Overclock.net banner
12981 - 13000 of 13162 Posts

·
Overclocker
Joined
·
7,038 Posts
@Katana1074

Just like you said man, yesterday was fine and now i got WHEA error again when testing with OCCT...

There must be something wrong with Windows 10 or its an CPU issue... What an clusterfuck of problems man..

Its not related to only 3000 serieus because even 5000 series CPU's have lots of issues man... A friend of mine switched to Intel a couple of months ago because of this and hes had not a single issue ever since he switched platforms..

I contacted my retail store and they said i could return the RAM but who's saying its an RAM issue? These problems did start when i switched to my new Samsung 970 EVO Plus SSD AND new G.Skill RAM... But if i test both i see no errors only interconnect according to WHEA....
 

·
Registered
Joined
·
2,329 Posts
I haven't gotten any WHEA errors with this new BIOS and hadn't seen any with the previous version. What I see now are just warnings saying "corrected hardware error occurred".
 

·
Registered
Joined
·
20 Posts
@Katana1074

Just like you said man, yesterday was fine and now i got WHEA error again when testing with OCCT...

There must be something wrong with Windows 10 or its an CPU issue... What an clusterfuck of problems man..

Its not related to only 3000 serieus because even 5000 series CPU's have lots of issues man... A friend of mine switched to Intel a couple of months ago because of this and hes had not a single issue ever since he switched platforms..

I contacted my retail store and they said i could return the RAM but who's saying its an RAM issue? These problems did start when i switched to my new Samsung 970 EVO Plus SSD AND new G.Skill RAM... But if i test both i see no errors only interconnect according to WHEA....
Are you overclocking your Bclk past 100?

Only reason i ask is my system doesnt like it at all,I would reflash your bios and load defaults then start from scratch and test docp timings 1st

Yeah i know what ya mean about ryzen being problematic, ive seen all the posts all over the net about it, it does make me wonder if there is some fundamental flaw somewhere

that maybe Amd are not letting on

Could be worse im sat here with the inlaws lenovo laptop that has shat its self,no backups ffs mind you its a ryzen 5 Ha !
 

·
Registered
Joined
·
1,807 Posts
makes me wonder if its worth going back to 2606....
only if you dont care about auto bullshit boosts and set basically everythingh on manual instead of auto.

Not had my old profiles so took me only 2 weeks to get settings back to sweet spots. So pc wont random reboot.
 

·
Registered
Joined
·
20 Posts
well i do use pbo and resizeble bar, id prob gain something back in pcie 4.0
 

·
Registered
Joined
·
215 Posts
All them new bioses are ****ing waste of time on Zen3
View attachment 2485081

Zero Errors Reboots problems...
I have had ZERO WHEA errors on previous bios and ZERO with the new "latest" one. It might be something to do with people's RAM OC.
 

·
Registered
Joined
·
35 Posts
All them new bioses are ****ing waste of time on Zen3


Zero Errors Reboots problems...
pardon me...what do you mean exactly? i am on 4201 with 5900X and never experienced any problems. No WHEA error no USB problems...no nothing! am i just lucky or whats the fuzz all about?

makes me thinking if i should flash the newest bios^^
 

·
Registered
Joined
·
20 Posts
Im gonna stick out with bios 4301, ive had one WHEA error in the last 24 hours, more tweaking to come,...
 

·
Overclocker
Joined
·
7,038 Posts
Very strange that one is affected and the other is not.. This month there is a new BIOS coming that supposedly fix the USB issues people are having, i wait for that and and in the mean time i stay on 3103 as that was the best BIOS for me so far. Its just very poor AMD AGESA code and Asus or any other manufacturer cannot fix this as its on AMD side..

Im gonna look over on the AMD forums to see if i can get some explanation about this erratic behavior as i did everything to prevent or fix this with no result so far. Im testing the 3103 BIOS in the mean time too.
 

·
Registered
Joined
·
10 Posts
Well, I'm still on 4007 bios.

Seems to be the best for me.

Currently running a 5800x @4.8ghz 1.232v with the Der8aer OC Bracket max 63c on cinebench.

Ram running at 3600mhz 14-15-13-28 @1.43v Flare x B die.

Not a single issue, I think this is the best stability I got out of Ryzen.

The ram is stable at 3800mhz 16-16-16-36 but my gaming results are actually worse and sometimes I get the stupid F9 error on a cold boot. 3600mhz is perfectly stable and no F9 error.

2485222


2485226
2485227
2485223
 

·
Registered
Joined
·
1,807 Posts
pardon me...what do you mean exactly? i am on 4201 with 5900X and never experienced any problems. No WHEA error no USB problems...no nothing! am i just lucky or whats the fuzz all about?

makes me thinking if i should flash the newest bios^^
missclicked zen2 well new one as i see it is zen2+ Basically using 4000+ bioses on Zen2 is waste of space but they work fine on zen3. @Katana1074

Update on 2606 @crakej
2485237


55 hours constant 100% load zero oproblems zero errors

@hurricane28 its not STRANCE at all cause new bioses operate memory subsystem other way. Bioses pass 3000 will all be crap and theert is no point to use them bioses anyway cause they are slower on zen2. You reverted to old bios and you are testing ?? Why to **** u dont have profile that you had on that bios ??
 

·
Overclocker
Joined
·
7,038 Posts
On my end it has nothing to do With BIOS man.. there must be some compatibility issue or something as this WHEA error also occurs on 3103 BIOS now.. Never had this before. Or it could be Windows acting up. Strange thing is that my memory is perfectly stable and have no issues until the error pops up in hwinfo64 and i need to restart system because it becomes unstable...

I read somewhere that 3000 series CPU's can go bad really quick and that this can cause these issues.. lets hope not as this CPU is pretty good OCer...
 

·
Registered
Joined
·
25 Posts
you cant compare 3000 series to 5000. There is a ~10ns reduction by default.

nice result though. may i ask the voltages you run the ram?
thanks m8

1.15 SOC
1.45 dram
1.5 dram boot
rest is auto
 

·
Overclocker
Joined
·
7,038 Posts
thanks m8

1.15 SOC
1.45 dram
1.5 dram boot
rest is auto
Would you mind share your results and settings? Like a screenshot of zentimings? Thnx
 

·
Overclocker
Joined
·
7,038 Posts
timings and results quoted above
will post complete bios settings when i get home if you want
Thank you, looking forward to it.
 

·
Registered
Joined
·
26 Posts
On my end it has nothing to do With BIOS man.. there must be some compatibility issue or something as this WHEA error also occurs on 3103 BIOS now.. Never had this before. Or it could be Windows acting up. Strange thing is that my memory is perfectly stable and have no issues until the error pops up in hwinfo64 and i need to restart system because it becomes unstable...

I read somewhere that 3000 series CPU's can go bad really quick and that this can cause these issues.. lets hope not as this CPU is pretty good OCer...
I've had similar issues with my 3800X. Whenever I'd start a game (3D graphics load only, any benchmark without it or "fake" loads wouldn't trigger it) my PC would run fine for a while and then after 15-20 min it'd restart with no blue screen in between and a bunch of WHEA errors logged in Windows' event viewer. I ended up rebuilding my PC, remounting the MB, the RAM and the CPU, cleaning everything and reinstalling Windows from scratch. No issues after that. I think there maybe were a few issues with the thermal paste bleeding into the socket (I didn't apply much back then, but I did try to clean my CPU multiple times and there was probably a point where I moved some dried thermal paste into the socket between the CPU edge and the MB pins). Other possibility is Windows got corrupted: I use a Kingston PCI-e 3.0 x2 (not x4) SSD and where I live sometimes the power goes out, who knows. Everytime I'd run sfc /scannow and it'd find some corruption or error in the system. Final possibility is compatibility with something you use in your PC, especially if you use any fan control or RGB software. In my case, I use a lot of those, actually. My PC is filled with RGB strips from Corsair and also Corsair RAM and keyboard, so it all syncs up through iCUE. It's a garbage program that eats single-threaded CPU cycles but it gives me the option to turn it all off if I want to (or change colors, tho' I rarely use it for that, most of the time is static - another reason why it shouldn't eat CPU cycles as much as it does). And I know some of these softwares use CPU (and even GPU, in case of iCue for example) and that can lead to incompatibility.

So, maybe if you have nothing to do due to the pandemic and are still at home, you could try rebuilding everything from scratch, reseating CPU/RAM, format Windows, etc. Not that it's the only solution (or even that it is, you might do all that and still have problems, who knows). Just sharing my experience. Now I use basically everything on Auto and some RAM timings manual, but everything else (including VRM control and PBO) are all set to Auto/Default or Disabled (PBO).

P.S.: I use this CPU to browse the web 90% of the time, studying and Zoom calls at most. So I don't care if I'm not hitting 4.35GHz all the time. Every now and again I do play games and now it doesn't crash anymore, but it is very limited. Also, I'm on 3103 now (having had these issues soon after using any of the 4000+ bioses). I'm waiting for the next AGESA update, so I'm not really keen to trying BIOS at random because of all that I've gone through - at some point I thought my 5700XT was dead and now it works perfectly normally).
 

·
Overclocker
Joined
·
7,038 Posts
I've had similar issues with my 3800X. Whenever I'd start a game (3D graphics load only, any benchmark without it or "fake" loads wouldn't trigger it) my PC would run fine for a while and then after 15-20 min it'd restart with no blue screen in between and a bunch of WHEA errors logged in Windows' event viewer. I ended up rebuilding my PC, remounting the MB, the RAM and the CPU, cleaning everything and reinstalling Windows from scratch. No issues after that. I think there maybe were a few issues with the thermal paste bleeding into the socket (I didn't apply much back then, but I did try to clean my CPU multiple times and there was probably a point where I moved some dried thermal paste into the socket between the CPU edge and the MB pins). Other possibility is Windows got corrupted: I use a Kingston PCI-e 3.0 x2 (not x4) SSD and where I live sometimes the power goes out, who knows. Everytime I'd run sfc /scannow and it'd find some corruption or error in the system. Final possibility is compatibility with something you use in your PC, especially if you use any fan control or RGB software. In my case, I use a lot of those, actually. My PC is filled with RGB strips from Corsair and also Corsair RAM and keyboard, so it all syncs up through iCUE. It's a garbage program that eats single-threaded CPU cycles but it gives me the option to turn it all off if I want to (or change colors, tho' I rarely use it for that, most of the time is static - another reason why it shouldn't eat CPU cycles as much as it does). And I know some of these softwares use CPU (and even GPU, in case of iCue for example) and that can lead to incompatibility.

So, maybe if you have nothing to do due to the pandemic and are still at home, you could try rebuilding everything from scratch, reseating CPU/RAM, format Windows, etc. Not that it's the only solution (or even that it is, you might do all that and still have problems, who knows). Just sharing my experience. Now I use basically everything on Auto and some RAM timings manual, but everything else (including VRM control and PBO) are all set to Auto/Default or Disabled (PBO).

P.S.: I use this CPU to browse the web 90% of the time, studying and Zoom calls at most. So I don't care if I'm not hitting 4.35GHz all the time. Every now and again I do play games and now it doesn't crash anymore, but it is very limited. Also, I'm on 3103 now (having had these issues soon after using any of the 4000+ bioses). I'm waiting for the next AGESA update, so I'm not really keen to trying BIOS at random because of all that I've gone through - at some point I thought my 5700XT was dead and now it works perfectly normally).

thnx for your insight.

the problem with Ryzen is that its just very very erratic due to poor AGESA code from AMD as far as i understand from reps. This results in poor BIOS an can cause multiple issues.

The WHEA error can be caused by so many things. It can even be caused by .m2 nvme drives: https://www.reddit.com/r/Amd/comments/cd7pqb
It can also be caused by CPU degradation or sudden breakdown.https://community.amd.com/t5/processors/whea-18-again-i-m-done/td-p/430145

It can be caused by so many things. The problem for me is the RAM i think as im running my M.2 in the bottom slot of my board which is the correct one and doesn't get lanes straight from the CPU but the chipset and my GPU stays as x16 instead of x8 in the top slot. Not that there is a huge difference between x16 and x8 but still.

The WHEA error can also be caused by bad PSU, or RAM compatibility issue, the list is endless as it can be Windows update as well. Windows handles RAM very poorly so any slight misconfig can cause erratic behavior.

The thing is that my current RAM is not faulty and there are no compatibility issus as far as i know so i can rule that out, RAM is also stable so there must be something else that is causing this. I test for hours and hour with TM5 without getting a single issue but testing with OCCT or letting the PC "rest" or light work might trigger another WHEA...

It can even be caused by loading the DOCP settings... I really don't want to give up this platform but sometimes its just a clusterfuck of problems.
 

·
Registered
Joined
·
66 Posts
thnx for your insight.

the problem with Ryzen is that its just very very erratic due to poor AGESA code from AMD as far as i understand from reps. This results in poor BIOS an can cause multiple issues.

The WHEA error can be caused by so many things. It can even be caused by .m2 nvme drives: https://www.reddit.com/r/Amd/comments/cd7pqb
It can also be caused by CPU degradation or sudden breakdown.https://community.amd.com/t5/processors/whea-18-again-i-m-done/td-p/430145

It can be caused by so many things. The problem for me is the RAM i think as im running my M.2 in the bottom slot of my board which is the correct one and doesn't get lanes straight from the CPU but the chipset and my GPU stays as x16 instead of x8 in the top slot. Not that there is a huge difference between x16 and x8 but still.

The WHEA error can also be caused by bad PSU, or RAM compatibility issue, the list is endless as it can be Windows update as well. Windows handles RAM very poorly so any slight misconfig can cause erratic behavior.

The thing is that my current RAM is not faulty and there are no compatibility issus as far as i know so i can rule that out, RAM is also stable so there must be something else that is causing this. I test for hours and hour with TM5 without getting a single issue but testing with OCCT or letting the PC "rest" or light work might trigger another WHEA...

It can even be caused by loading the DOCP settings... I really don't want to give up this platform but sometimes its just a clusterfuck of problems.
Been running Ryzen since the launch of the 2700x and just upgraded recently to the 5800x, have had zero issues in all those years but i run things pretty much stock, are you trying for OC or are these issue's at stock settings?
 

·
Registered
Joined
·
215 Posts
Man he got probnlems since 2017 and ONLY fix is moving to intel for him. I had every zen besides new ones all super stable on max silicone. People talk bullshit about degradation my 3900x is under load on 1.337vcore after drop soon will be two years no problems.
I had bought into all Zen gens, 1700x, 2700x, 3900x, 5900x. The 2700x and 3900x had decent OC on them and a lot of play around none got degraded after a year. Both CPUs are still been used atm the 2700x for my mum pc and the 3900x my GF uses it for 3d modeling and animations. Intel silicone might be more mature but TSMC is not trash lol. The only "problem" zen has is RAM compatibility and some chipset issues that aren't deal-breaker at least for the majority of people.
 
12981 - 13000 of 13162 Posts
Top