Question Do all AM5 risk frying a 7800x3D or is it only ASUS and gigabyte?

[wolter.jeremy]

Curious if anyone has heard of a msi or ASRock board frying a 7 series Ryzen or if it's only been mostly Asus and a rare gigabyte

 

[Roland Of Gilead]

Hey there,

IIRC it was all mobo's with this particular AGESA bios. But as long as your bios is up to date, and you have the latest one, then I think the fix is already done.

It's mainly to do with the the SOC voltage being above 1.3v.

 

[drea.drechsler]

I think it's just a concern if you enable XPO overclocking. So don't do that and you should be OK until a fix is developed and available. It will most likely be a BIOS update but could be chipset drivers too so keep current with those on your motherboard support site.

From what I gather the biggest beef with Asus is the weasling way they're shirking responsibility by reserving the right to retroactively void warranty if you used a BETA BIOS. Even one they put up to "fix" this problem and subsequently take down should they find it doesn't or injects other problems. Voiding warranty for overclocking (which enabling EXPO does) is one thing, but for just using a BIOS they furnished (even a BETA) is another. Personally, that just adds to the reason to never buy them again; I'm growing very disatisfied with my B550m Tuf Gaming.

 

[Roland Of Gilead]

I think it's just a concern if you enable XPO overclocking. So don't do that and you should be OK until a fix is developed and available. It will most likely be a BIOS update but could be chipset drivers too so keep current with those on your motherboard support site.

I think the XPO issue arrived in the AGESA fix. Fixes one thing and then breaks another

 

[wolter.jeremy]

I think it's just a concern if you enable XPO overclocking. So don't do that and you should be OK until a fix is developed and available. It will most likely be a BIOS update but could be chipset drivers too so keep current with those on your motherboard support site.

From what I gather the biggest beef with Asus is the weasling way they're shirking responsibility by reserving the right to retroactively void warranty if you used a BETA BIOS. Even one they put up to "fix" this problem and subsequently take down should they find it doesn't or injects other problems. Voiding warranty for overclocking (which enabling EXPO does) is one thing, but for just using a BIOS they furnished (even a BETA) is another. Personally, that just adds to the reason to never buy them again; I'm growing very disatisfied with my B550m Tuf Gaming.

My understanding is that XPO is like the new XMP? Will you be able to get everything out of your RAM speeds with this disabled?

 

[Roland Of Gilead]

My understanding is that XPO is like the new XMP? Will you be able to get everything out of your RAM speeds with this disabled?

Yes, it's possible to run ram with rated speeds with manual timings and voltage. This is what XMP/XPO do automatically. With that said, it can involve tinkering around with , as opposed to 'working out of the box'.

 

[TerryLaze]

Yes, it's possible to run ram with rated speeds with manual timings and voltage.

The question is if that voltage is going to be safe, if you have to manually put in more than 1.3V to make them run just as fast then that's not helping.

 

[drea.drechsler]

My understanding is that XPO is like the new XMP? Will you be able to get everything out of your RAM speeds with this disabled?

Its like XMP but with optimizations specific to AMD's processors. So I'm not sure you can get the full rated speed of the kit without using it. I'm also not sure if you can use XMP instead or if it has the same potential to overvolt your processor if it can be enabled.

It seems to me most of the tech press are more interested in pounding on the problem... and the parties responsible... than helping users to mitigate potential for damage. I'd like to know if you can use EXPO but manually set safe voltages and what would be the best way to see if your kit and motherboard have this problem when you do enable EXPO.

I've also gathered that even when it does overvolt it's not so serious unless also pushing the processor with heavy overclocking. I'm left to guess it means, if you set EXPO, you won't likely do any damage while you check it out to see if yours is experiencing the problem. But I wouldn't want to keep it running that way for a long time.

But even if you can't use it and EXPO is the only way to overclock your RAM without delving into manual settings, this will be a temporary limitation. AMD's working on a complete solution and we'll see something soon enough. But it will take time: AMD's going to test their solution, then it's handed off to the mfr's for them to integrate into BIOS's specific to their boards and they will test it too. More thoroughly this time, I think we all hope.

 

[Deleted member 2838871]

Curious if anyone has heard of a msi or ASRock board frying a 7 series Ryzen or if it's only been mostly Asus and a rare gigabyte

It's been Asus with the rare Gigabyte... and rare MSI.

I know because my MSI Mag x670e Tomahawk got smoked along with my 7950x3D back on April 22nd... 3 days after I built PC.

Rebuilt on April 23rd... updated to new BIOS on April 28th and all has been good. EXPO enabled.

RMA'ed dead parts... didn't lose anything except a Saturday afternoon of gaming.

The question is if that voltage is going to be safe, if you have to manually put in more than 1.3V to make them run just as fast then that's not helping.

No issues here. SoC is locked at 1.3v and I'm getting the advertised EXPO speeds. Too much performance left on the table not to run EXPO and I'm not gonna run my PC scared of what might happen.

 

[drea.drechsler]

It's been Asus with the rare Gigabyte... and rare MSI.

I know because my MSI Mag x670e Tomahawk got smoked along with my 7950x3D back on April 22nd... 3 days after I built PC.

Rebuilt on April 23rd... updated to new BIOS on April 28th and all has been good. EXPO enabled.

RMA'ed dead parts... didn't lose anything except a Saturday afternoon of gaming.



No issues here. SoC is locked at 1.3v and I'm getting the advertised EXPO speeds. Too much performance left on the table not to run EXPO and I'm not gonna run my PC scared of what might happen.

So to understand...your replacement board has an updated BIOS that locks in VSoC? or did you manually locked VSoC at 1.3V too?

If it's the updated BIOS alone that fixes it that's great, but if it's also manually setting VSoC that's important to know so that everyone can be sure to do that too.

 

[Deleted member 2838871]

So to understand...your replacement board has an updated BIOS that locks in VSoC? or did you manually locked VSoC at 1.3V too?

If it's the updated BIOS alone that fixes it that's great, but if it's also manually setting VSoC that's important to know so that everyone can be sure to do that too.

I updated to a new BIOS on 4/28 that locked the voltage at 1.3. I didn't do anything manually. The BIOS dated 4/14 that I was running when mine died was removed from MSI's website.

 

[drea.drechsler]

I updated to a new BIOS on 4/28 that locked the voltage at 1.3. I didn't do anything manually. The BIOS dated 4/14 that I was running when mine died was removed from MSI's website.

I don't know why it's so hard for the tech press to tell that story too so that people can know.

 

[Deleted member 2838871]

I don't know why it's so hard for the tech press to tell that story too so that people can know.

Yeah I'm not even salty about it. This is my first AMD build since the Athlon XP 1800+ in 2001. Been with Intel for 20+ years and went AM5 for the easy upgrade path vs Intel's LGA 1700 that is EOL.

The x3D chips much lower power draw and thermals was another factor in my decision.

I'm not disappointed at all in the choice. I mean yes it definitely sucked that the system blew up on me but the RMA process was easy and all I lost was a Saturday afternoon of gaming. After it blew up that morning I ordered a new board and cpu for next day Amazon delivery. I rebuilt the next day... then sent the dead parts back to Newegg on Monday... and had new parts the following Monday that I returned to Amazon for my $1000 back that had been in limbo for a week.

Only way to avoid being without a PC for a week... and Amazon is good with returns so that's what I did. Gamers Nexus offered to buy the dead parts for analysis when their video came out last week but I had already returned them to Newegg.

I've always used MSI boards and this is first time I've ever had an issue. As said though... all is good now. PC has been fine since the rebuild and EXPO has been on since the new BIOS on 4/28 and have had no further problems.

The PC is an absolute monster for 4K gaming.

 

[drea.drechsler]

...
I've always used MSI boards and this is first time I've ever had an issue. As said though... all is good now. PC has been fine since the rebuild and EXPO has been on since the new BIOS on 4/28 and have had no further problems.
...

I've had ASUS boards before, but always one of their low end boards. So when I got this one, a much more capable one (B550m Tuf Gaming Plus) I figured it would be just as good as the VRM design with 8 DrMOS power stages for the Vcore. I was shocked when I found out it didn't expose the sensors for monitor VRM temperature nor DIMM voltage.

Then I found out there's also a back door enabler in their BIOS: it instructs the OS to download and install their Aura command center software without asking me. It can be disabled but you have to know to do that before it trash-wares up your nice fresh Windows install.

Now, I found that Realtek has updated network and audio drivers but I can't update to them because Asus has a proprietary implementation. And they aren't updating their drivers on their support web site either, I assume since it's an obsolete board. That just sucks...but luckily there's this dude on ElevenForums.com that creates an installer that will install the latest WHQL signed Realtek drivers. I shouldn't have to go to that sort of extreme just to keep my system updated.

Their behavior with this VSoC issue is the absolute end. I'm done with Asus. I had one MSI board and was pleased as peaches with it. Even though they were a bit slow getting a BIOS update out to fix the 3700X boosting issues they did get it out...and it worked. I was very pleased with it and have thought about taking it back from my son to use for my 5800X system and give him the TUF. That means giving up PCIe gen 4, but oh well.

 

[Deleted member 2838871]

Well I have never purchased anything Asus... and now most likely never will.

 

[TerryLaze]

I've also gathered that even when it does overvolt it's not so serious unless also pushing the processor with heavy overclocking. I'm left to guess it means, if you set EXPO, you won't likely do any damage while you check it out to see if yours is experiencing the problem. But I wouldn't want to keep it running that way for a long time.

As far as I understood, the increased voltage from XMP messes up the thermal protection of the CPU, that causes normal PB to not have a thermal ceiling anymore because the thermal protection doesn't get triggered anymore and that is causing the heavy overclock cenario.
It's not manually doing heavy overclocking, it's normal PB which is just turbo boosting to infinity because the CPU doesn't know how hot it is getting.

 

[drea.drechsler]

It's not manually doing heavy overclocking, it's normal PB which is just turbo boosting to infinity because the CPU doesn't know how hot it is getting.

So it seems a bit of a two-faced dragon. The over volting is bad enough as that alone will present a long-term problem typical of running with excessive voltage. But it's thermal meltdown from boosting to infinity that can kill it quickly.

So the next question is: do you have to also enable PBO (I would not consider that heavy overclocking, btw) to set up the scenario for it to "boost to infinity" instead of observing normal thermal limits? Or does does it boost uncontrolled even with PBO in AUTO, or disabled?

 

[TerryLaze]

So it seems a bit of a two-faced dragon. The over volting is bad enough as that alone will present a long-term problem typical of running with excessive voltage. But it's thermal meltdown from boosting to infinity that can kill it quickly.

So the next question is: do you have to also enable PBO (I would not consider that heavy overclocking, btw) to set up the scenario for it to "boost to infinity" instead of observing normal thermal limits? Or does does it boost uncontrolled even with PBO in AUTO, or disabled?

As far as I know the only difference between the two are the limits of V W A, temperature is a global limit that is always in place unless you manually change it or if it fails.

 

[drea.drechsler]

... temperature is a global limit that is always in place unless you manually change it or if it fails.

It sounds like that's what is happening when it's overvolting...the global limit either fails, or simply ignored at least. My question is whether or not it has to also have PBO enabled to cause the global thermal limit to fail. It's not really important, just something I'm curious about.

Apparently the best way for someone with an affected board and CPU to avoid catastrophe is to simply not enable EXPO. But if someone wanted to see if they're affected what would be a way to check it out without taking a chance of near-instant death? Could they enable EXPO with PBO disabled and check VSoC voltage? or is there still a chance the CPU could boost to infinity, killing itself and the board?

 

[Deleted member 2838871]

Apparently the best way for someone with an affected board and CPU to avoid catastrophe is to simply not enable EXPO.

I'm not doing that. I rebuilt my system and updated to a new BIOS which locked SoC at 1.3v. EXPO is enabled and I've had no further issues for the last 17 days now. I'm not expecting any more issues either.

Some people may be the type to run and hide afraid to use their PC to its fullest potential because they are scared of what might happen due to what already has happened... but that ain't me.

AMD uses EXPO in their marketing campaigns... LOL... imagine that. There's too much performance left on the table with these AMD processors not to use EXPO. If you are gonna go that route might as well rebuild with Intel parts.

 

[drea.drechsler]

I'm not doing that. I rebuilt my system and updated to a new BIOS which locked SoC at 1.3v. EXPO is enabled and I've had no further issues for the last 17 days now. I'm not expecting any more issues either.

Some people may be the type to run and hide afraid to use their PC to its fullest potential because they are scared of what might happen due to what already has happened... but that ain't me.

AMD uses EXPO in their marketing campaigns... LOL... imagine that. There's too much performance left on the table with these AMD processors not to use EXPO. If you are gonna go that route might as well rebuild with Intel parts.

I do agree...use Expo to get the most out of it. But if you want to avoid drama, at least, then do it only if you know you have updated the board with a fixed BIOS. You seem to be good in that regard as MSI seems to have fixed your BIOS for you.