Jump to content

Unstable AMD 3900x in Windows and Boot

Simon Bruun
Go to solution Solved by Radium_Angel,
Just now, Simon Bruun said:

I've attached the 4 minidumps.

Step 1: Reset everything in the BIOS to default.

Step 2: Test that way.

Step 3: Use memtest86 to test RAM if it's still giving you problems.

Step 4: report back.

Hi

 

I build a new AMD PC back in October 2019. 

I had a little trouble getting the build op and running stable, fiddling mostly with Memory settings in UEFI. But i felt like i did get to run pretty stable.

With a few hiccups it has been running really good for about 2 months. But yesterday the PC became very unstable. 

 

In windows it either just shuts down or gives a BSOD. 

After a crash it usually also have problems booting again, powering on and off a few times before starting on safe mode, and i need to press F1 to go change some settings.  

The crashes can occur at any point in time, at idle, or at load (gaming or rendering), I haven't found any sign of a pattern in the crashes. 

 

I have a feeling it my problem is somewhere between the CPU and the Memory. and the frequency and Latency settings. 

So I'm looking for help to either getting my system stable again, or help figuring out if one or more of my Components should be RMA'ed. 

 

PC:

  • Windows 10 Pro for Workstations
  • Asus x570 ROG CROSSHAIR VIII HERO (BIOS version 1001, latest according to EZ Update software)
  • AMD 3900x at stock speed 
  • 4x16GB G.Skill TridentZ Neo F4-3600C16Q-64GTZN (God! there is a lot of settings around RAM. This is where i've been trying to debug the most, and atm. i think i'm at 1600 22-22-22-53. )
  • Zotac RTX 2080Ti
  • Samsung nvme SSD 970 1TB boot drive
  • 4 x Samsung SSD 850 EVO 1TB storage drives 
  • Corsair RM850x, 850W 

 

I'm not a 100% sure what my Memory setting where for the two stable months, but i'm pretty sure it was running 3600 16-16-16-36. 

 

I've attached the 4 minidumps.

 

 

 

MiniDumps.zip

Link to comment
Share on other sites

Link to post
Share on other sites

Just now, Simon Bruun said:

I've attached the 4 minidumps.

Step 1: Reset everything in the BIOS to default.

Step 2: Test that way.

Step 3: Use memtest86 to test RAM if it's still giving you problems.

Step 4: report back.

NOTE: I no longer frequent this site. If you really need help, PM/DM me and my e.mail will alert me. 

Link to comment
Share on other sites

Link to post
Share on other sites

Running memtest atm, It is taking a while.

It failed the first memtest run, this was before going back to defaults. 

There was an error and the machine froze, so it never completed the first run.

It has been running for longer now after the defaults. 

 

I have screenshots I'll post when i get my PC back in my hands.

 

 

I tried different voltages, don't renemmber what it is right now. I'll check after memtest completes

 

Manged to attach the failed run screenshot from mobile.

IMG_20200107_224057.jpg

Link to comment
Share on other sites

Link to post
Share on other sites

Oh man, just realized that it is still only on pass 1 of 4.. 

 

I guess I'll leave it overnight, 

 

I'm at 92% on the first run. This has taken 57 minutes. And No errors so far.

Link to comment
Share on other sites

Link to post
Share on other sites

After changing UEFI to "Optimized Defaults" I've been able to run MemTest for 9½ Hours. But haven't had time to test windows idle for long, or any games. i'll have to do that later today.

 

dram voltage looks like it is set at 1.2V

 

If i read Ryzen Master or CPU-Z correctly, i looks like the timings are 15-15-15-36 running at 1064 Mhz.

 

 

Memtest9Hours.jpg

UEFIDRAM.jpg

CPU-Z.png

Link to comment
Share on other sites

Link to post
Share on other sites

I've been using windows for an hour without a crash, and booting is going fine too, seems like the defaults made the system more stable.

 

Can't help but feeling a little bad about having RAM rated for something they can't handle, or the CPU can't handle.

Also it looks like the "Infinity frabric" is not in sync with mem speed, witch I've read loads about being a bad idea. But hey if this is stable, I'll roll with it for now. 

 

Thanks for the help so far.

FaricClock.png

Link to comment
Share on other sites

Link to post
Share on other sites

I would say things have gotten much more stable after the defaults, but still had a few crashes.

 

I did monitor my temps yesterday and they do go pretty high when i put stress on it, like gaming, it is around 85-90C 

So i ordered my first AIO cooler. 

This one was on offer: Cooler Master MasterLiquid ML360P Silver 

I liked the look of it, and it should do way better then the stock fan cooler i'm running with now. 

Unless you guys know something particular bad with this cooler, I didn't plan on going into a longer discussion on what cooler to choose. 

I did consider coolers when I made this build, but after some reviews I got the feeling that the stock cooler for 3900x was pretty decent for a non-overclocked build. but of cause not nearly as good as a 3rd party cooler. 

 

Today it just crashed again (minidump attached), but for the first time the POST screen said CPU Temps were high - see screenshot

So fingers crossed all my problems have been temperatures. 

 

One thing is bothering me on this last crash, I wasn't stressing my PC, just you-tubing.

And last night i was gaming for quite a while with no crashes, but with high temps. 

 

Right now all i'm doing is writing this message, and my Temp is 63C

 

Do you guys think it is my temperatures, or something else ?

IMG_20200111_095149.jpg

011120-13593-01.zip

Link to comment
Share on other sites

Link to post
Share on other sites

Just installed the new AIO cooler, sad to say it doesn't look like it helped. I just had a similar random crash

 

It did help with my temperatures tho. So I'm pretty sure it not my CPU overheating. 

 

I've attached two new miniDumps. 

 

I'm back at blaming either Faulty CPU or RAM. 

 

Think i'm gonna try removing some RAM Sticks, and or reinstalling windows. 

 

Any suggestions ?

MiniDumps.zip

Link to comment
Share on other sites

Link to post
Share on other sites

Uptime of almost 4 hours with only 2 sticks of memory - in other words, no crashes so far with only 2 sticks. 

 

 

Link to comment
Share on other sites

Link to post
Share on other sites

Pushed my luck, and set the the two sticks of RAM to D.O.C.P.

It booted without a problem, and now running at 3600-16-16-16-36.

 

I'm going to do more testing tomorrow. and i'll try swapping the sticks with the ones i've taken out. 

Fingers crossed it will crash with those two, so i can RMA.

Link to comment
Share on other sites

Link to post
Share on other sites

After running mem test all night with only two sticks it completed after 6 hours. 

 

I switched to the two other sticks (in the same mobo slots) and the system became instantly unstable.

 

Then I tried running with only one stick, and that worked fine. 

Switched the other single stick, and problems booting again. 

 

I'm ready to conclude that all of this is caused by a faulty mem stick.

At least i really hope this is the problem, so i can get this behind me :D

Link to comment
Share on other sites

Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now

×