PSA: AMD AM5 issue with over voltage requiring BIOS update
- Thread starter dansilvkai
- Start date
Reading between the lines, along with the whispers in the shadows, it seems it may have been an error on AMDs part. It seems more likely as every board manufacturer has the same issue. This suggests that it's been a spec configuration handed to them and set by AMD, otherwise there would be variation.
The V1202 BIOS isn't the SoC patch though, it was just the latest that was available the other day when I was updating. The SoC patch is 1301 on BETA which I'll install tonight. I wrongly though it to be 1203 but got the naming wrong.
I fully expect my RAM to fail with this update, but we shall wait and see. I've got it running at 6400Mhz right now
I was in touch about my build in the queue awaiting production. Confirmed that they do install the latest BIOS but only if it's verified and tested.
I think for now when we get our systems that turning EXPO off until resolved should help alleviate any fears and then when fully fixed with new BIOS updates it should be fine to turn back on. Best to take a bit of a performance hit for a short period vs. a fried board!
I think for now when we get our systems that turning EXPO off until resolved should help alleviate any fears and then when fully fixed with new BIOS updates it should be fine to turn back on. Best to take a bit of a performance hit for a short period vs. a fried board!
And this is for the MB I chose. Severely doubt PCS will install this in a beta but hopefully get a fully verifed non beta shortly.
TUF GAMING X670E-PLUS WIFI BIOS 1412
Version 1412
Beta Version
9.11 MB
2023/04/25
SoC voltage for Ryzen 7000X3D series limited to a maximum of 1.30V to protect the CPU and motherboard.
Please note that this is a beta BIOS version of the motherboard which is still undergoing final testing before its official release. The UEFI, its firmware and all content found on it are provided on an “as is” and “as available” basis. ASUS does not give any warranties, whether express or limited, as to the suitability, compatibility, or usability of the UEFI, its firmware or any of its content. Except as provided in the Product warranty and to the maximum extent permitted by law, ASUS is not responsible for direct, special, incidental or consequential damages resulting from using this beta BIOS.
TUF GAMING X670E-PLUS WIFI BIOS 1412
Version 1412
Beta Version
9.11 MB
2023/04/25
SoC voltage for Ryzen 7000X3D series limited to a maximum of 1.30V to protect the CPU and motherboard.
Please note that this is a beta BIOS version of the motherboard which is still undergoing final testing before its official release. The UEFI, its firmware and all content found on it are provided on an “as is” and “as available” basis. ASUS does not give any warranties, whether express or limited, as to the suitability, compatibility, or usability of the UEFI, its firmware or any of its content. Except as provided in the Product warranty and to the maximum extent permitted by law, ASUS is not responsible for direct, special, incidental or consequential damages resulting from using this beta BIOS.
And at least we know to disable EXPO if not, which is extremely easy to do in the BIOS (just flipping a drop down menu) until we get a fully verified fix.
As much as BETAs can often be sketchy, the only reason this is BETA is because they have impacted a fix to one particular setting and haven't had the time to check the impact on the other aspects.
It's not really a BETA like the typical use of the term. There shouldn't be anything sketchy in play with it so it may be an exception to the rule.
It's not really a BETA like the typical use of the term. There shouldn't be anything sketchy in play with it so it may be an exception to the rule.
Figure this one out. Installed the BETA for the ProArt (full version not available) and I ended up with a performance boost
I had configured MSFS to be right on the ragged edge of its capability with the GPU running 90+% continuously while forcing 40fps in the headset. Went in for a game and it was around 80%. Bumped up a number of settings to high, then ultra and it finally started to struggle. Pulled the clouds back to medium and it's now where it was before.
Same with PC2, managed to max out all the settings and I'm getting a solid 90 FPS. I had to put MSAA to medium previously and got some reprojection from time to time. After the update, it was flat, so moved it to high.... it's still flat. Turned every setting to max.... still flat. Can't figure it out but I've got more performance than I had before.
I think I had it on the edge of stability before. It was scoring well, but stumbling from time to time. This new cap on the SoC has tightened it a bit and it's not on the edge anymore, no stumbling and thus more gaming performance.
Oh, it now sits at 1.24v as well and I didn't need to touch any of the RAM settings at all.
To think.... this is only going to get better with improvements being made
I had configured MSFS to be right on the ragged edge of its capability with the GPU running 90+% continuously while forcing 40fps in the headset. Went in for a game and it was around 80%. Bumped up a number of settings to high, then ultra and it finally started to struggle. Pulled the clouds back to medium and it's now where it was before.
Same with PC2, managed to max out all the settings and I'm getting a solid 90 FPS. I had to put MSAA to medium previously and got some reprojection from time to time. After the update, it was flat, so moved it to high.... it's still flat. Turned every setting to max.... still flat. Can't figure it out but I've got more performance than I had before.
I think I had it on the edge of stability before. It was scoring well, but stumbling from time to time. This new cap on the SoC has tightened it a bit and it's not on the edge anymore, no stumbling and thus more gaming performance.
Oh, it now sits at 1.24v as well and I didn't need to touch any of the RAM settings at all.
To think.... this is only going to get better with improvements being made
dansilvkai
Member
My build has also gone into testing, I did speak to someone on the phone and we went through this issue and I mentioned my concerns. I had asked them to update my BIOS and told them its in BETA but they seemed to be okay to put my message across, I will have to check and see what version is installed on arrival but its good they are willing to listen and help, heck they even told me about a conflict in my RAM choice.. having issues of reverting back to default settings randomly.. this was sorted swiftly.. In all with all thats going on I can't fault PCS.
SpyderTracks
We love you Ukraine
GamersNexus have done a pretty deep dive on this issue, and are just about to upload a video, I'll post it as soon as it's available.
SpyderTracks
We love you Ukraine
Gamers Nexus video is in, I'll add a TLDR once I've watched it through
TLDR: So it's a compounded issue from both AMD on the CPU and board partners (Asus and Gigabyte tested but assume all partners affected) messing up their over voltage protections on the motherboard.
BUT it's an issue that compounds over time, say 6 months or so, it's almost impossible to just trigger instantly, it's something that takes months of overvoltage slowly baking the CPU.
To protect: Flash latest BIOS (preferably not Beta)
For now, don't leave the PC on overnight, and don't overclock the CPU (including PBO)
Both are solvable within the BIOS revision but the current SOC voltage limit that AMD have applied is just an umbrella protection to prevent the issue occuring while they sort out a proper fix, further updates are required from board partners for a full fix.
This is NOT an EXPO issue, EXPO doesn't adjust SOC voltages at all, that's down to the board partners to set. That needs to be clear as EXPO is a really good feature, don't go blaming it.
As found on the Asus boards they specifically tested, Asus were reporting 1.35V set on the SOC in the BIOS options, but were actually pushing over 1.4V. This is lazy implementation by Asus probably to increase compatibility for more RAM DIMMS, but it has to be addressed by Asus and likely better direction given by AMD in future on limitations like this.
This needs further investigation, and GN are liaising with a 3rd party silicon evaluator, so there will be further updates on this in the coming weeks.
TLDR: So it's a compounded issue from both AMD on the CPU and board partners (Asus and Gigabyte tested but assume all partners affected) messing up their over voltage protections on the motherboard.
BUT it's an issue that compounds over time, say 6 months or so, it's almost impossible to just trigger instantly, it's something that takes months of overvoltage slowly baking the CPU.
To protect: Flash latest BIOS (preferably not Beta)
For now, don't leave the PC on overnight, and don't overclock the CPU (including PBO)
Both are solvable within the BIOS revision but the current SOC voltage limit that AMD have applied is just an umbrella protection to prevent the issue occuring while they sort out a proper fix, further updates are required from board partners for a full fix.
This is NOT an EXPO issue, EXPO doesn't adjust SOC voltages at all, that's down to the board partners to set. That needs to be clear as EXPO is a really good feature, don't go blaming it.
As found on the Asus boards they specifically tested, Asus were reporting 1.35V set on the SOC in the BIOS options, but were actually pushing over 1.4V. This is lazy implementation by Asus probably to increase compatibility for more RAM DIMMS, but it has to be addressed by Asus and likely better direction given by AMD in future on limitations like this.
This needs further investigation, and GN are liaising with a 3rd party silicon evaluator, so there will be further updates on this in the coming weeks.
Last edited:
SpyderTracks
We love you Ukraine
Yes, SOC is dynamic and will fluctuate, you take the max reading. You can monitor this (and it will state the max for that run) with Hwinfo, you don't need to monitor it within the bios
SpyderTracks
We love you Ukraine
I'm quickly turning to HWInfo over HWMonitor, it's a little more advanced, but seems to have better support and compatibility. I've had it a number of times now where HWMonitor fails to show any CPU temps
HWMonitor and HWInfo both report it fine with the 7800X3D.
I turned to HWInfo after I failed to spot the SoC label in HWMonotor, only to find it on a revisit
Make sure EXPO is disabled and you won't have any issues. When turning on EXPO the boards cap the SoC at 1.35v, but it often creeps to 1.4. With EXPO off, it sits around 1.2v, peaking around 1.25v.
The update sees similar values even with EXPO enabled
I turned to HWInfo after I failed to spot the SoC label in HWMonotor, only to find it on a revisit
Make sure EXPO is disabled and you won't have any issues. When turning on EXPO the boards cap the SoC at 1.35v, but it often creeps to 1.4. With EXPO off, it sits around 1.2v, peaking around 1.25v.
The update sees similar values even with EXPO enabled
Still waiting on my system, currently in "Testing - problem" stage, hopefully nothing too serious!
I can see the BETA AGESA BIOS update for my board on the site now, assume I will have to update when my system gets here. Will ask for permission of course (as it may impact warranty) but I guess we have to update regardless. It's too important an update not to. ASUS have automated software in Windows too that does it all for you so risk is very low as long as you are not using BETA versions.
I can see the BETA AGESA BIOS update for my board on the site now, assume I will have to update when my system gets here. Will ask for permission of course (as it may impact warranty) but I guess we have to update regardless. It's too important an update not to. ASUS have automated software in Windows too that does it all for you so risk is very low as long as you are not using BETA versions.