![[H]ard|Forum](/styles/hardforum/xenforo/logo_dark.png){kind=logo}
Lots of issues for few weeks with my PC. 5900x / Dark Hero
- Thread starter Elric82
- Start date
Furious_Styles
Supreme [H]ardness
- Joined
- Jan 16, 2013
- Messages
- 4,520
Doubtful.
I had this same problem for some time. Fine for 67 months. Then regular crashes with the hierarchy error. I am certain its related to voltage and powering up the cores. It always happen at idle when I click some random app, usually Edge, and whoop. Black screen straight into restart.
I have updated bios, removed 2 sticks of ram, upgraded my PSU, reinstalled my cpu and reinstalled Windows.
Now it works most of the time. One crash per week.
At one point I managed to provoke a crash using Ryzen Master to powercycle the cpu's. I haven't retried it, and it might just be how Ryzen Master works.
I have a 5900x on an Asus wifi board, with a Seasonic 1000 gold I think it is. Changed all power cables too.
I have updated bios, removed 2 sticks of ram, upgraded my PSU, reinstalled my cpu and reinstalled Windows.
Now it works most of the time. One crash per week.
At one point I managed to provoke a crash using Ryzen Master to powercycle the cpu's. I haven't retried it, and it might just be how Ryzen Master works.
I have a 5900x on an Asus wifi board, with a Seasonic 1000 gold I think it is. Changed all power cables too.
GoldChain
n00b
- Joined
- Jan 6, 2016
- Messages
- 62
This is telling me that you have a core, or chiplet that is sensitive to voltage fluctuations. Essentially the curve optimizer is going to try to undervolt the cores until its unstable. What I would do is set your Load Line Calibration to a flat line, leave PBO off, or RMA the chip. Also, I know you got a new PSU, but adding a battery backup might also alleviate some of your problems. Hard V Drop from a wall fluctuations can show through in a PSU and the battery backup is the easiest way to stop that.
I too have been having this issue on my 5950X on the Dark Hero. Started happening this summer, so I was thinking it was related to the increased ambient temperature. I turned off PBO and went back to stock settings and it was still crashing from time to time. So what I've been doing to try and solve it is turn PBO back on and use a positive offset (instead of the usual negative) on the Core that keeps crashing according to the Windows Event Viewer. I've been increasing the positive offset by +1 everytime it crashed (about once a week). I'm at +3 at the moment. We'll see if it holds. If all else fails, I like your idea of using LLC to try and get it to stabilize since it's clearly a problem with idle power values and its transitions.
Else, anyone have experience RMAing one of these chips with AMD for similar issues ? Are they hard to deal with ?
I RMA'd mine to the seller. They ran some stress tests, memtest, prime95 and some gpu test on it and declared it was fine. Billed me $50 and sent it back.
Now that I have a way to reproduce the problem, I'm considering what to do next. It took them 14 days last time to run that. Thats a long time to not have my deck.
Tamlin_WSGF
2[H]4U
- Joined
- Aug 1, 2006
- Messages
- 3,114
Positive offset is what is often used to solve crashing with ASUS PBO fmax enhancer and 5XXX series if I remember correctly. Try turning that off. Many reports that its not that good with Zen3 and more for Zen2 CPUs.
I've been using the AMD PBO from the overclocking menu and not the one provided from ASUS Extreme menu, if that's what you meant. Either way, with or without PBO it was crashing about once a week.
Have you considered contacting AMD directly ? Reading on their website, it does looks like they expect you to go with your dealer first, if you are still under warranty with them.
But yah, I couldn't wait that long for sure.
Some updates =)
I've flashed the 4006 bios and reinstalled windows.
I've reset my OC settings as they were before and got rid of the custom power plan I was using to use the balanced one. Temps are a bit higher on idle but the same or even lower during gaming actually.
I changed the internal usb of the ryujin aio
All seems stable for few days, I've played for few hours and work for few days, no issues so far..
I've flashed the 4006 bios and reinstalled windows.
I've reset my OC settings as they were before and got rid of the custom power plan I was using to use the balanced one. Temps are a bit higher on idle but the same or even lower during gaming actually.
I changed the internal usb of the ryujin aio
All seems stable for few days, I've played for few hours and work for few days, no issues so far..
Last edited:
Tamlin_WSGF
2[H]4U
- Joined
- Aug 1, 2006
- Messages
- 3,114
Did you disable PBO fmax enhancer, not the PBO from the AMD OC menu? Apperantly, auto might means its on and it might be enabled even though you use AMD PBO (it also reverts after bios update):
Ahh yes I get what you mean now. No it wasn't disable and was set on auto. I will get rid of it. Thank you for the tip!
Last edited:
Tamlin_WSGF
2[H]4U
- Joined
- Aug 1, 2006
- Messages
- 3,114
No problem! Many have had that issue that have been directly related to the fmax enhancer and 5XXX series, believing that it was their CPU or other things. Might not be your problem, but it have solved it for others turning this feature off.
I hope you're right! While not the end of the world considering its not that frequent, random crashes are still annoying.No problem! Many have had that issue that have been directly related to the fmax enhancer and 5XXX series, believing that it was their CPU or other things. Might not be your problem, but it have solved it for others turning this feature off.
Vengance_01
Supreme [H]ardness
- Joined
- Dec 23, 2001
- Messages
- 7,215
Core VIDs are perfectly safe. They are 1.450 at full load. 5900x should be safe up to 1.550, but its a bit high. Here is from AMD themselves:
https://www.reddit.com/r/Amd/comments/ciajef/comment/hnk8wy1/?context=3
Since your issues started after you installed the cooler, could you have made a mistake? There are some AIO mistakes to be made, even by those that have build computers their whole life
I can't even imagine a 5900x at that voltage. Shit my 5600 at 1.2ish running full tilt gets up there in temps....
Tamlin_WSGF
2[H]4U
- Joined
- Aug 1, 2006
- Messages
- 3,114
5900x and 5600 are different beasts and you must remember we are not talking about static voltage here and we are talking about stock settings (where it regulates itself within specs).
He has below 60C with 1.450V and boost at 4750mhz, spikes/highs at 70.5C with 1.531V at 4.950mhz @ stock settings. This is how modern CPU works and is within safe limits. Short boosts with higher voltage and frequency is normal boost behavior when temps and load allows it. Higher substained boosts with higher voltage is also normal when temps and load allows it, though it often uses more voltage "just in case" then what is nessesary normally for the higher boost @ stock for stability reason and CPU variations. If you have issues with temps on a 5600 @ 1.2V @ stock settings on full load, I would really check if the cooling is done right.
sounds like what going on with my 3960x rig was fine with 4.45 pob/ 1850 mem core for 3 years then stated crashing under no load after a power cycle. After many many hours of troubleshooting had to settle for 4ghz fixed all core and for some reason the imc is very "weak" now and 1600/3200mhz is the best it can do.
I've seen enough reports about amd 3x+ chips acting up that im not sure I would buy another amd cpu.
I had the same problem with Intel HEDT chips when maxing out the RAM and running it as fast as it would go, 2.5-4 years lifespan before they all had to go back to stock ram or got RMAd due to no longer being able to do even that.
It's really driving me nuts.
It was fine for like two weeks
Monday, my pc froze completely, Ryujin screen turned off , had to restart and all was ok after that.
Today, I was just browsing and...
Screen turned black, fans turned off.
Ryujin screen turned off
Q code 00
I had lots of issues since I've installed the Ryujin 2 . Could it be faulty and cause that kind of problem or is it more motherboard related?.
It was fine for like two weeks
Monday, my pc froze completely, Ryujin screen turned off , had to restart and all was ok after that.
Today, I was just browsing and...
Screen turned black, fans turned off.
Ryujin screen turned off
Q code 00
I had lots of issues since I've installed the Ryujin 2 . Could it be faulty and cause that kind of problem or is it more motherboard related?.
Small update on this. Too make a long story short, I figured that my problem was probably coming from the latest bios (4201). I almost went back to the previous version, but decided to try and reload the bios default settings first and run full stock too see what would happen (RAM included). Rebooting the system at stock seemed to have done something for some reason. I was able to incrementally reapply all my previous aggressive settings (RAM, PBO, etc. with fmax explicitly disabled) without any immediate crashes. Anyway, I ended up reading about CoreCycler and was able to figure out a much more stable curve for each core. What I though was stable really wasn't, with some cores throwing errors in Prime95. I would recommend anyone having issue with crashes to give that little script a try.No problem! Many have had that issue that have been directly related to the fmax enhancer and 5XXX series, believing that it was their CPU or other things. Might not be your problem, but it have solved it for others turning this feature off.