[SOLVED] PC Ran fine for 6 months, suddenly won't boot past 2133Mhz

May 12, 2020
8
0
10
Hi everyone. I hope I can find some help here.
I have a Ryzen 5 2600, Gigabyte B450M S2H board, and Corsair Vengeance 3600mhz 8x2g RAM.

When I got my PC a few months back I enabled XMP and forgot all about it. Suddenly, two days ago, I was starting to get really bad stability issues, my computer would just die, no BSOD, no restart, it would just turn off and wouldn't turn back on until I unplug the PSU and put it back in.

The only stable way my PC would run is at stock bios settings - which is 2133mhz and all timings on auto.
I tried manually tuning with Ryzen DRAM Calculator for 2933Mhz, my bios wouldn't even boot - had to clear CMOS everytime.
I tried setting it at 2666Mhz 16-18-18-35 like the motherboard support list for memory says, and my computer crashes after about 10-30 mins of gaming.

Would appreciate any help, and especially would like to know why nothing happened for a few months until one day it started acting up?
I dont have the best specs out there and I can really see the FPS difference even between 2133 and 2666 on my ryzen.

thanks alot!
 

Ralston18

Titan
Moderator
Power down, unplug, open the case.

Clean out dust and debris.

Then reseat all cards, connectors, RAM, and jumpers to ensure that all are fully and firmly in place.

Look in Reliability History for errors that correspond with the crashes.

Make and model PSU: wattage, age, condition?
 
  • Like
Reactions: gg83
May 12, 2020
8
0
10
Power down, unplug, open the case.

Clean out dust and debris.

Then reseat all cards, connectors, RAM, and jumpers to ensure that all are fully and firmly in place.

Look in Reliability History for errors that correspond with the crashes.

Make and model PSU: wattage, age, condition?

Thanks for the response. I reseated my RAM and graphics card numerous times.
PSU is Antec Earthwatts Gold 550W
 
May 12, 2020
8
0
10
Any errors in Reliability History and Event Viewer?

Use the calculators in the following link to total up the wattage requirements of your build.

https://www.tomshardware.com/reviews/best-psus,4229.html

Your PSU may be nearing a premature EOL (End of Life) especially if it has been running at the high end of its' true power capabilities.

I used 2 calculators and both of them showed a max wattage consumption of no more than 380W for my system

As of now, my system isn't even stable running 2 sticks of 8GB anymore, even at 2133MHZ stock bios settings, even though I let my PC run memtest all night yesterday with both of them with no problem.

Right now I'm down to using only 1 stick of 8gb because thats the only way my system doesin't die on me :(
desperate for help. could this be a motherboard issue? could my PSU really be dying prematurely, a gold rated power supply?
 
May 12, 2020
8
0
10
Thank you for your reply. I have tried almost all memory configurations and timings - XMP settings from HWmonitor, xmp settings from Ryzen Dram calculator, Timings from gigabyte supported memory list, etc.

It used to run fine with both sticks in at 2133mhz and would be unstable above that. right now, I cant even use both sticks, as I said. Am currently running one stick at 2133mhz. By the way, if I try more than 1.2v for memory (reccomended is 1.35 for 2666+) the pc won't even post - just bricked until i reset CMOS.

Would a new CMOS battery help?
 
May 12, 2020
8
0
10
You may have PSU problems and what do your temps look like
Could it be my cpu or motherboard aswell? I heard alot about memory controllers dying on those.

about temps - I had it die on me even when both cpu and gpu were below 50c while gaming. temps are completely normal, CPU 65-70c under full load and GPU around 60c under load
 
D

Deleted member 14196

Guest
well, it could be mobo or cpu (cpu is very unlikely in most cases, they rarely fail--i have only seen 2 cpu die suddenly in my time) but we have to rule things out.

if you have a spare psu you could try it to rule out that aspect. But yes, it very well could be the motherboard as well.
 
What might have changed since all was well?

Possibly you have faulty ram.
Run memtest86.
It boots from a usb stick and does not use windows.
You can download the free edition here:
https://www.memtest86.com/download.htm

If you can run a full pass with NO errors, your ram should be ok.
If the test fails, try individual sticks at default settings to see if you can find a good stick.

Ram will have a lifetime warranty if it proves defective.

Many ryzen motherboards will have bios updates that address ram issues.
 
May 12, 2020
8
0
10
What might have changed since all was well?

Possibly you have faulty ram.
Run memtest86.
It boots from a usb stick and does not use windows.
You can download the free edition here:
https://www.memtest86.com/download.htm

If you can run a full pass with NO errors, your ram should be ok.
If the test fails, try individual sticks at default settings to see if you can find a good stick.

Ram will have a lifetime warranty if it proves defective.

Many ryzen motherboards will have bios updates that address ram issues.

Well i just completed a pass of memtest86 about 4 hours - no errors whatsoever.
what might have changed? biggest thing I could think of is the ambient weather getting alot hotter here. system temps still looking fine though.
I installed Valorant and it has some intrusive kernel based anti - cheat thats been making people have problems, even after deleting that it happends.
I would try reinstalling windows but considering the fact that PC wont even post with XMP on, i don't know if its driver or software related.
 
May 12, 2020
8
0
10
It turned out to the PSU. Plugged my friend's corsair vengeance, still no go. my memory worked great in his system.
1st boot with both sticks and xmp on with his psu went smooth and even did a few games like a champ.
Ordered myself a new seasonic 650w gold, so happy it ended up like this and not worse!