctly. I'm suspecting my motherboard to be faulty and I've ordered a spare one, I'll return the original one if it turns out to be the problem.

I will try tonight on a non-conductive surface with the SSD unplugged. Gonna give a try to Linux with the motherboard in the case too.

I see you're still avoiding double posts.

Haven't had time yet to get my motherboard out of my case and test it, although I built a Linux key already and ran it with the mobo still in the case: I don't get past the Ubutu loading screen.
So far my computer doesn't crash once I'm in TF2 unless the map changes so I just disconnect while the server is changing level.
A friend of mine offered me to borrow some of his HDDs to try and run W10 on it. The spare motherboard arrived but I'm waiting to get around the non-conductive test and my friends' HDDs to try to swap.

Found the strength to double post.
I just tried what you told me, taking out my mobo from the case, ran it with a Linux key. Direct HDMI feed from the motherboard was giving me no signal, I then plugged my GPU and a Displayport cable, got an image and managed to get to the Ubuntu loading screen but the same usual crash shortly after. No components were hot btw.
About to swap with the spare motherboard right now.
Edit: same issue with the spare motherboard (same model I probably should've tried with a different one but w/e)
Running out of ideas :(

There's no HDD/SSD plugged in, the mobo isn't shorting out and has been replaced and the PSU isn't exactly working hard during boot (not like a G3 is going to crap itself like that anyway) which means there's only 3 components left.
1. The CPU. Obviously difficult to test.
2. The RAM. Run memtest, maybe just one stick, etc., maybe replace it if you can borrow some.
3. The GPU. Sadly you've got no integrated GPU (that's why you're not getting any signal with HDMI plugged into the mobo) so you'll need a different GPU to test it. Or test your GPU in a different PC.

I ran memtest86 for 9 hours the other night and had no errors. I can probably find some other sticks. I'll give my GPU to someone to try out. If all of this comes out negative I guess I'll just buy another CPU and eat ice cubes for the rest of the month..

Yeah, my money is on the GPU then.

I'd expect quite a few people to upgrade to Ryzen 3000 CPUs, so if there's any way I would take advantage of the market being flooded with used Ryzen 1000/2000 next month. Even if you really can't wait you could abuse return policies. Buy a new CPU, wait for the launch, buy a used CPU and return the new one.

Hi,
The new GPU has arrived. After I installed it, I did not get the "No signal" screen anymore, however by trying to boot on a linux key with the motheboard on a non-conductive surface it first failed and then showed me this screen: http://fr.tinypic.com/r/15wf1as/9
Any idea on what it could mean?

Is that new? I mean it's always possible that there was a combination of issues.

But against all odds it seems that the CPU is FUBAR.
I mean if I read this correctly it's "Bank 6: baa0000000000118" which according to google has never happened before and according to the MCE decoder means
Bank: Floating Point Unit (FP)
Error: Physical register file (PRF) parity error (PRF 0x0)
You shouldn't even have to know what any of those words mean because none of this should ever break.

Bank 5: bea0000000000108 is fairly well known and fixed by disabling C-states, so you could try that but it shouldn't be that.
I think that's one for AMD support. It just screams warranty/DOA.

First time I've seen this screen, however it flashed extremely quickly and I easily could've missed it during testing, it took me several tries to get the picture.
I'm gonna follow your previous advice and use return policies too in order to get something up & running asap, I don't want to play an official on a laptop ever again.
As well as contacting AMD support ofc.

Hi, finally some good news.
Turns out my CPU was indeed faulty all along, as soon as I swapped it fixed everything. I'll return what I don't need asap and contact AMD. Thanks for helping me through this Setsul, think I owe you a drink next lan.

Glad it works now.
Sorry that took so long, but it was the least likely thing to happen.

I won't be at LAN this year so don't worry about it. Maybe Eemes needs a drink after all the missed scrims.