Alright, So I have an issue after a completely new PC build

[jdkzombie]

The only thing I have from my old PC is my HDD, and I am running into a new issue that I have never experienced on any computer I have ever owned.

 

 

A red screen of death, or a complete system shutdown during and only during Fallout 4. I can run a Passmark Bench and Cinebench, and OCCT for over an hour and hit 260 FPS with no issues at all on either power supply, or graphics, or processor.

 

Build specs are:

Ryzen 1800x stock clock with Liquid cooling

RX 460 8Gb Gfx stock clock

x370 Gaming K4 Mobo

Anarchy 16Gb 2400 memory DDR4 DIMM

EVGA 1000w Supernova G3 psu

 

HDD is a Velociraptor 1Tb 10k rpm drive, Windows 10 Pro.

 

Old build was:

FX-8350 liquid cooled at 5.0Ghz OC

R9 390 8Gb 1500Mhz clock

G Skill 16Gb DDR3 DIMM iirc freq was 1633? the lowest of the low whatever that is, 1333 maybe?

Themaltake 800w Bronze psu

MSi 970 Gaming Mobo

 

 

What I'm thinking is possibly the power supply may be bad right out of the box, but it does this at no other time. I wondered if the graphics card was acting up, but you'd figure a hardcore hour plus stress test would find issues, temps never exceeded CPU at 34c and Card at 47c, cool enough you could keep your hand on the metal back plate even.

 

I did a fresh driver install last night, so we shall see whats up, but I changed the Win10 advanced system settings to prevent crashes from automatically restarting the pc in hopes that maybe it would chug through the issue, so now instead of just rebooting with no error, or beep, or message, or anything, it hit a red screen once, and that was enough to warrant me wiping all AMD drivers and reinstalling fresh, updating my BIOS, and reseating the GFX card, making sure the 6+2 power connector was properly seated, and even removing the processor cooling block, and processor from the board, cleaning all thermal paste off and reseating and applying new paste in hopes that maybe there was a bubble or something allowing a hot spot to form.

 

I know that a red screen is usually a heat issue, "usually", so I've taken steps, but I really hope I don't have to RMA with Newegg on a Gfx and PSU....that would be very annoying lol

[MissingMeshTV]

I was going to say it sounded like a possible overheating issue. Maybe yor fan speeds need to be adjusted? There is a thread a few below this one where overheating was the culprit...maybe the last post has info that could help...

 

https://forums.nexusmods.com/index.php?/topic/4740170-multicoloredrainbow-texture-bug/page-2&do=findComment&comment=49242412

[jdkzombie]

The weirdest part is this only happens on THIS game.

 

So I wondered if maybe a recent patch hit and f*#@ed things up or not, and how to roll back to a previous patch.

[Fonger]

Install Armored Warfare - no need for the HD version, the SD will do

 

set your graphics to almost anything

 

then let it idle in the garage

 

if you don't have similar issues, its something to do with this game

 

if it shuts you down - its the PSU

[jdkzombie]

Alright so I found what the issue was for me anyways.

 

Apparently the ASRock x370 Gaming K4 motherboard can only have certain frequency RAM modules in certain RAM channels on the board.

 

 

A1 A2 B1 B2

 

You can use 1333-1600 in all four, or 1600-2400 in A1 and B1, any higher, like 2666 can only be ran in slots A2 and B2.

 

I found this after I spent 168 bucks with geek squad and they didnt find anything wrong with any of my other stuff, I came home, started up Fallout, made it 5 minutes and it Red Screened on me, total Red out, no text or anything, rebooted and immediately Blue Screened with a Memory Management failure on it, which led me to dig deeper into the memory, and find my solution.

 

How this pertained to Fallout was through ENB. Enblocal was configured to use all 16Gb of system memory to allocate if necessary plus the 8Gb on the Graphics card, well since my Motherboard was registering that I had all 4 slots used, but was trying to dial down 2 different slots to its acceptable speed, it kept crashing from memory management error, so I dropped to 16Gb memory, and changed RAM channels, and voila, haven't had a single issue at all with it crashing. I had foolishly set Enblocal to try and use all 40Gb of memory I had available and it literally couldn't I guess. That, and never use PNY memory. Those sticks are gone, switched to Corsair 2666 Vengeance DDR4 instead.