X570 Tomahawk wifi + Ryzen 7 5800x workaround

[Griffo77]

Getting WHEA errors, random black screens and reboots after installing my new Ryzen 7 5800x in a X570 Tomahawk wifi. I was so desperate that I almost RMA the CPU. After some digging installed Bios 7C84v151(Beta version), cleared cmos, updated chipset drivers (from AMD), disabled PBO and CPB. Enabled XMP P2 (3600). No more reboots or errors but the cpu was stuck at 3800MHz. Got to the Bios and set Vcore to 1.35V, Vsoc to 1.075 and CPU multiplier to 44.

It is stable, including games, and running for 24 hours now at 4400 Mhz. Max temps in Cinebench r20 near 78ºC in multicore. Reducing multiplier to 42.5 tomorrow, ... just in case ;-)

Waiting for a fix.

I hope it helps!

 

[dummykc]

yes, PBO is not working correctly on my set up as well, here are the scenarios:
1. Under setting, set PBO to advance, set scalar to auto, and PBO with 200 MHZ, I can login to windows, and pass CB20 with multi-core test. However, I will get WHEA BOSD with single core test.
2. Under Setting, set PBO to Advance, set scalar between 1x to 10x, and PBO with 200 MHZ, I will get BOSD once the system enter the windows desktop.
3. Under setting, set PBO to advance, set scalar between 1x to 10x, tried curve from -1 to -20, system will have black screen and reboots 2 or 3 times, then I have to turn off power manually, then reset my BISO.
4. Under setting, set PBO to either auto/enable/disable, they system will operate without issue.

 

[Griffo77]

Update 1
Put back auto voltages, multiplier and enabled CPB (PBO disable).
System boot into Windows and pass 3 Cinebench R 20 with no errors. HWINFO reported a red 80% value in "Power Reporting Deviation".
Crashed (black screen and reboot) loading Star Wars: Jedi Fallen Order.
Going back to manual voltings. Something very wrong with boost (not only PBO).

 

[Griffo77]

Update 2
Set Vcore to 1.325 Volts, Vsoc to 1.1 volts and multiplier to 46 (PBO and CPB off).
Stable for 24 hours without errors at 4600 Mhz. Max temps in multi-core around 78ºC in Cinebench R20 and 49ºC in games.
So i gess issues are not related to a faulty CPU.
(Maybe it placebo effect but i think MOS temps are quite lower relative to auto settings...)

 

[Griffo77]

Update 3
Disabled Global C States and enabled PBO and CPB (multiplier on auto) and manualy set vcore to 1.375 volts and vsoc to 1.125 volts. It WORKS!!!! Stable and no errors and boosting to 4.850MHz. Something with auto voltage is wrong!!

 

[CaptFwiffo]

Update 3
Disabled Global C States and enabled PBO and CPB (multiplier on auto) and manualy set vcore to 1.375 volts and vsoc to 1.125 volts. It WORKS!!!! Stable and no errors and boosting to 4.850MHz. Somenthing with auto voltage is wrong!!

YES!! Did the same in my bios ver 7C35v1D2 (MEG x570 Ace) with disabling C States, etc, I have auto voltage and set FCLK manually to 1800Mhz with XMP Profile 1 active. No reboots so far.

 

[CaptFwiffo]

YES!! Did the same in my bios ver 7C35v1D2 (MEG x570 Ace) with disabling C States, etc, I have auto voltage and set FCLK manually to 1800Mhz with XMP Profile 1 active. No reboots so far.

Damn, still getting event 41 ntfs.sys reboot errors. Dammit! Anyone else know what else I can do?????

 

[Griffo77]

My problem is autovoltage. I think it is to low. Setting to manual solved all problems. I AM running now stable at 4575Mhz all Core at 1.25v and LLC m6 with max temps about 60 degrees C. I am happy with these settings for now.

 

[Griffo77]

Damn, still getting event 41 ntfs.sys reboot errors. Dammit! Anyone else know what else I can do?????

I have issues with VRAM on auto and xmp enabled. Did you setting a manual Vram?

 

[joesquin159102e2]

I swapped out a 5600x to a 5900x. Immediately starting getting these pc reboots on a b550 gaming plus with a whea 18 error in event viewer. I didn’t change vcore, but I manually set vsoc, vddp, vddg, and dram voltages and also disabled global c state control. If I turn c states on I immediately get reboots, so a combination of all of these were needed. Knocking on wood, but pc has been on for 12 hours with no reboots yet, compared to the 6 I had yesterday.

 

[CaptFwiffo]

I have issues with VRAM on auto and xmp enabled. Did you setting a manual Vram?

I manually set my RAM to DDR4-3200 (using 3600 RAM). and turned off XMP. Seems to be a bit more stable at the moment.

 

[navyada154d02de]

hey -
i have a 5950x and MSI unify X570.

my system is completely unstable at stock. the only way I can boot windows and remain stable is if I disable CPB and this will lock my CPU to base speed of 3.4.

i can all core OC just fine. i am stable 4.4@ 1.225 v.

maybe i should try a manual voltage for my 5950x? anyone know what i should try?

also - what is the point of Vsoc voltage, vddp and vddg?

 

[Griffo77]

Vsoc can help with RAM stability.

 

[Griffo77]

Update 4
Set Vcore to 1.25 Volts, Vsoc to 1.1 volts and multiplier to 45,50 (PBO and CPB off).
Stable without errors at 45.50 Ghz. Max temps in multi-core around 62ºC in Cinebench R20 and 49ºC in games even with Cyberpunk.
I am going to stick with this for now!

 

[Griffo77]

Update 5
Digging AMD forum, i found several people saying that it is a bios problem (something related with digital fuse) and there is a guy who posted this "solution":

Go into BIOS. Disable CBP, save, reboot. Go back into BIOS.
Set the performance enhancer to level 3 (i dont know this feature in MSI mobo)
Then go to the PBO settings and
set PBO to enabled
set Fmax enhancer to enabled (i dont know this one to)
Open the advanced tab and open AMD overclocking.
Select PBO here
set PBO to advanced and the limit to motherboard.
(The main thing) EDC - Set Spike VRM-out current limit to 200A. (later we can back this down, probably AGESA 1.2)
(Just in case) PPT - Set the socket power limit to 130W. (depends on mobo)
(Just in case) TDC - Set the vrm thermal limit to 85. (depends on mobo and proc)
EDC is a temporary massive increase, PPT here a decrease. TDC depends on your cooling of VRMS and such.

Leave at zeros all the rest in that menu
Set Idle Voltage to Typical
Set Global C-states control to Disable
MAKE SURE THAT ECO MODE IS OFF.
NOW you can reboot, go into BIOS and set Core Precision Boost back to On, everything should work.

Does anyone know what is the msi equivalent to Fmax enhancer and performance enhancer level are?

 

[pnjoram158802dc]

First of all, let me thank you, for the first time since i bought my new computer, im able to play games on it. I was just wondering what kind of cooler do you use for your cpu 77griff159e02ea? With these settings im in the mid 90ºC. Im using Kraken X63.

 

[Griffo77]

You are welcome mate! Glad you get It running.
What settings did you go? I am on update 4 with 360 cooler master aio with custom fans.
90 degrees is to hot...

 

[pnjoram158802dc]

Im using your update 4 setting aswell, although i cant set the soc to 1.1 When i do it jumps to a nearby fixed setting, dont remember what it was but 1.0xx i think. Before i starte tweaking anything in bios my cpu was around 55C idle in Windows, but When i Google it i found out that was what it was, so i didnt think more of it.

Im not an experienced oc'er or bios adventurer xD but i change only those setting that you provided. Arr you able to take a few pictures of your bios setting so i can double check with mine? Last time i bought an aio water cooler, it was broken and i got a new one, its Just too good to be true that it has happens a second time in a row.

EDIT: When i got home from work, the computer would start for a second, then shut off straight away. I tried a few things, but the only thing that fixed it was to clear CMOS. Could it be that tge computer did not get enough power at the start cyclus?
Trying to apply same setting again.

 

[Griffo77]

I am not at home thus o cant take pictures of the BIOS. But settings are Simple:
-Global c states DISABLE
-CPU multiplier 45.55
-Vcore OVERRIDE 1.25 volts
-Vsoc OVERRIDE 1.1 volts
-Load like calibration MODE 6
-Cpb and pbo DISABLE
-XMP ON profile 2 (3600 mhz/ IF 1800 mhz)
-VRAM manual 1.35 volts

Temperatures are quite low: 30 degrees idle; 49 degrees in gaming and 64 degrees in Cinebench r20 multicore. (20 degrees room temperature.)

Never had a single crash or whea errors with these settings.

 

[Griffo77]

EDIT: When i got home from work, the computer would start for a second, then shut off straight away. I tried a few things, but the only thing that fixed it was to clear CMOS. Could it be that tge computer did not get enough power at the start cyclus?
Trying to apply same setting again.

You have to find your own sweet spot. Your temps are to high. Try to low your multiplier to 44 or less. Test and see what happens.
Check your RAM configuration. My system usualy prompts to clear CMOS with RAM issues.