ROG STRIX X670E-E GAMING WIFI Latest BIOS 0925

[Kei]

The Soc voltages left on Auto now with Expo Tweak enabled are showing 1.341 and 1.359 under Cinebench Tests and before was up to 1.492v. The Latest Bios has capped the Voltage for 7000X processors including my 7950X. Performance is still good though and scores 37,480 multi core test in Cinebench. It was a shame this issue wasn't addressed earlier as we all don't know what stress it put the processor under to begin with. I also think we are going to see tweaks to bioses throughout May as The Motherboard Venders address all voltages. Cheers

Agreed, I think there will be quite a few tweaks from all of the vendors for the next 30-45 days or so. I thankfully never saw any crazy high voltages on my rig whether it was the 7700x or the new 7800x3D. I definitely never saw anything as sky high as your 1.492v! I wasn't looking because of the (unknown at the time) issues that are going m right now...I just like to check out my systems pretty thoroughly when I install them or make a big update. I just so happened to get lucky to have bought the 7700x in January and then the 7800x3D in April.

Hopefully all is well for the long term, but I have zero regrets with my purchase as both processors had been more than fantastic for me. If something had(does) happen, I'd still buy another one the same day. I imagine I'd think differently if it weren't for VR and simulations that the 3D processors are so good at.

I also am more lenient because of my career and a very deep understanding of how unknowns can pop up even after testing and problem solving in regards to quality control and assurance. The processor industry is one I hold in extremely(!!!) high regard with quality control/assurance and how well they've gotten their yields dialed in over the years. I'm sure there will be an extreme deep dive behind closed doors to improve the process over time.

 

[Nam-A-M-D]

Hi All. After doing some testing i have found setting the Soc Votage to manual and using 1.225v keeps the voltage down to 1.270v - 1.279v on Ryzen 7950X as it is still high at 1.350v on auto. It is far better to lower it and be safe. I will also do some more tests over the next 48 hours to see if i can go any lower stable. Hopefully X3D processors should be ok now due to the restrictions set in place, but It has amazed me that Asus motherboards are one on the highest setting Soc voltages than some of the other vendors. Steve from Gamers Nexus did a video which you can watch here:-

.

I don't think this is the last we are going to see with Over Current Protection voltages within the settings. Not everyone uses the same hardware which in itself will bring different issues and tollerances. We all want to trust the vendors, but this neglect until now may cause our processors to die earlier than expected. I have had my 7950X for five months now with no update to the bios until Feb 2023 which only stated memory optimisation support and stability. Nothing came to light until someone stated their processor basically blew up in the socket. Asus just brushed it all under the carpet by removing the other bioses they had on their website. They only do something when they get found out which is not good enough when it costs so much money to move to the new platform. Cheers

 

[DAMEND]

In 1401ver BIOS.I try use offset mod.I set -0.02v offset of the cpusocV in my X670EEGAMING.

I get socV up to 1.22v.Memory set in 6000c30 expotweaked.After 2days working,it's stable in my working.Hope that helps,Cheers.

 

[Mussels]

The Soc voltages left on Auto now with Expo Tweak enabled are showing 1.341 and 1.359 under Cinebench Tests and before was up to 1.492v. The Latest Bios has capped the Voltage for 7000X processors including my 7950X. Performance is still good though and scores 37,480 multi core test in Cinebench. It was a shame this issue wasn't addressed earlier as we all don't know what stress it put the processor under to begin with. I also think we are going to see tweaks to bioses throughout May as The Motherboard Venders address all voltages. Cheers

Explains a lot, these boards were kicking up voltages on auto in the hope they'd make them appear to be better overclockers (expecting droop under load or something)

Pretty sure all future BIOS updates are gunna include some more aggressive voltage limits that the board makers cant bypass, and AMD will put the pressure on them to stop screwing with voltages and VRM's so badly (It's not like power delivery issues are a new problem)

 

[Nam-A-M-D]

There is a new bios again.
ROG STRIX X670E-E GAMING WIFI BIOS 1410

Version 1410 Beta Version
9.21 MB 2023/05/02
"1. Update AGESA version to Combo AM5 PI 1.0.0.7.
2. Support 48/24GB high-density DDR5 memory module.
3. SoC voltage for Ryzen 7000 series limited to a maximum of 1.30V to protect the CPU and motherboard.

Before running the USB BIOS Flashback tool, please rename the BIOS file (SX670EE.CAP) using BIOSRenamer.

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."

ROG Strix X670E-E Gaming WiFi Motherboard

ASUS ROG Strix X670E-E Gaming WiFi features 18+2 power stages, DDR5 support, four M.2 slots with heatsinks, PCIe® 5.0, USB 3.2 Gen 2x2 front-panel connector, and onboard WiFi 6E. It also features AI Cooling II, AI Networking, Two-Way AI Noise Cancelation. ROG Strix X670E-E Gaming WiFi caters to...

rog.asus.com

-----------------------------------------------

I hope this one works well

 

[Ryodu]

I wish I would know what this

"1. Update AGESA version to Combo AM5 PI 1.0.0.7."

actually changes compared to 1.0.0.6

 

[Nam-A-M-D]

I wish I would know what this

"1. Update AGESA version to Combo AM5 PI 1.0.0.7."

actually changes compared to 1.0.0.6

AMD are working on a new AGESA update (version 1.0.0.7) that will add support for "non-binary" DDR5 24GB and 48GB DRAM modules to their AM5 platform. This will allow users of AMD's Ryzen 7000 series processors to support these new DDR5 DIMM to remove annoying compatibility issues.

 

[Mussels]

I wish I would know what this

"1. Update AGESA version to Combo AM5 PI 1.0.0.7."

actually changes compared to 1.0.0.6

2. Support 48/24GB high-density DDR5 memory module.
3. SoC voltage for Ryzen 7000 series limited to a maximum of 1.30V to protect the CPU and motherboard.


^ Those
They arent a seperate changelog, they're whats new


Some people have been perfectly fine with those new odd sized memory modules, but not everyone - I guess it mattered how many you used.

 

[Nam-A-M-D]

I have replaced the 7950x for a new one and i haven't had any further crashes like before. It looks like there has been damage done to the IGPU in the 7950X due to the Soc voltage exceeding 1.492v on workloads. The latest Bios has restricted this from happening now with the lower set voltage at 1.3v. I purchased mine originally from Amazon and i have now sent it back for a full refund to save the hastle of RMA with AMD. If mine has degraded this much in 5 months i am concerned that their are alot of processors out there also have been put under the same stress. I never overclocked the processor because i didn't need to but like most videos on Utube would not share information with anyone. AMD and Asus would guide people to leave most of the settings on Auto. Unfortunately the damage has been done and we all have to move on and deal with the issue head on. Anyone experiencing issues with performance or crashing then it would be a good idea to contact the place you brought the processor from for advice and possible RMA. Don't just leave it and put up with it as this issue will get worse over time. I will still test out the new Bioses as they come out and let you all know what i think. Cheers

 

[Kei]

I have replaced the 7950x for a new one and i haven't had any further crashes like before. It looks like there has been damage done to the IGPU in the 7950X due to the Soc voltage exceeding 1.492v on workloads. The latest Bios has restricted this from happening now with the lower set voltage at 1.3v. I purchased mine originally from Amazon and i have now sent it back for a full refund to save the hastle of RMA with AMD. If mine has degraded this much in 5 months i am concerned that their are alot of processors out there also have been put under the same stress. I never overclocked the processor because i didn't need to but like most videos on Utube would not share information with anyone. AMD and Asus would guide people to leave most of the settings on Auto. Unfortunately the damage has been done and we all have to move on and deal with the issue head on. Anyone experiencing issues with performance or crashing then it would be a good idea to contact the place you brought the processor from for advice and possible RMA. Don't just leave it and put up with it as this issue will get worse over time. I will still test out the new Bioses as they come out and let you all know what i think. Cheers

Thanks for posting your experience Nam. I haven't had any issues with thus far with my 7800x3D (or 7700x prior to this) outside of having the white error GPU light on my motherboard turn on during boot most of the time. That wasn't something new though and has pretty much been since the first boot with the 7800x3D and never with the 7700x it replaced. When I installed the 7800x3D I also had to install a new bios so it's not raised any flags since it happened exactly then.

To add to that, it ONLY happens when I use the display port cable but never on the HDMI cable. There are zero performance issues with anything and there were people talking about this issue prior to any of this stuff ever coming close to being an issue, specifically with the 4090. I've done the display port update that came out for the 4090 and the issue still happens but not every time.

Definitely keeping an eye on how things progress from here though just in case. Either way if something happened, I'd want another 7800x3D asap because it's great.

 

[ir_cow]

@Nam-A-M-D did you at least say it was bad? Otherwise it will be sold as open box to someone else:/

 

[Nam-A-M-D]

@Nam-A-M-D did you at least say it was bad? Otherwise it will be sold as open box to someone else:/

The processor is being returned faulty so it won't be sold again.

 

[Mussels]

To add to that, it ONLY happens when I use the display port cable but never on the HDMI cable.

To be fair, i've got a few displays around the house here that wont work on DP to enter the BIOS at boot, but work fine in the OS
Then the situation reverses on another GPU, where DP works and not HDMI

It's really bizarre what goes on with some monitors at times
(1440p 165Hz and a 1440p 144Hz, with intel IGP and nvidia 9/10 series GPUs)

Glad you got a refund and a replacement CPU, this is a really crappy thing to be happening that could be affecting a *lot* of users - even seeing the amount of people with DDR5 6000 has surprised me since I assumed most people knew that was an overclocking dream and not guaranteed, but i should have realised that people would google "the best" to get and go for it with high budget setups

 

[Nam-A-M-D]

Hi All.
I am not sure if anyone has seen the latest Gamers Nexus Video in regards the complete mess from Asus with the way they are implementing the Bioses and voltages. I will put a link below:

I have been checking the voltages and doing some tests and found a couple of things to set in the bios manually, which will bring down the voltages to a safer level. Originally i set the Soc voltage to 1.250v, but this was still pushing the voltage to 1.305v under stress testing with this bios. Asus set the max voltage to 1.3v which is wrong and I wouldn't be surprised if we don't get another bios revision out by the weekend. The CPU VDDIO/MC Voltage was also running at 1.350v which is also set to high. Asus need to take this serious and stop messing about and get this fixed else we are going to see a big increase in processor fails. Only time will tell though

The two settings which need to be changed are:
CPU SOC = 1.2v
CPU VDDIO/MC Voltage = 1.2v

These settings should work fine for all AM5 Motherboards
cheers

 

[Mussels]

Jayztwocents had the x670 crosshair board in his personal system, and he's had one die and one CPU die so far in the last two months.

At this stage no one knows if the lower end boards have the same problems with overvolting, EXPO profiles, and disabled safety features so... play it safe guys.

 

[Haytham]

New Bios Today

ROG STRIX X670E-E GAMING WIFI BIOS 1416

Anyone tested ?

 

[Nam-A-M-D]

Hi. Just noticed the update. I have just updated with no problems so everyone should be fine to do so.I noticed they changed the voltage to 1.250v which i feel could be lowered more. I set the Soc voltage to manual to 1.2v and also the CPU VDDIO/MC to 1.2v. Just doing some benchmarks now and noticed the CPU Core voltage behaving better doing multicore test on Cinebench R23 @ 1.208 when before it was all over the place. The single cores are boosting correctly between 5700 - 5775Mhz max voltage 1.474v. The CPU core boosting seems to be better overall. Asus has made some under the hood changes.
Give it a go for yourselves and let me know how you get on. Try the 1.2v on the soc voltage and CPU VDDIO/MC 1.2v to see if your system is stable. Remember i am using my 7950x. I will be testing the 7950X3D soon .cheers

ROG STRIX X670E-E GAMING WIFI | Gaming motherboards｜ROG - Republic of Gamers｜ROG Global

ASUS ROG Strix X670E-E Gaming WiFi features 18+2 power stages, DDR5 support, four M.2 slots with heatsinks, PCIe® 5.0, USB 3.2 Gen 2x2 front-panel connector, and onboard WiFi 6E. It also features AI Cooling II, AI Networking, Two-Way AI Noise Cancelation. ROG Strix X670E-E Gaming WiFi caters to...

rog.asus.com

Jayztwocents had the x670 crosshair board in his personal system, and he's had one die and one CPU die so far in the last two months.

At this stage no one knows if the lower end boards have the same problems with overvolting, EXPO profiles, and disabled safety features so... play it safe guys.

Hi Mussels.
I think all motherboards will be having updates to regulate the voltages correctly

 

[Mussels]

Hi. Just noticed the update. I have just updated with no problems so everyone should be fine to do so.I noticed they changed the voltage to 1.250v which i feel could be lowered more. I set the Soc voltage to manual to 1.2v and also the CPU VDDIO/MC to 1.2v. Just doing some benchmarks now and noticed the CPU Core voltage behaving better doing multicore test on Cinebench R23 @ 1.208 when before it was all over the place. The single cores are boosting correctly between 5700 - 5775Mhz max voltage 1.474v. The CPU core boosting seems to be better overall. Asus has made some under the hood changes.
Give it a go for yourselves and let me know how you get on. Try the 1.2v on the soc voltage and CPU VDDIO/MC 1.2v to see if your system is stable. Remember i am using my 7950x. I will be testing the 7950X3D soon .cheers

ROG STRIX X670E-E GAMING WIFI | Gaming motherboards｜ROG - Republic of Gamers｜ROG Global

ASUS ROG Strix X670E-E Gaming WiFi features 18+2 power stages, DDR5 support, four M.2 slots with heatsinks, PCIe® 5.0, USB 3.2 Gen 2x2 front-panel connector, and onboard WiFi 6E. It also features AI Cooling II, AI Networking, Two-Way AI Noise Cancelation. ROG Strix X670E-E Gaming WiFi caters to...

rog.asus.com

Hi Mussels.
I think all motherboards will be having updates to regulate the voltages correctly

SoC voltage does heat up the CPU more, so you're very likely running colder now - if you're wattage limited by PBO settings, more of that wattage is reaching the CPU Cores vs the SoC.


The problem is the BIOS udpates are at AMD's end, so they're putting a cap on VID - requested voltage.
If the boards lie with an offset (as many have been caught doing) or use LLC to compensate for loads, it can still go higher than expected even when manually set by users.

The problem with the ASUS board(s) is that enabling EXPO doesn't load SoC from AMD or the DRAM - it's upto the BIOS itself. It's a hidden value that could change every BIOS update or accidentally revert, EXPO I and EXPO II had nothing to do with the RAM's EXPO profiles, with EXPO II changing secondary timings and raising voltages higher - it's an overclocking preset, the kind we always advise people to never use but labelled as if its from the RAM itself.

Asus were both using a high SoC preset and a high level of LLC at least on the crosshair, and that's something software cant tell end users, only probing the boards.

 

[Nam-A-M-D]

SoC voltage does heat up the CPU more, so you're very likely running colder now - if you're wattage limited by PBO settings, more of that wattage is reaching the CPU Cores vs the SoC.


The problem is the BIOS udpates are at AMD's end, so they're putting a cap on VID - requested voltage.
If the boards lie with an offset (as many have been caught doing) or use LLC to compensate for loads, it can still go higher than expected even when manually set by users.

The problem with the ASUS board(s) is that enabling EXPO doesn't load SoC from AMD or the DRAM - it's upto the BIOS itself. It's a hidden value that could change every BIOS update or accidentally revert, EXPO I and EXPO II had nothing to do with the RAM's EXPO profiles, with EXPO II changing secondary timings and raising voltages higher - it's an overclocking preset, the kind we always advise people to never use but labelled as if its from the RAM itself.

Asus were both using a high SoC preset and a high level of LLC at least on the crosshair, and that's something software cant tell end users, only probing the boards.

Hi. The system seems to be the same at between 42-50 degrees running normal windows and up to 60 - 75 degrees under stress on a water loop the higher being benchmarking for a few hours. The System has run flawlessly today and passed all tests with no problems. I would say that lowering the voltage manually is better than the 1.3v set by Asus. All should be good to give it a go. cheers

 

[Mussels]

Hi. The system seems to be the same at between 42-50 degrees running normal windows and up to 60 - 75 degrees under stress on a water loop the higher being benchmarking for a few hours. The System has run flawlessly today and passed all tests with no problems. I would say that lowering the voltage manually is better than the 1.3v set by Asus. All should be good to give it a go. cheers

For sure.
The only remaining concern is that if asus EXPO sets 1.30v and spikes to 1.40v, if you set 1.20v what's it really spiking to?

Users only copy-paste advice from others, and if using the same voltage on asus results in higher than everyone else, asus users are going to end up with hotter temps and higher failure rates - didn't gigabyte and MSI get in trouble for similar to this with the "hardware deviation" being added to HWinfo to expose it?

 

[Ryodu]

In every Bios I have used so far I loaded the better timings for my Memory from this Video:

It sets the Voltage manually to 1.25v and even if I run Cinebench R23 it tops out at 1.305v which I think is absolutly OK.

 

[Nam-A-M-D]

For sure.
The only remaining concern is that if asus EXPO sets 1.30v and spikes to 1.40v, if you set 1.20v what's it really spiking to?

Users only copy-paste advice from others, and if using the same voltage on asus results in higher than everyone else, asus users are going to end up with hotter temps and higher failure rates - didn't gigabyte and MSI get in trouble for similar to this with the "hardware deviation" being added to HWinfo to expose it?

The Soc voltage is running at 1.243v under normal windows and 1.252v on Cinebench R23. This is keeping the voltage below the 1.3v Stable. This should work for all processors. The lower voltage the better and the scores in cinebench R23 have increased between 37580 - 38450v

 

[Ryodu]

what Vcore do you guys have under full Load? Mine goes up to 1.45v on Auto setting which I find quite high

 

[Nam-A-M-D]

what Vcore do you guys have under full Load? Mine goes up to 1.45v on Auto setting which I find quite high

mine goes up to 1.472v on my 7950X running @ 5725Mhz at 50 degrees. I leave mine on Auto because Manually setting this voltage causes more instability issues due to the fluctuating voltages on the cores running higher clocks

 

[Mussels]

The Soc voltage is running at 1.243v under normal windows and 1.252v on Cinebench R23. This is keeping the voltage below the 1.3v Stable. This should work for all processors. The lower voltage the better and the scores in cinebench R23 have increased between 37580 - 38450v

No argument there - *if* its really 1.252v that's a happy voltage.
The crosshair problem is that it would then have the hardware add another 0.05v on top, so your 1.25 became 1.30v (still safe, in your case)

Asus had the problem of hardware and bios adding in voltage offsets, and because of that no one really knows what the heck is going on - different BIOSes have different defaults, and the hardware offsets arent showing in the software readings.


TL;DR: Manually set the lowest that's stable for you. Make sure you give it long term testing at low load/idle as well, as thats when AM4 played up the most with low SoC.