PC Processors

CrispyCrunch · ‎11-19-2020

Mainboard: MSI x570 Unify
Mainboard-BIOS: 7C35vA82 (Beta version)
CPU: Ryzen 5900x
RAM: Crucial Ballistix BL2K32G36C16U4B 3600 MHz, 64GB (32GB x2)
Drive: M.2 Samsung 970 Evo+ 1TB SSD
Graphics: SAPPHIRE Nitro+ Radeon RX 5700 XT
PSU: be quiet straight power 11 750w Platinum
OS: Win 10 Pro (64bit) - all updates installed
Chipset driver: 2.9.28.509 (released 2020-11-09)

I first assembled the PC with a Ryzen 3800x a week ago because it was unclear if and when I would get the Ryzen 5900x I ordered. Worked with the included AMD Prism Wrath CPU cooler for one week without any problems.

- Today I installed a Ryzen 5900x and a Scythe Fuma 2 CPU cooler.
- After 20 min the first crash/restart with the following entries in the Event Viewer: WHEA-Logger ID 18 and critical error Kernel-Power ID 41.
- Happens irregularly again and again, sometimes after minutes, sometimes longer: Windows freezes for a few seconds and then the PC reboots. Doesn't matter if load or not.
- CPU temperature between 30 and 40 °C
- Updated to BIOS and chipset driver mentioned above: Problem still exists
- XMP Profile disabled (RAM on 2600 MHz): problem still exists
- CMOS Reset: Problem still exists

Either there is a compatibility problem of something with the CPU, or the CPU is defective?
What to do? Really frustrating.

Jazzinplayer · ‎11-23-2020

Im having a similar issue, x570 aorus and 5600x. Have same errors on windows.

Disable CBP and PBO and run it at default settings (3.7 ghz and xmp on). That works for me.

View solution in original post

CrispyCrunch · ‎12-03-2020

I got a new angle on this. So deactivating PBO and CBS definetely works, PC was running stable for a week now. But you'll loose performance.

So I wrote to the MSI support and the AMD support.

MSI suggested to try increasing the DRAM Voltage by 0.05 V, which I did. System seems to be stable, no crashes so far - neither in idle or while gaming.

View solution in original post

thunk_stuff · ‎12-11-2020

@j96j wrote:
My only temporary fix is to set my CPU core voltage to 1.3v. Tried 1.25v and I can finish RDR2 without any crashes. But when I play cyberpunk, it crashes. So I use the 1.3v.
However with this setting of CPU core voltage (offset mode 1.3v), my CPU clock speed is running at base clock (3800 MHz).
Is there any way to set CPU core voltage to 1.3v and run CPU clock speed to auto?

Do you have curve optimizer in your BIOS? Have you tried setting it to a positive voltage setting for all core, somewhere bewteen +2 and +10? This would be with all other voltages/overclock at default.

This is what has made my system stable. I know it has worked for one other person. I am still waiting to hear from someone in this thread that it doesn't work for, to know how similar the problem is between me and everything else with these WHEA errors that seem affected by voltage.

tim716 · ‎12-11-2020

Do you have curve optimizer in your BIOS? Have you tried setting it to a positive voltage setting for all core, somewhere bewteen +2 and +10? This would be with all other voltages/overclock at default.
This is what has made my system stable.

Actually, this. Thank you for this info.

EDC limit = 200A.

Curve optimizer = +8.

Looks like it works for me too. Of course your CPU might need more or less curve.

If this works for many people, I can even give a ~~conspiracy~~ theory, explaining this.

Looks like the AMD casino took the silicon lottery to a new level.

The usual gambling used to be - how well you can overclock your CPU, but the base specified peroformance was guaranteed to you. Not anymore. Now, to make the Ryzen great again, the performance AMD specifies is the performance of an AVERAGE CPU. But of course that doesn't mean AMD is going to put a half of the CPU yield which is below that average down the trash and lose profits. That means a half of the buyers downvolts their CPUs to overclock them (the "awesome" new feature much advertised by AMD), and another half OVERvolts their CPUs to UNDERclock them to make them work somehow. This thread is the home of this second half losers. And, miraculously, these attempts to make this crap work voids the warranty, so AMD doesn't even have to take their crap back. Casinos never lose!

Of course this can be corrected by BIOSes (and will be, when AMD is tired of RMAs) by just raising the default voltages and/or cutting the turboboost (together with the performance).

Also it can be easily explained why the systems mostly BSOD or reboot at idle or some plain low load tasks, and remain stable under burn-in. The problem is not overheating, the problem is inability of a given ~~crappy~~ CPU to work stable at a given frequency with a given voltage. (just the same as if you undervolt it too much). The larger the frequency, the more chance of a BSOD to occur. The fully all-core loaded CPU works at LESSER frequences to stay within the TDP. But when you stop your burn-in and start to watch a video, just one or two cores (pre-heated by the previous burn-in) work, but they work at the MAXIMUM frequences. And - say Hi to a BSOD or reboot.

If the above turns out true, I'd advise everyone having a similar problem th RMA their CPUs ASAP.

thunk_stuff · ‎12-11-2020

@tim716 wrote:

Actually, this. Thank you for this info.
EDC limit = 200A.
Curve optimizer = +8.
Looks like it works for me too. Of course your CPU might need more or less curve.
If this works for many people, I can even give a ~~conspiracy~~ theory, explaining this.
Looks like the AMD casino took the silicon lottery to a new level.

Glad it helped. One thing I am curious as hopefully more people test it out, is how many can get it stable at +2, +4, or do people need to raise it higher. +6 almost worked for me, but crashed after a day. +8 is the lowest I can get and be stable. Maybe +7 would work, but I'm just not in the mood to deal with bluescreens right now.

I was kind of thinking along the same lines that we are on the losing end of silicon lottery. It would make sense given how a simple voltage tweak fixes the problem and makes it 100% stable. I am on day 11 of this fix working for me using my computer daily at all hours. This fix also manifest as me having a low quality binned chip running at lower frequency: I can run stable but my clocks are 100-200Mhz less than at default, and my chip runs hotter.

If this is the case, AMD should have discovered this in the testing phase and put them aside for a lower end chip.

But we may have a clearer idea when AGESA 1.1.8.0 gets released this month and PBO2 is official. The 5000 series was designed to run best with PBO2. Maybe AMD had our chips in mind but couldn't get a fix out in time. Crossing finger that this is the case.

If not, we have full reason to RMA this chip as we are not able to run stable at default or get expected clocks.

mackbolan777 · ‎12-11-2020

I found AMD updated the chipset driver on 10/20, it specifically addresses fixes for the 5000 series CPU's, support for the 6000 GPU's and some other things. Might be worth a try.

"It worked before you broke it!"

Deocrone · ‎05-18-2021

Wow, having the BSOD for a while after building a new system. Haven't had the time to investigate some more.. Was doubting myself.. after having checked all things.. ripping apart the pc.. almost reinstalling windows..

Came to this thread finally by searching for BSOD in combination with 5900x.. (what could possibly be wrong with this new chip huh?).. x570 Aorus elite Bios version F32 has a Vcore Loadline calibration.. I've set this from Auto to medium (High the first time) and disabled the cool and quite function (could result in low voltage on the chip, was my theory). Haven't had any BSOD for a few hours on Idle.. which crashed every 15 minutes or so yesterday.. Also had CPB disabled.. but turned it back on to auto... Hope this helps others confirming these many crashes are result of a CPU voltage combination..

Greetings.. 1st time self builder.

Cmdr-ZiN · ‎05-18-2021

@Deocrone are you actually getting BSODs if so what is the error code? Most of us aren't getting blue screens and the system is power cutting and restarting while on idle.

If you're having the same issue check page 68, it's the full troubleshooting from when I contacted AMD tech support. I fixed the issue by setting to Typical Idle Current, I don't know if if my PSU should handle power draw low enough for the CPU or not. Your solution might be a different one on that list.

If that doesn't solve your problem contact AMD tech support and see what they say.

Deocrone · ‎05-26-2021

I'm actually getting a blue screen.. telling me there is a WHEA external error blabla error... Always in Idle.
Never whenever a game or something is running. It can go for hours when in game (even when not playing) without crashing.

But as soon as i leave it on just windows or browser for 15-30 minutes.. it has restarted itsself.
I've cut my way through all the settings and options on page 68. Still crashing (well it does crash less fast).

Setting my entire Bios to default and just settting the Vcore line up a notch from auto to high or medium.. No issues anymore. Feels a bit weird to RMA this with applying this (small) fix.. or am I missing a potential risk on this V-core setting ?

Do you still advice contacting AMD tech support ?

Thanks, Deocrone

NewAMDGuy · ‎05-26-2021

So a lil update from me, ive disabled PBO and the game is running stable till now

Cmdr-ZiN · ‎05-26-2021

@Deocrone you shouldn't have to go to such lengths it should run stable at stock settings.

It might not be the CPU it could be so many things that a BIOS update might fix. However it's presenting as a hardware failure so should be fixed.

Mine was stable for months, granted I haven't had a lot of time to use it over that time, but it died while updating Star Citizen corrupting the install completely, I needed to fully remove it to fix. This was annoying and a waste of time, but what if you lost something important, I'd want a system I could trust.

Deocrone · ‎05-26-2021

True.. I shouldn't have gone through all this trouble.

It should have been plug and play. (well,.. installing all drivers and updating the Bios is logic).
Currently going on the X570 Aorus Elite F32 Bios update (Latest update without a Letter behind it).

But going as far as a V-core loadline adjustment.. shouldn't have been necessary. So i'll contact someone at AMD..
But nontheless.. Setting this Loadline, fixed the idle crashing 100% for me.

Does anybody have an idea if this is a risk to the durability of the system or anything ?

j96j · ‎12-17-2020

@thunk_stuff wrote:
Do you have curve optimizer in your BIOS? Have you tried setting it to a positive voltage setting for all core, somewhere bewteen +2 and +10? This would be with all other voltages/overclock at default.
This is what has made my system stable. I know it has worked for one other person. I am still waiting to hear from someone in this thread that it doesn't work for, to know how similar the problem is between me and everything else with these WHEA errors that seem affected by voltage.

So I tried this method of using curve optimizer, setting positive voltage all core to +2. Other settings at default, except for this curve optimizer. My PC haven't restarted at idle or at light gaming. Haven't tried stress testing, since I'm fairly new to this stuff. What is the recommended application for stress testing CPU?

Usually at stock settings with no curve optimizer, my pc will go into constant reboots randomly. Even at idle, or disabling CPB PBO etc. I'm still not sure if the above method is the temporary solution though. I've been troubleshooting this PC for 3 weeks.

I'm using 5800x, x570 tomahawk, RAM 3600c18.

If anyone is wondering, the settings of curve optimizer in x570 tomahawk is:
BIOS advanced mode > Settings > Advanced > AMD Overclocking > Accept > Set PBO to Manual > Curve Optimizer should show up

thunk_stuff · ‎12-17-2020

@j96j wrote:
So I tried this method of using curve optimizer, setting positive voltage all core to +2. Other settings at default, except for this curve optimizer. My PC haven't restarted at idle or at light gaming. Haven't tried stress testing, since I'm fairly new to this stuff. What is the recommended application for stress testing CPU?

Best application is using it in a variety of ways and observing if it crashes at all over the next couple of days.

You're luckier than me that it's so far stable at +2. When I set it to +2, it was still crashing within a couple of minutes.

When I set it to +6, it was much more stable. Passed every benchmark and stress test I threw at it. Eventually it did crash the next day (after 18 hours or so turned on), when doing a 3D benchmark (spikes in CPU usage).

That's when I set it to +8, and it was stable for nearly two weeks, no crashes or any warnings in the event logger. I've now RMA'd it to AMD and will see what happens. So my test didn't go beyond 2 weeks.

How to test in a a variety of ways:

Continuously low usage (streaming video, music)
Low usage with spikes (Web browsing)
Watching videos
Playing variety of video games
Content creation
High CPU usage tests (Cinebench, Blender) that stress the cores, single and multi
Gaming benchmarks (Heaven 4.0 Benchmark, Superposition). Heaven can run continuously.
Passmark Performance Test (tests all components of your system)
prime95 (stresses CPU over long period of time)

nxvkumar · ‎12-17-2020

I'm getting Kernal Power event 41 as well with my 5600x everytime i try to to load into a game or play a game for about 5 mins or so.

I'm running an ASrock b450m hdv r4.0, does anyone know the BIOS settings for i need to change for an ASrock motherboard in order to have a stable pc while gaming.

Thanks

BraveKiwi · ‎12-17-2020

An update for my particular case.

I followed MSI's advice, and today i've been running the system with CPB and PBO on, but adjusting NB/SoC voltage with a +0,05V offset. The system has been stable all day, i'll leave it on idle through the night to check if i keep crashing.

MSI basically instructed me to either keep CPB off, or to try a slight voltage offset on NB/SoC, while we wait for a BIOS update:"If the pc can work stable, you do not need to adjust the voltage, if the issue still appears, you can adjust a little of the voltages to check. And we will continue to update the bios, please update the new bios in the future."

Guess we're stuck until we get a new BIOS update, i'll report on how effective the voltage offset is, it might be a temporary solution that doesn't hinder performance while we wait for said update.

BraveKiwi · ‎12-18-2020

@BraveKiwi wrote:
An update for my particular case.
I followed MSI's advice, and today i've been running the system with CPB and PBO on, but adjusting NB/SoC voltage with a +0,05V offset. The system has been stable all day, i'll leave it on idle through the night to check if i keep crashing.
MSI basically instructed me to either keep CPB off, or to try a slight voltage offset on NB/SoC, while we wait for a BIOS update:"If the pc can work stable, you do not need to adjust the voltage, if the issue still appears, you can adjust a little of the voltages to check. And we will continue to update the bios, please update the new bios in the future."
Guess we're stuck until we get a new BIOS update, i'll report on how effective the voltage offset is, it might be a temporary solution that doesn't hinder performance while we wait for said update.

An update on this. I've been running for almost 24 hours now with NB/SoC voltage offset +0,05V, completely stable (CPB on), in case this is useful to any of you.

860lacov · ‎12-18-2020

MSI Advice - turn pbo on

And bios says - turning pbo on can validate warranty :D

Ok so let's sum up

I don't need pbo for now but prefer to have CPB

What motherboard do you have? Can't remember if my tomahawk has got offset for NB/SoC

What is safe value of soc if I I will have to adjust it manually?

Instead of adding offset can I set constant value ?

A would like to have stock value increased by 0.05 but don't know what is default

Am I correct?

You have

CPB on

Xmp on

Soc + 0.05V

And everything is working fine?

BraveKiwi · ‎12-18-2020

@860lacov wrote:
MSI Advice - turn pbo on
And bios says - turning pbo on can validate warranty :D

Ok so let's sum up
I don't need pbo for now but prefer to have CPB

What motherboard do you have? Can't remember if my tomahawk has got offset for NB/SoC
What is safe value of soc if I I will have to adjust it manually?
Instead of adding offset can I set constant value ?
A would like to have stock value increased by 0.05 but don't know what is default

Am I correct?
You have
CPB on
Xmp on
Soc + 0.05V
And everything is working fine?

MSI didn't instruct me to turn PBO back on, they nudged me to try the voltage offset.

I've got an MSI MPG X570 Gaming Plus. I don't know the safe values for X570 and Ryzen 5000, i (a risky move, i admit) just tried adjusting NB/SoC voltage offset to +0,05V and kept an eye on my voltages. CPU doesn't seem to peak any higher than before, so no problems there.

And yes, you are correct: CPB on, XMP on, NB/SoC +0.05V offset, and everything is working fine... for now.

860lacov · ‎12-18-2020

I'm not sure but NB SoC is connected with memory controller so maybe it has impact on ram and not cpu?

Could you tell me what settings do you have for vcore and if it's auto what is bios showing?

In my case vcora auto and:

With CPB on bios shows 1.4V vcore

System shows 1.0 idle and 1.1load

With CPB off

Bios shows 1.0V

System shows under idle 0.9V

Under load 1.0V

BraveKiwi · ‎12-18-2020

@860lacov wrote:
I'm not sure but NB SoC is connected with memory controller so maybe it has impact on ram and not cpu?

Could you tell me what settings do you have for vcore and if it's auto what is bios showing?

In my case vcora auto and:
With CPB on bios shows 1.4V vcore
System shows 1.0 idle and 1.1load

With CPB off
Bios shows 1.0V
System shows under idle 0.9V
Under load 1.0V

Hmm, your voltages are different than what i'm seeing on my system. I have not touched vcore, it's set at auto.

With CPB on and my offset applied, i'm looking at 0.9V idle and 1.3 - 1.35V sustained under load (peaking at 1.44V very shortly, at least in HW monitor - Ryzen Master shows around 1.35V under load).

860lacov · ‎12-18-2020

Thats why I mentioned bios and system

If you can please check what bios says.

I use linux so it may have readings that are not very precise I think.

jpee80 · ‎12-18-2020

I left a post in Reddit about this issue in r/AMDHelp.

https://www.reddit.com/r/AMDHelp/comments/kfyst7/random_bsods_with_amd_5000_series_processor/

A google search about BSODs with AMD 5000 Series CPU return with many results

j96j · ‎12-18-2020

@BraveKiwi wrote:
MSI didn't instruct me to turn PBO back on, they nudged me to try the voltage offset.
I've got an MSI MPG X570 Gaming Plus. I don't know the safe values for X570 and Ryzen 5000, i (a risky move, i admit) just tried adjusting NB/SoC voltage offset to +0,05V and kept an eye on my voltages. CPU doesn't seem to peak any higher than before, so no problems there.
And yes, you are correct: CPB on, XMP on, NB/SoC +0.05V offset, and everything is working fine... for now.

Tried your method, setting NB/SoC offset mode +0.0500V, XMP on, other settings default. PC still reboots while gaming. No WHEA, just Kernel-P and audits.

Even disabling PBO, CPB, cstate, reboots my pc. Don't know any other solutions.

BraveKiwi · ‎12-18-2020

@j96j wrote:
@BraveKiwi wrote:
MSI didn't instruct me to turn PBO back on, they nudged me to try the voltage offset.
I've got an MSI MPG X570 Gaming Plus. I don't know the safe values for X570 and Ryzen 5000, i (a risky move, i admit) just tried adjusting NB/SoC voltage offset to +0,05V and kept an eye on my voltages. CPU doesn't seem to peak any higher than before, so no problems there.
And yes, you are correct: CPB on, XMP on, NB/SoC +0.05V offset, and everything is working fine... for now.
Tried your method, setting NB/SoC offset mode +0.0500V, XMP on, other settings default. PC still reboots while gaming. No WHEA, just Kernel-P and audits.
Even disabling PBO, CPB, cstate, reboots my pc. Don't know any other solutions.

Then you might be having a different problem than the ones described in this thread. If you're not getting a WHEA-logger ID 18, and your system is rebooting under load, i might look at the PSU.

My system reboots with the WHEA-Logger ID 18 event, and it only happens when under light load/idle.

Flaunt · ‎12-17-2020

@thunk_stuff wrote:
@j96j wrote:
So I tried this method of using curve optimizer, setting positive voltage all core to +2. Other settings at default, except for this curve optimizer. My PC haven't restarted at idle or at light gaming. Haven't tried stress testing, since I'm fairly new to this stuff. What is the recommended application for stress testing CPU?
Best application is using it in a variety of ways and observing if it crashes at all over the next couple of days.
You're luckier than me that it's so far stable at +2. When I set it to +2, it was still crashing within a couple of minutes.
When I set it to +6, it was much more stable. Passed every benchmark and stress test I threw at it. Eventually it did crash the next day (after 18 hours or so turned on), when doing a 3D benchmark (spikes in CPU usage).
That's when I set it to +8, and it was stable for nearly two weeks, no crashes or any warnings in the event logger. I've now RMA'd it to AMD and will see what happens. So my test didn't go beyond 2 weeks.
How to test in a a variety of ways:
Continuously low usage (streaming video, music)
Low usage with spikes (Web browsing)
Watching videos
Playing variety of video games
Content creation
High CPU usage tests (Cinebench, Blender) that stress the cores, single and multi
Gaming benchmarks (Heaven 4.0 Benchmark, Superposition). Heaven can run continuously.
Passmark Performance Test (tests all components of your system)
prime95 (stresses CPU over long period of time)

How long from when you first initiated contact to when they sent you a shipping label? I contacted them and then 10 days later they asked for more info. Now nothing for 4 days.

thunk_stuff · ‎12-18-2020

@Flaunt wrote:
How long from when you first initiated contact to when they sent you a shipping label? I contacted them and then 10 days later they asked for more info. Now nothing for 4 days.

13 days from initiating RMA to getting a response. Submitted details and two days later it was approved.

Flaunt · ‎12-18-2020

@thunk_stuff wrote:
@Flaunt wrote:
How long from when you first initiated contact to when they sent you a shipping label? I contacted them and then 10 days later they asked for more info. Now nothing for 4 days.
13 days from initiating RMA to getting a response. Submitted details and two days later it was approved.

Any ETA on getting the chip back?

thunk_stuff · ‎12-18-2020

No idea. I'll post back when they get back to me and if I get a new chip.

j96j · ‎12-18-2020

Best application is using it in a variety of ways and observing if it crashes at all over the next couple of days.
You're luckier than me that it's so far stable at +2. When I set it to +2, it was still crashing within a couple of minutes.
When I set it to +6, it was much more stable. Passed every benchmark and stress test I threw at it. Eventually it did crash the next day (after 18 hours or so turned on), when doing a 3D benchmark (spikes in CPU usage).
That's when I set it to +8, and it was stable for nearly two weeks, no crashes or any warnings in the event logger. I've now RMA'd it to AMD and will see what happens. So my test didn't go beyond 2 weeks.
How to test in a a variety of ways:
Continuously low usage (streaming video, music)
Low usage with spikes (Web browsing)
Watching videos
Playing variety of video games
Content creation
High CPU usage tests (Cinebench, Blender) that stress the cores, single and multi
Gaming benchmarks (Heaven 4.0 Benchmark, Superposition). Heaven can run continuously.
Passmark Performance Test (tests all components of your system)
prime95 (stresses CPU over long period of time)

Update, pc reboots after 4-hours, light browsing and idle. Now trying a +4 curve optimizer.

Is RMA the same as warranty claim? Because I tried checking this CPU to my local shop that I bought it from. After 2-days of their testing, it never reboots on their system. They're using ASUS B550 motherboard though. So now I'm stuck with this CPU.

thunk_stuff · ‎12-18-2020

@j96j wrote:
Is RMA the same as warranty claim? Because I tried checking this CPU to my local shop that I bought it from. After 2-days of their testing, it never reboots on their system. They're using ASUS B550 motherboard though. So now I'm stuck with this CPU.

RMA falls under the warranty claim. If the RMA shop tests it and can't replicate the issue, I'm not sure what happens at that point. SOL?

I wonder if the board they tested it at your shop was running a different AGESA version.

860lacov · ‎12-18-2020

I don't know if this ia cpu or not

I left CPB enabled but set core multiply manually to 42

Auto in my case is at 4080MH

I set vcore from auto (in auto with CPB bios shows 1.4 and system shows 1 -1.1)

to 1V + 0.125 offset

Bios constantly shows 1.15V but system shows 0.9 idle and 1.1 load

With those settings I coukd pass blender benchmark

It's not the solution. Its another way around to get performance closer where it should be.

It seems that auto core boost + auto vcore instead of best stability gives us f...d up system.

Still don't know if its cpu or just bios

j96j · ‎12-18-2020

@thunk_stuff wrote:
RMA falls under the warranty claim. If the RMA shop tests it and can't replicate the issue, I'm not sure what happens at that point. SOL?
I wonder if the board they tested it at your shop was running a different AGESA version.

Another update, PC still reboots using curve optimizer +4 +6 when playing games. What's weird is it runs fine when stress testing in cinebench and AIDA64. What's different is: In event viewer, usually there is a WHEA error. Now there's no WHEA, only Kernel-Power and audit etc.

When I tried +8 curve, it has gone into a reboot loop in BIOS.

Now I've revert back to turning off PBO, CPB and global csate.

For your settings do you only have +8 curve optimizer and have the other settings at default?

thunk_stuff · ‎12-18-2020

@j96j wrote:
@thunk_stuff wrote:
RMA falls under the warranty claim. If the RMA shop tests it and can't replicate the issue, I'm not sure what happens at that point. SOL?
I wonder if the board they tested it at your shop was running a different AGESA version.
Another update, PC still reboots using curve optimizer +4 +6 when playing games. What's weird is it runs fine when stress testing in cinebench and AIDA64. What's different is: In event viewer, usually there is a WHEA error. Now there's no WHEA, only Kernel-Power and audit etc.
When I tried +8 curve, it has gone into a reboot loop in BIOS.
Now I've revert back to turning off PBO, CPB and global csate.
For your settings do you only have +8 curve optimizer and have the other settings at default?

Yes, only +8 in optimizer and no other BIOS tweaks besides enabling RAM XMP.

So... +2 to +6 are all better than default, but still crash in games randomly. You have a kick ass motherboard so that shouldn't be it.

I read that some people had instability with BIOS 3001 that they didn't have before. You could try an older AGESA version to see if it's more stable (if this is even possible):

https://www.reddit.com/r/ASUS/comments/k97mm6/bios_version_3001_vs_version_2816_rog_strix_x570e/

860lacov · ‎12-18-2020

ok

Checked the solution with changing soc from auto to offset

at offset +0.05 blender benchmark passed 2 times in a row. First time so good.

During regular stuff (image conversion. Not very cou demanding ) pc rebooted.

I changed soc to + 0.075

Don't know if this is safe but I was able to finish image conversion.

Still have no idea if this is cou, mobo or bios problem.

Wanted to change mobo from msi to strux x570-e (I wanted it from beginning but it seems that it's not connected to manufacturer so not sure if its a good idea)

paez17 · ‎12-19-2020

+0.05 offset method seems worked with my 5900x

It passed blender 6-7 times without an issue. If cpu isn't stable it was crashing even in loading screen of blender.

Beside offset method only solution worked for me was disabling pbo cbp and using at constant 3700mhz.

Especially vram temps got high with offset +0.05. 15 min torture to full system made cpu temp 76 degree with water cooling. Before + offset i never seen in above 70.

If it crashes again i'll send it back to the rma on monday.

860lacov · ‎12-19-2020

maybe it's worth to set constsnt value?

+0.05 helped in for example blender benchmark but pc still crashed on idle.

I changed to 0.075 and then idle started to work. But still I'm not sure if bios update will help . If not then I think 5900x is not worth keeping.

For me this is an insult. It wasn't small price and AMD without any help showed what do they think about customers

colesdav · ‎12-19-2020

See: https://community.amd.com/t5/processors/best-motherboards-for-amd-ryzen-5000-cpus-x570-b550-for-gami...

860lacov · ‎12-19-2020

I wonder...

If this is a bios problem then why only some people have this problem?

paez17 · ‎12-19-2020

If we look at 100 pc which have same hardware and software and compare temps, voltages, performance etc You'll see 100 different results. And here We talk about 0.05 voltage ups or downs for stability surely it depends on the system we don't have same nb soc voltages, vcores, dimm voltages.

If a bios isn't stable we can't find a common solution for a common problem it may change setup to setup.

deepcuts · ‎12-20-2020

@860lacov wrote:
I wonder...
If this is a bios problem then why only some people have this problem?

Because they haven't figured out how to stabilize crappy silicon yet.

Decent to excellent silicon = no problems even with crappy AGESA and crappy BIOS
Bottom of the barrel silicon = whea and reboots.

Podersen · ‎12-20-2020

Returned my 5900x because of me having all the same issues as aforementioned no whea error 18. Just kernel 41.

Bought a 3700x and I don't know how but that chip is displaying the same behaviour. This is on a new X570 unify and gskill 3600 cl16. I don't think there is anything wrong with power. So am I just so incredibly unlucky or is every amd chip like this? I just came over from Intel.

PC Processors

Ryzen 5900x: System constantly crashing/restarting WHEA-Logger ID 18 and critical error Kernel-Power