• Some users have recently had their accounts hijacked. It seems that the now defunct EVGA forums might have compromised your password there and seems many are using the same PW here. We would suggest you UPDATE YOUR PASSWORD and TURN ON 2FA for your account here to further secure it. None of the compromised accounts had 2FA turned on.
    Once you have enabled 2FA, your account will be updated soon to show a badge, letting other members know that you use 2FA to protect your account. This should be beneficial for everyone that uses FSFT.

My 9800X3D died

Dude, use both CPU power cables, lol.
That doesn’t really matter. Each of the those EPS 8pin cables are rated to deliver 300-380w power. You can easily and without issue run most if not all modern CPUs on a single 8 pin cable. It’s only for serious overclocking that you actually need to use both cables. Most motherboards have 2 8pin cables for redundancy and to slightly lower the temps of each connector. With a 9800X3D only really drawing 150-160w of power at full load he was in no danger of even getting close to the capacity of a single cable much less 2 of them.
 
load on PSU or mobo VRMs?
Cables and connectors, don't think it makes much difference for psu or vrms. Although maybe it helps certain vrm designs at high load.
 
I hope this doesn't become a 9800X3D issue like Intel's 13th and 14th gen. I've heard and read rumblings.
The worrying thing for me is how the vcore vid max voltages spike into the 1.45s by default. Which was what killed the 14900ks etc although I think the Intel chips were a little more "generous" on the voltage spikes prior to all the latest BIOS fix patches but the ryzens are pretty high up too imo.
 
  • Like
Reactions: Niner
like this
That doesn’t really matter. Each of the those EPS 8pin cables are rated to deliver 300-380w power. You can easily and without issue run most if not all modern CPUs on a single 8 pin cable. It’s only for serious overclocking that you actually need to use both cables. Most motherboards have 2 8pin cables for redundancy and to slightly lower the temps of each connector. With a 9800X3D only really drawing 150-160w of power at full load he was in no danger of even getting close to the capacity of a single cable much less 2 of them.
why? heard its for 300+ W
I mean, sure, just use one CPU power cable when there's two CPU power plugs. I just take the side of caution since I don't own any design schematics on how the power delivery works for the motherboard. Maybe there's load balancing, maybe there's more stability, maybe maybe maybe...
 
  • Like
Reactions: haste
like this
The worrying thing for me is how the vcore vid max voltages spike into the 1.45s by default. Which was what killed the 14900ks etc although I think the Intel chips were a little more "generous" on the voltage spikes prior to all the latest BIOS fix patches but the ryzens are pretty high up too imo.

VID can be ignored if the Vcore is between the correct margin, VID can have very high spikes even on idle and mean nothing, vid is the voltage identificator requested by the CPU to the motherboard, but not necessarily what the CPU is receiving, before that is applied vdroop and LLC and the CPU receive the applied voltage. and the Vcore at 1.392 without any manual adjustment isn't bad for those cpus.
 
VID can be ignored if the Vcore is between the correct margin, VID can have very high spikes even on idle and mean nothing, vid is the voltage identificator requested by the CPU to the motherboard, but not necessarily what the CPU is receiving, before that is applied vdroop and LLC and the CPU receive the applied voltage. and the Vcore at 1.392 without any manual adjustment isn't bad for those cpus.
1771805995629.png

1771806018331.png
 
VID can be ignored if the Vcore is between the correct margin, VID can have very high spikes even on idle and mean nothing, vid is the voltage identificator requested by the CPU to the motherboard, but not necessarily what the CPU is receiving, before that is applied vdroop and LLC and the CPU receive the applied voltage. and the Vcore at 1.392 without any manual adjustment isn't bad for those cpus.
I suppose it's a divided crowd. Some users are ok with the vcore voltage kissing 1.4v, and some are definitely more comfortable keeping it below 1.3v somewhere in the 1.275v range. Also not sure if the juice is worth the squeeze for that extra 100 or 200 mhz. The 9850X3D is already receiving harsh criticism of how unimpressive and unnecessary boosting the voltage and power consumption up higher for a couple percent % of fps in the best cases. Not to mention the additional heat from higher voltage in my PC room it's not even a question for me to keep the voltage low for comfort of temperature and comfort piece of mind for the silicon longevity and stability.
 
Max safe SOC voltage is 1.3 and you are well below that.

Either the CPU had a random unknown failure. Or the motherboard firmware did something weird, which isn't indicated by the reported settings.
 
  • Like
Reactions: erek
like this
“We are pleased to inform you that your warranty claim has been approved for the following product(s) under RMA# Xxxxxxx. OPN 100-100001084WOF, Serial Number Xxxxxxx Please note this RMA is only valid for the product(s) with the OPN and Serial Number mentioned above. Unauthorized / unapproved items that are not listed above will be discarded or returned at your own cost.”
(Just moments ago)
 
“We are pleased to inform you that your warranty claim has been approved for the following product(s) under RMA# Xxxxxxx. OPN 100-100001084WOF, Serial Number Xxxxxxx Please note this RMA is only valid for the product(s) with the OPN and Serial Number mentioned above. Unauthorized / unapproved items that are not listed above will be discarded or returned at your own cost.”
(Just moments ago)
The king is dead... long live the king!
 
9800x3d on asus tuf gaming x670e plus wi-fi

is this weak? the 9800x3d is on the validated list

https://www.asus.com/us/motherboard...qvl_cpu?model2Name=TUF-GAMING-X670E-PLUS-WIFI
I think it's fine, solid VRM's. Was validated since BIOS 2613. It's just difficult to know what caused the issue without a lot of data. Not just from you, but from AMD, board manufacturers, other people, and custom computing companies like Puget Systems. If it's not a CPU batch problem (AMD), or a motherboard BIOS problem, it's typically BIOS voltage settings or user error. High voltages (VCore or SoC), poor cooling, or maybe the CPU cooler was too tight on one side. Who knows.
 
  • Like
Reactions: erek
like this
This is why I generally always keep an eye on the voltages the for the mb/CPU. Very interested what they were previously. Both Intel and AMD run voltages higher than they should be shortening the CPUs lifespan. Try to always run hwinfo so you can see the average and max spikes after gaming sessions.
I believe even one of AMD's guys said something about 1.20v or something being safe but anything over causes degradation. Its why I use that guide to control my volts to 1.20v while keeping my temps at 80C max.

Found it...
https://www.reddit.com/r/ryzen/comments/1gq1yu9/universal_guide_to_configuring_all_ryzen_9000/
 
I believe even one of AMD's guys said something about 1.20v or something being safe but anything over causes degradation. Its why I use that guide to control my volts to 1.20v while keeping my temps at 80C max.

Found it...
https://www.reddit.com/r/ryzen/comments/1gq1yu9/universal_guide_to_configuring_all_ryzen_9000/
Typically 1.1V's is fine. If you're overclocking RAM, sometimes you'll need it higher. But scaling the SoC voltage down and testing your stability is the best practice.
 
Stock ootb may enable pbo on certain boards, I know there were a few am4 boards, i think, which did this. Or if not default, then if you enabled xmp/expo. I don't know if they still do, there was a fair amount of backlash over it.
 
Sure. Point is, a 14 phase VRM is overkill. For all current AM5 CPUs, actually.
what’s the deal with all these 16 and now 18+ phase boards being recommended anymore?

https://wccftech.com/best/best-motherboards-for-amd-ryzen-7-9850x3d/

View attachment 787824
“The 24+2+2 power phase design saw a peak of 62°C, but after half an hour dealing with our Ryzen 9 9950X at full load, only one thermal probe topped 60°C with the rest sitting in the mid to high 50s. A decent result thanks to a large dual heatpipe upper heatsink array and rear cooling too.“

https://www.techpowerup.com/review/asus-rog-crosshair-x870e-glacial/5.html
 
  • Like
Reactions: Niner
like this
1772151303120.png


---

also ASUS just released a Beta bios and mentioning my generation of CPUs,

Version 3841
Beta Version17.45 MB2026/02/26SHA-256 :816FA8D561BAF2AEA75BD2643D257390D566E61BEA82AD4601B77366EDDA56CE"

1. AGESA updated to ComboAM5 PI 1.3.0.0a.
  1. Provide additional stability margin during high-frequency DDR5 training.
  2. Addresses potential boot failures and stability issues observed on certain Ryzen 9000 configurations.
  3. Resolves an issue where systems might unexpectedly prompt for a BitLocker recovery key during startup.
 
I mean, sure, just use one CPU power cable when there's two CPU power plugs. I just take the side of caution since I don't own any design schematics on how the power delivery works for the motherboard. Maybe there's load balancing, maybe there's more stability, maybe maybe maybe...
Just FYI CPU power connectors are wired together. It's just to spread the load across more contact points, it won't hurt to run more cables and I would say you may as well but if the power supply only has 1 then it's not an issue.
 
Yes AMD RMA is pretty solid. I RMAed a 5600g without any issues.
 
  • Like
Reactions: erek
like this
Back
Top