Overclock.net banner
Cancel
Create thread New Forums

New Ryzen Build Freezing then Restarting on Windows 10 x64 [SOLVED]

Jump to Latest
1 - 18 of 18 Posts
S

SkillzKillz

· Registered
Joined
·
4,319 Posts
Discussion starter · #1 ·
My new build is:
  • AMD Ryzen 5600x with stock cooling (purchased used)
  • MSI B550 Gaming Plus (purchased used)
  • 32gb (2x16) Kingston Fury Beast 3600mhz CL18 (purchased new)
  • Sapphire RX 570 4GB Nitro+ (old rig)
  • WD Black SN750 500gb M2. SSD (purchased new)
  • Seasonic 750W Gold (old rig, 2012)
  • Windows 10 Pro x64 (my first time on W10)
  • 3 HDDs, WD/Seagate = 8TB total (old rig)
  • No overclocking
Randomly in the day—sometimes when I'm doing something, sometimes when PC is idle—my computer will freeze, lock up for about 6-8 seconds, then reboot. This happens about twice per day. No error messages have come up. Sometimes right before it freezes it stutters a bit.

Troubleshooting (tried):
  1. Installing Windows Updates
  2. Repairing Windows (doesn't let me)
  3. Passed MEMTEST86, two full passes; 14 tests, over 2 hours
  4. CHKDSK no errors
  5. Rolled back Radeon drivers to Dec 2020 and tried 4 different drivers throughout 2021
  6. Turned off SysMain / Superfetch
  7. Put pagefile to 2x my RAM (64GB) or let system manage
  8. Power Options: sleep = never, turned off PCI Link State Management, turn off hard drive = 999999 minutes
  9. Updated motherboard drivers
  10. CPU Temps highest they reached is 69C. Nothing is overheating; GPU is about 55, everything else is under 40.
  11. Passes Prime95 and Furmark
I've tried so much I'm at a total loss of what to do.
 
#2 ·
We have had lots of reports of 5600X not being quite stable at stock settings using 1800 FCLK / 3600 memory. :(

I suggest a small increase in the SOC voltage, perhaps tweaking VDDG and/or VDDP as well.
If that doesn't help, a small positive offset to the CPU voltage might. Put the SOC voltages back to stock first, at least for testing.

SkillzKillz said:
Power Options: sleep = never, turned off PCI Link State Management, turn off hard drive = 999999 minutes
Click to expand...
0 also disables hard drive power down. :)
 
Discussion starter · #3 · (Edited)
Asmodian said:
We have had lots of reports of 5600X not being quite stable at stock settings using 1800 FCLK / 3600 memory. :(

I suggest a small increase in the SOC voltage, perhaps tweaking VDDG and/or VDDP as well.
If that doesn't help, a small positive offset to the CPU voltage might. Put the SOC voltages back to stock first, at least for testing.



0 also disables hard drive power down. :)
Click to expand...
Can I just turn off XMP and run 2400mhz to troubleshoot? In the meantime, I will try SOC at 1.1v as it's been running at 1.056

Haha thanks, I saw a pic online of someone's power setting as "off" but mine can't do that and I was confused lol.
 
Discussion starter · #4 · (Edited)
Tried and failed:

12. Updated MSI bios, v15 16 17 don't work.
13. Uninstalled all video drivers
14. Updated chipset to latest on AMD's site
15. Manually set:
  • DRAM: 3600mhz 18-22-22 at 1.4v
  • SOC: 1.1v
  • VDDP: 1.0v
  • VGGD: 1.0v
  • 1800mhz FCLK
  • Vcore [AUTO]: range of 0.234 - 1.422
*This was the first thing that made a difference. Instead of freezing, locking up for 8 seconds, then rebooting, this time the PC slowed way down, some software stopped responding, but Windows File Explorer and HWMonitor were working perfectly normal & I did a manual reboot. Maybe this is evidence of a hardware problem or compatibility issue?

Failed (9/24 - 9/29):
16. Currently testing MSI bios beta v18 => AMD ComboAM4PIV2 1.2.0.3c = AGESA 1.2.0.3c (pass)
17. Running 1 stick of RAM at stock 2400mhz (single channel in CPU-Z, 1200mhz) (pass)
18. Unplugged all HDDs, only running M.2 boot SSD (pass)
19. Changed "Turn off Display" to "never" in power options (pass)

New:
20. Using ethernet instead of wireless adapter
21. New surge protector
22. New wall outlet
23. One monitor instead of two
24. No speakers (RealTek Audio Drive may cause crashes)

If fail, next steps:
  • Safe mode w/ networking & ethernet
  • Fresh install windows
  • Sub for friend's PSU
  • Sub for friend's RAM
  • Install after-market cooling and play around with dozens of voltage & speed settings
  • Smash it with a baseball bat and set it on fire
  • Ask MSI and Kingston for help, despite not being on QVL
The worst part out of all of this is that I usually have to wait 6-12h before it fails every time I make a change. This has been ongoing for 2 weeks.

EDIT 9/26: Above "new" changes passed 24h without crashing... a new record as I always crash between 6h-10h. I've since added back RAM, running two sticks @ 3600mhz XMP.

EDIT 9/30: Got 64h stable with changes on 9/24-9/29. HUGE progress! But crashed after moving back to the other floor. Continuing troubleshooting. It's down to dual monitors, or wireless internet, or speakers/audio drivers.
 
#5 ·
SkillzKillz said:
Tried and failed:

12. Updated MSI bios, v15 16 17 don't work.
13. Uninstalled all video drivers
14. Updated chipset to latest on AMD's site
15. Manually set:
  • DRAM: 3600mhz 18-22-22 at 1.4v
  • SOC: 1.1v
  • VDDP: 1.0v
  • VGGD: 1.0v
  • 1800mhz FCLK
  • Vcore [AUTO]: range of 0.234 - 1.422
*This was the first thing that made a difference. Instead of freezing, locking up for 8 seconds, then rebooting, this time the PC slowed way down, some software stopped responding, but Windows File Explorer and HWMonitor were working perfectly normal & I did a manual reboot. Maybe this is evidence of a hardware problem or compatibility issue?

New:
16. Currently testing bios beta v18
17. Running 1 stick of RAM at stock 2400mhz
18. Unplugged all HDDs, only running M.2 boot SSD

If fail, next steps:
  • Safe mode w/ networking & ethernet
  • Fresh install windows
  • Sub for friend's PSU
  • Sub for friend's RAM
  • Install after-market cooling and play around with dozens of voltage & speed settings
  • Smash it with a baseball bat and set it on fire
  • Ask MSI and Kingston for help, despite not being on QVL
The worst part out of all of this is that I usually have to wait 6-12h before it fails every time I make a change. This has been ongoing for 2 weeks.
Click to expand...
Did you checked the Windows event viewer, did it report WHEA errors when it crashes?

To be honest, you are checking all the box of a bad or weak cpu, as we keep seeing posts and threads like yours, as soon the Ryzen 5000 series launched.
If you did diligently all the test, like the bios up to date, checked RAM and SOC voltage, tried w/o XMP, and you're still getting reboot at idle, i would RMA the cpu.

Be sure to record a video of the bad behaviour with the reboots at idle and the WHEA, if any, because the AMD support will ask you which test you performed.
Usually, when users also showed a video of the issues, the AMD support accepted the RAM without further questions.

Another clue of a weak cpu, is USB deconnections and drop out on the peripherals, you can try to disable the C-STATES and switch to PCI-E Gen3.
If this solves or alleviate even partially your issues, i would definitively RMA the cpu, you are not the first nor will be the last with these kinds of issues.
 
#6 ·
I feel like you are getting ton loads of WHEA errors down there, memtest86 doesn't prove anything close to the term "System Stability" it may give you a thumbs up only on the memory department but elsewhere anything can happen..I suggest to use everything as it is (stock) and try to find out where are your issues, from there, once you have been able to find the culprit, then you can gradually start tuning your system..

AMD isn't for the people who don't want to spend time with the platform, its really going to be time consuming taming it, you can enjoy it as it is, but the potential is lost there specially if you bought the high end parts..
 
#7 ·
You are running a almost 10 year old psu... Double check 12v rail etc..
 
Save
Reply Quote Rep+ Rep+
Discussion starter · #8 · (Edited)
1devomer said:
Did you checked the Windows event viewer, did it report WHEA errors when it crashes?

To be honest, you are checking all the box of a bad or weak cpu, as we keep seeing posts and threads like yours, as soon the Ryzen 5000 series launched.
If you did diligently all the test, like the bios up to date, checked RAM and SOC voltage, tried w/o XMP, and you're still getting reboot at idle, i would RMA the cpu.

Be sure to record a video of the bad behaviour with the reboots at idle and the WHEA, if any, because the AMD support will ask you which test you performed.
Usually, when users also showed a video of the issues, the AMD support accepted the RAM without further questions.

Another clue of a weak cpu, is USB deconnections and drop out on the peripherals, you can try to disable the C-STATES and switch to PCI-E Gen3.
If this solves or alleviate even partially your issues, i would definitively RMA the cpu, you are not the first nor will be the last with these kinds of issues.
Click to expand...
With my last changes (16. 17. 18.) I've been running for over 14h, which is the longest I've ever run on this setup. Could be dumb luck, who knows. Other things that have changed—because I put desktop on another floor— 19. ethernet instead of wireless 20. different surge protector 21. different wall outlet. So I guess that's 6 things that are different and this PC hasn't crashed... yet. I will leave it on overnight- if it will crash that will surely do it.

I'm really hoping the August Beta BIOS has fixed it. This might be a compatibiilty issue of having Kingston's new RAM that launched on July 29.

Right now I only have a few seconds video showing my cursor locked up and the screen going black for the reboot. It's otherwise impossible to catch any other action because it happens so fast. I don't know how to trigger it either. Oh but sometimes I can go like 8h with no problem, then it will freeze & reboot and within 5 minutes of the startup it will freeze and reboot again. Then it's good for another 6h-12h. No USB issues.

My 5600x was made in 2020, week 50. The previous owner purchased it in February.

In the RMA process, 1) if I can't narrow down what the problem is... which component manufacturer do I contact for support? 2) If I RMA a component, but they verify that it functions properly, what happens?

kairi_zeroblade said:
I feel like you are getting ton loads of WHEA errors down there, memtest86 doesn't prove anything close to the term "System Stability" it may give you a thumbs up only on the memory department but elsewhere anything can happen..I suggest to use everything as it is (stock) and try to find out where are your issues, from there, once you have been able to find the culprit, then you can gradually start tuning your system..

AMD isn't for the people who don't want to spend time with the platform, its really going to be time consuming taming it, you can enjoy it as it is, but the potential is lost there specially if you bought the high end parts..
Click to expand...
I was gonna check Event Log this morning but totally forgot. The last crash was 9/24: Kernel-Power EventID: 41. However, bug codes are all blank 0s.

But a few hours before that it crashed at 6:48:09 and at 6:49:47 WHEA-Logger reported EventID: 18. and Here's the details. I've gotten this errror at least 8 times in the past two weeks when it crashes. Six of those events I was running all bios settings at stock... except XMP was probably on for most of them, couple of them was running JEDEC.

I don't think AMD is particularly challenging unless you're overclocking RAM. And it's okay, it's a learning process.
o1dschoo1 said:
You are running a almost 10 year old psu... Double check 12v rail etc..
Click to expand...
12v rail measures 12.456
 
#9 · (Edited)
SkillzKillz said:
With my last changes (16. 17. 18.) I've been running for over 14h, which is the longest I've ever run on this setup. Could be dumb luck, who knows. Other things that have changed—because I put desktop on another floor— 19. ethernet instead of wireless 20. different surge protector 21. different wall outlet. So I guess that's 6 things that are different and this PC hasn't crashed... yet. I will leave it on overnight- if it will crash that will surely do it.

I'm really hoping the August Beta BIOS has fixed it. This might be a compatibiilty issue of having Kingston's new RAM that launched on July 29.

Right now I only have a few seconds video showing my cursor locked up and the screen going black for the reboot. It's otherwise impossible to catch any other action because it happens so fast. I don't know how to trigger it either. Oh but sometimes I can go like 8h with no problem, then it will freeze & reboot and within 5 minutes of the startup it will freeze and reboot again. Then it's good for another 6h-12h. No USB issues.

My 5600x was made in 2020, week 50. The previous owner purchased it in February.

In the RMA process, 1) if I can't narrow down what the problem is... which component manufacturer do I go with? 2) If I RMA a component, but they verify that it functions properly, what happens?


I was gonna check Event Log this morning but totally forgot. The last crash was 9/24: Kernel-Power EventID: 41. However, bug codes are all blank 0s.

But a few hours before that it crashed at 6:48:09 and at 6:49:47 WHEA-Logger reported EventID: 18. and Here's the details

I don't think AMD is particularly challenging unless you're overclocking RAM.

12v rail measures 12.456
Click to expand...
Yeah, that's why he sold it, i guess.
It is an early sample, most of these issues are due to weak and badly binned cpu, happened especially when the 5K cpu launched last year.

Once again, you are checking all the boxes, it should not happen, there are already huge threads about, one on this same forum and also the AMD forum.
The reboot at idle, even at stock, was the most prominent issue!

It has been better when users begin to complain, AMD increased a bit the silicon quality of their cpu, chiplets.
Still, even nowadays, some samples are still impacted by these issues.

If the old owner provided the proof of purchase, RMA it, don't spent more time on it.
AMD will not push you back too much if you provide the proofs as advised, AMD has a page dedicated to the troubleshooting, you can find here.

As you can see, not a lot of users will tell you the whole story, keeping telling you to tune your system, it is unfortunate, but it is as it is.
It is a sad situation for AMD, that obviously didn't have been correctly acknowledged.
Don't buy second hands AMD 5K series cpu, that are dated near the 5K launch, late 2020, begin 2021.

Good luck!
 
Discussion starter · #10 ·
1devomer said:
Yeah, that's why he sold it, i guess.
It is an early sample, most of these issues are due to weak and badly binned cpu, happened especially when the 5K cpu launched last year.

Once again, you are checking all the boxes, it should not happen, there are already huge threads about, one on this same forum and also the AMD forum.
The reboot at idle, even at stock, was the most prominent issue!

It has been better when users begin to complain, AMD increased a bit the silicon quality of their cpu, chiplets.
Still, even nowadays, some samples are still impacted by these issues.

If the old owner provided the proof of purchase, RMA it, don't spent more time on it.
AMD will not push you back too much if you provide the proofs as advised, AMD has a page dedicated to the troubleshooting, you can find here.

As you can see, not a lot of users will tell you the whole story, keeping telling you to tune your system, it is unfortunate, but it is as it is.
It is a sad situation for AMD, that obviously didn't have been correctly acknowledged.
Don't buy second hands AMD 5K series cpu, that are dated near the 5K launch, late 2020, begin 2021.

Good luck!
Click to expand...
Yeah I found a thread of people all reporting same problem as me.

I do have PoP. He purchased from BHP US, shipped to Canada.

I asked him why he's selling. Said he switched to Intel because it's better for gaming. Said he didn't overclock. He probably switched to Intel because of the problems with AMD and dumped this lemon on me.

Thank you for your help I will contact AMD.
 
#11 ·
SkillzKillz said:
But a few hours before that it crashed at 6:48:09 and at 6:49:47 WHEA-Logger reported EventID: 18. and Here's the details. I've gotten this errror at least 8 times in the past two weeks when it crashes. Six of those events I was running all bios settings at stock... except XMP was probably on for most of them, couple of them was running JEDEC.
Click to expand...
your chip may be defective, no wonder previous owner sold it..event 18 is something hard to fix (other than raising voltages) unlike event ID 19's..
 
#12 ·
My 5800x out of the box wouldnt get stable until I threw 1.12 vcore soc @ it. At higher ram speeds, It needs up to 1.25V vcore soc to stay stable. I think I am currently around 1.16 or 1.18
 
Save
Reply Quote Rep+ Rep+
Discussion starter · #13 ·
Insidious Supra said:
My 5800x out of the box wouldnt get stable until I threw 1.12 vcore soc @ it. At higher ram speeds, It needs up to 1.25V vcore soc to stay stable. I think I am currently around 1.16 or 1.18
Click to expand...
Were you getting WHEA errors on it?

I just ran 1 stick of ram for 24h at 1200mhz with no crash, setting a new record by 13h. Then I did a 3h MEMTEST on it, which passed. Now I'm back to two sticks @ 3600mhz trying to hone in on the issue.
 
#14 ·
Yep, was getting a lot. vcore is for the memory controller.... so i recon that more sticks and more speed requires more voltage. I run 4 sticks because my board is t-topology, if it was daisychain I would run 2 sticks.
 
Save
Reply Quote Rep+ Rep+
Discussion starter · #15 ·
My PC has been crashing since I built it on September 13 and after 25 changes and tests I've finally found the problem.

Last week, I moved my PC to another floor and it ran stable for 64h from 9/27 to 9/30. The changes I made at this transition were:

20. Using ethernet instead of wireless adapter
21. New surge protector
22. New wall outlet
23. One monitor instead of two
24. No speakers (RealTek Audio Drive may cause crashes)

But moving back to the other floor it crashed again. So, I tested the monitors, audio, and all cables. Finally, I uninstalled the wireless adapter and I've been stable for almost 3 days now with my original setup but a different wireless adapter.

The culprit crashing my PC:
  • ASUS PCE-AC55BT Dual-band Wireless AC1200 PCIe Adapter
  • Intel PROSet/Wireless Driver 21.10.1 for Windows 10 x64
I'm so relieved that all my expensive hardware functions great, I don't have to RMA anything, and my system is rock solid stable at stock settings. It's been three and a half weeks of troubleshooting but despite that, I'm so grateful for the outcome. Thank you everyone for your help!
 
#17 ·
SkillzKillz said:
My PC has been crashing since I built it on September 13 and after 25 changes and tests I've finally found the problem.

Last week, I moved my PC to another floor and it ran stable for 64h from 9/27 to 9/30. The changes I made at this transition were:

20. Using ethernet instead of wireless adapter
21. New surge protector
22. New wall outlet
23. One monitor instead of two
24. No speakers (RealTek Audio Drive may cause crashes)

But moving back to the other floor it crashed again. So, I tested the monitors, audio, and all cables. Finally, I uninstalled the wireless adapter and I've been stable for almost 3 days now with my original setup but a different wireless adapter.

The culprit crashing my PC:
  • ASUS PCE-AC55BT Dual-band Wireless AC1200 PCIe Adapter
  • Intel PROSet/Wireless Driver 21.10.1 for Windows 10 x64
I'm so relieved that all my expensive hardware functions great, I don't have to RMA anything, and my system is rock solid stable at stock settings. It's been three and a half weeks of troubleshooting but despite that, I'm so grateful for the outcome. Thank you everyone for your help!
Click to expand...
Glad you sorted out.
I admit, i also wouldn't have bet on the wireless card adapter.

Just out of curiosity, what replaced the Asus AC1200, as wireless adapter, if you are still using any wireless at all?
 
Discussion starter · #18 ·
1devomer said:
Glad you sorted out.
I admit, i also wouldn't have bet on the wireless card adapter.

Just out of curiosity, what replaced the Asus AC1200, as wireless adapter, if you are still using any wireless at all?
Click to expand...
There was a liquidation sale for a local Tiger Direct years back. They had a bunch of USB wireless AC adapters in a box for $20 so I grabbed one just for ****s, not even needing it. Adapters were a lot more expensive back then, at least $80 so I figured why not. Well this little black USB, made by Engenius , has saved my ass so many times now. It's plug-and-play on any system- immediately you can search for wireless networks and it gives you full speeds just as any other wireless does.

Next step... I'm drilling holes in the floor to wire ethernet in throughout my place.
 
1 - 18 of 18 Posts
About this Discussion
17 Replies
6 Participants
Last post:
SkillzKillz
Overclock.net
posts
28M
members
562K
Since
2004
The premier forum for overclocking experts and enthusiasts. Discuss hardware optimization, custom builds, benchmarking, cooling solutions, and pushing the boundaries of computing performance. From beginner guides to extreme overclocking, join our technical community to master system tuning.
Full Forum Listing
Explore Our Forums
Intel CPUs Hardware News NVIDIA PC Gaming Water Cooling