Question All games crashing... Completely out of ideas

[eric1906]

Hello everyone,

Never thought I would someday post on a forum about hardware issues as I'm quite experienced, but this one drives crazy.
So about a year ago all my games started crashing randomly. Sometimes came with an error message (Unreal Engine games would either report a GPU HUNG or memory error), but most games will actually just close and nothing more.

My initial build was an i7 8700K, 32 gb of ram (4*8), and an RTX 3080, 850W corsair psu.

I changed EVERY SINGLE component, one by one. Changed the motherboard twice, and then the cpu. RMA'd my gpu, got a new one. Bought another couple of memory sticks. Bought two new nvmes and installed a clean Windows 10.
Changed my PSU too.

Now my setup has a Ryzen 5 5600X, 32 gb of ram (not the same sticks), an RTX 3080 Ti, another new 850W psu, two new NVMe drives.

I am left with not a single component from my original build and I still get the exact same crashes...

When I put my old GTX 1080 Ti, everything works flawlessly... But I would believe it is mathematically impossible to have 2 faulty GPUs in a row from the factory.

No overheating ; HWiNFO shows some thermal limitations though, but even when downclocked (-500Mhz to both gpu and memory clocks) same issue... XMP enabled too.

I spent roughly 1k€ to replace every single component but to no avail...

Any idea is welcome...

Thanks!

 

[Roland Of Gilead]

Hello everyone,

Never thought I would someday post on a forum about hardware issues as I'm quite experienced, but this one drives crazy.
So about a year ago all my games started crashing randomly. Sometimes came with an error message (Unreal Engine games would either report a GPU HUNG or memory error), but most games will actually just close and nothing more.

My initial build was an i7 8700K, 32 gb of ram (4*8), and an RTX 3080, 850W corsair psu.

I changed EVERY SINGLE component, one by one. Changed the motherboard twice, and then the cpu. RMA'd my gpu, got a new one. Bought another couple of memory sticks. Bought two new nvmes and installed a clean Windows 10.
Changed my PSU too.

Now my setup has a Ryzen 5 5600X, 32 gb of ram (not the same sticks), an RTX 3080 Ti, another new 850W psu, two new NVMe drives.

I am left with not a single component from my original build and I still get the exact same crashes...

When I put my old GTX 1080 Ti, everything works flawlessly... But I would believe it is mathematically impossible to have 2 faulty GPUs in a row from the factory.

No overheating ; HWiNFO shows some thermal limitations though, but even when downclocked (-500Mhz to both gpu and memory clocks) same issue... XMP enabled too.

I spent roughly 1k€ to replace every single component but to no avail...

Any idea is welcome...

Thanks!

Hey there,

For a start, I'd only use ram that you have in a 2 x 8 kit. Also, Ryzen systems like 2 x kits. 4 x kits sometimes cause issues with compatibility, and also more stress on the memory controller. Mixing ram is not advisable. It can cause random crashing as you are epxeriencing.

With that said, let us know what Bios you are running. If it's not up to date, this also could be the issue.

 

[eric1906]

Thank you for your reply

I never mixed rams. I have a 2x16Go kit and a 4x8 kit. I tried everything RAM-wise: a single 8Go, a single 16Go, and every possible 2x8Go combinations.
Also it has nothing to do with Ryzen not liking it, since I had the same issue with my i7 8700K.
I'll perform a bios update but again, I'm sure its not the issue since I tried with 3 different motherboards.