News AMD's 'Ryzen Burnout' Fixes Reportedly Plagued With Bugs

[sitehostplus]

All I had done was enable EXPO. Didn’t touch CPU at all. Imagine losing a processor and board using a feature that AMD advertises. 😂

It is again enabled.

Timex advertises 'Takes a licking and keeps on ticking". Does that mean I can beat my Timex watch with a hammer?

You can go ahead and use EXPO, and I don't care. Just thought I'd point that out is all.

 

[Deleted member 2838871]

Timex advertises 'Takes a licking and keeps on ticking". Does that mean I can beat my Timex watch with a hammer?

You can go ahead and use EXPO, and I don't care. Just thought I'd point that out is all.

All good... not really the same thing but not interested in debating the comparison... and I was just pointing out that AMD advertises EXPO with their processors... and Intel does the same. To market a feature and then deny warranty claims over it because it is technically "overclocking" wouldn't hold up to lawsuits I don't think... but I'm not a lawyer.

At the end of the day I don't much care either. I'm not going to hide in fear and not use my PC out of fear of what might happen and/or has already happened. I rebuilt and ran with EXPO disabled for 5 days until a BIOS update was released... then turned it back on and have ran the last 10 days without issue.

Am I expecting further issues? No. Do I plan to OC my CPU? No.

Cheers.

 

[InvalidError]

To market a feature and then deny warranty claims over it because it is technically "overclocking" wouldn't hold up to lawsuits I don't think... but I'm not a lawyer.

While it probably wouldn't, chances are you would end up spending more on lawyers and court fees than eating the cost. Since AMD and Intel both tell you this is technically overclocking and may void your warranty, it is unlikely you will get awarded anything beyond replacement goods or a refund.

 

[Deleted member 2838871]

While it probably wouldn't, chances are you would end up spending more on lawyers and court fees than eating the cost. Since AMD and Intel both tell you this is technically overclocking and may void your warranty, it is unlikely you will get awarded anything beyond replacement goods or a refund.

Yep. I am glad my part failures happened on day 3 where I didn't have to deal with AMD or MSI on the issue... I just RMA'ed them to Newegg which was 100% hassle free.

If I have any further problems though I'm just gonna sell dead parts to Gamers Nexus. They offered to buy these at full retail price for their analysis videos but I had already returned them. 🤣 🤣 🤣

 

[bit_user]

While it probably wouldn't, chances are you would end up spending more on lawyers and court fees than eating the cost.

A class-action lawsuit is very plausible, if AMD denied these warranty replacements en masse.

Since AMD and Intel both tell you this is technically overclocking and may void your warranty, it is unlikely you will get awarded anything beyond replacement goods or a refund.

Maybe, but I think awarding of punitive damages is at the discretion of juries?

 

[Amdlova]

I got refund from amd... my 5700g now I will try sell the motherboard ram and other things to build a new intel.
After 35 days waiting...

 

[InvalidError]

Maybe, but I think awarding of punitive damages is at the discretion of juries?

You aren't going to get punitive damages unless you can prove there was ill will and neither company officially supports running anything beyond stock specs. Even Intel's overclocking insurance was only a one-shot deal.

 

[Deleted member 2838871]

neither company officially supports running anything beyond stock specs.

Yeah... neither support it... but both use it in marketing their products. 🤣

At the end of the day though I don't want anything free... I am more than happy to pay market price for quality items that work correctly... and don't care if it's AMD or Intel who makes it. I used Intel processors for the last 20 years and went AMD this time because I felt the timing was right with AM5, the power/thermal differences, etc... and my opinion hasn't changed any despite the 7000 series hiccups.

Just call me a beta tester. 🤣

 

[zx128k]

If this were Intel or nVidia there would be a storm of toxic posts. Remember the 12 pin power connector and nVidia? CPU should never get too hot that the STIM melts.

 

[TerryLaze]

The reddit user wasn't using PB or PBO. They just set a fixed frequency and voltage, mistakenly thinking they were setting maximum limits. Setting static values overrides the normal dynamic frequency/voltage behavior. Same thing happens with Intel CPUs, IIRC.

Manual overclocking of Ryzen 7K also overrides the default temperature target of 95 C, allowing the CPU to get up to the actual TJmax of 115 C before throttling/shutting down, which explains the high temps the reddit user was seeing.

In other words, everything behaved as expected, the 'issue' is purely user error.

I see your point, even though manual settings will be below what PBO sets as maximum it's still considered a different thing.
Correction:
I still see your point, but any change to default PB is overriding it so it is PBO, it's not THE default PBO but it's still PBO.
It's like setting the PL on intel to 300W and saying it's a different type of overclock from having PL unlimited.

Cements my point even more that people get confused with all the wordings and all the different limits. It would make sense for manual set limits to be maximum limits and to not mess with the max temp limit automatically.

 

[ddcservices]

You always have people who try to do things manually, and they end up making mistakes. Ryzen 9 7900X with ASRock X670e Taichi motherboard, 1.21 BIOS with AGESA 1.0.0.6, and the only settings I adjusted were to set the memory profile to EXPO with Aggressive for those setting, no other changes made. HWInfo 64 showing most cores going to 5.35GHz with my cooling being a Corsair H115i Pro AIO, so not even a 360mm version. I didn't even need to tell it to turn on PBO to get these speeds. One of the fans is even set to silent on the AIO because it needs to be replaced.

So, there might be issues with AGESA 1.0.0.7, or all of this may be a lot of noise about nothing. It may also be that Asus is STILL screwing up with CPU voltages in its BIOS code.

 

[ddcservices]

So I guess I'm confused with who exactly is at risk here. Is it only overclockers or even at stock settings? I know some articles places originally only suggested the 3D chips in the series, but now it sounds all 7000 CPUs.

AMD didn't put direct limits into AGESA to prevent motherboard makers from sending too much voltage to the CPU, and as a result, some processors and even some motherboards were damaged. So, with AGESA 1.0.0.6 and 1.0.0.7, AMD had to lock down the voltages because motherboard makers can't be trusted.

 

[TerryLaze]

lock down the voltages because motherboard makers can't be trusted.

But we know that because this happens for a decade or so already, and if we know that AMD knows it way much better.
If AMD knew that the CPU can't handle above a certain voltage then it would have been on them to make sure that it only gets that much voltage .

 

[zx128k]

But we know that because this happens for a decade or so already, and if we know that AMD knows it way much better.
If AMD knew that the CPU can't handle above a certain voltage then it would have been on them to make sure that it only gets that much voltage .

I think I remember a statement were thermal protection wasn't working as well. Leads to a runaway thermal situation and kills the cpu.

 

[TerryLaze]

I think I remember a statement were thermal protection wasn't working as well. Leads to a runaway thermal situation and kills the cpu.

Yes,but it's not "as well" ,it's supposed that the high Volts are causing the thermal protection to fail.
Then, after that, if the mobo doesn't have proper protections (OCP) it keeps running at super high voltages until it fries both the CPU and MOBO.
From your link:

It seems that using EXPO memory modules - AMD's answer to Intel's XMP - which apply correct memory frequencies, voltages and timings, was somehow allowing the SoC voltage to rise to unsafe levels. This in turn, according to Tom's hardware, could 'destroy the chips' thermal sensors and thermal protection mechanisms, completely disabling its only means of detecting and protecting itself from overheating'.

 

[blacknemesist]

Do you have a source indicating that using a beta BIOS voids your warranty? I have not seen that suggested before.

On Asus support it states, under a beta bios, the following :

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."

PRIME X670E-PRO WIFI｜Motherboards｜ASUS Global

ASUS PRIME X670E-PRO WIFI is an AMD X670E (Ryzen AM5) ATX motherboard with PCIe® 5.0, four M.2 slots, DDR5, USB 3.2 Gen 2x2 Type-C®, front USB 3.2 Gen 2 Type-C, Thunderbolt™ 4 support and 2.5 Gb Ethernet. It also features AI Cooling ll and Two-Way AI Noise Cancelation. PRIME X670E-PRO WIFI...

www.asus.com

 

[Deleted member 2838871]

On Asus support it states, under a beta bios, the following :

ASUS is not responsible for direct, special, incidental or consequential damages resulting from using this beta BIOS."

Yeah I never saw any kind of warning like this with my MSI board BIOS. If it's on the page somewhere I can't find it. I'm mentioning it because when mine blew up 3 days after the build I was on a new "optimized for 7000 series" beta BIOS.

I even took a little heat (no pun intended) here on the forums "I feel bad for you but not as much as I did" for using a beta. 🤣🤣

Within 24 hours though the news broke that it was a widespread issue.

After rebuilding I updated to a new "this is AMD's fix" BIOS on 4/28 that locked SOC voltage at 1.3v but it also shows as beta on my board page. The BIOS that I was running when my cpu and board failed was removed from the board page.

Again though... no CYA disclaimers from MSI.

PC has been fine for last 11 days with EXPO enabled.

 

[TJ Hooker]

I see your point, even though manual settings will be below what PBO sets as maximum it's still considered a different thing.
Correction:
I still see your point, but any change to default PB is overriding it so it is PBO, it's not THE default PBO but it's still PBO.
It's like setting the PL on intel to 300W and saying it's a different type of overclock from having PL unlimited.

Cements my point even more that people get confused with all the wordings and all the different limits. It would make sense for manual set limits to be maximum limits and to not mess with the max temp limit automatically.

Manually setting a fixed frequency and voltage is not PBO.

"Unlike a manual overclock which can lock cores at a particular frequency, with Precision Boost Overdrive SenseMI is left in control of the processor [...]" https://hothardware.com/news/amd-un...to-amp-up-ryzen-5000-performance-even-further


"PBO is an extension to AMD's normal Precision Boost that offers an alternative overclocking solution to manual overclocking [...]" https://en.wikichip.org/wiki/amd/pbo

The fact that the redditor did something unintended isn't necessarily a result of it being confusing, it's a result of them messing around in the BIOS without doing any research and not understanding what they're doing.

 

[blacknemesist]

Yeah I never saw any kind of warning like this with my MSI board BIOS. If it's on the page somewhere I can't find it. I'm mentioning it because when mine blew up 3 days after the build I was on a new "optimized for 7000 series" beta BIOS.

I even took a little heat (no pun intended) here on the forums "I feel bad for you but not as much as I did" for using a beta. 🤣🤣

Within 24 hours though the news broke that it was a widespread issue.

After rebuilding I updated to a new "this is AMD's fix" BIOS on 4/28 that locked SOC voltage at 1.3v but it also shows as beta on my board page. The BIOS that I was running when my cpu and board failed was removed from the board page.

Again though... no CYA disclaimers from MSI.

PC has been fine for last 11 days with EXPO enabled.

AMD and ASUS support are too messy to deal with so I avoid touching anything that can be a problem for them. That being said my 7700X was bought in november and the mobo was replaced 2 weeks ago and came with 1.0.0.6 which does nothing for me(SoC is over 1.3v with EXPO) so if there is so kind so CPU weardown I can not see any indication of it and might come to bite me later(or not at all). My thermal protection is also gone and can get over 95ºC so yeah, kind of want to return the CPU but have nothing to justify it.

 

[Deleted member 2838871]

AMD and ASUS support are too messy to deal with so I avoid touching anything that can be a problem for them. That being said my 7700X was bought in november and the mobo was replaced 2 weeks ago and came with 1.0.0.6 which does nothing for me(SoC is over 1.3v with EXPO) so if there is so kind so CPU weardown I can not see any indication of it and might come to bite me later(or not at all). My thermal protection is also gone and can get over 95ºC so yeah, kind of want to return the CPU but have nothing to justify it.

I've checked mine with HWInfo and it's showing locked at 1.3v with EXPO on.

As for thermals... the highest I've seen is 89C under a 30 min cpu burn test. Within 1 minute it hit 89C... then stayed there for 29 mins.

Noctua D15 air cooler.

 

[Phaaze88]

View: https://www.youtube.com/watch?v=fFNi3YNJXbY

 

[Deleted member 2838871]

Watching it now. 👍👍