AMD Ryzen 7 7800X3D CPU Burns Up

rare · 17 May 2023 at 21:30

Linus did talk about this on his podcast video thingy. To be honest, but sure we really need any more coverage…

ASUS is in the Doghouse

Luke and Linus address the ASUS Beta BIOS Warranty disclaimer controversy. (Gamers Nexus)Watch the full WAN Show: https://www.youtube.com/watch?v=nXNIdz535I0...

youtu.be

Iron Man · 17 May 2023 at 21:58

I have not been following this but what’s the outcome? Is the Asus tax now not just money but a dead cpu?

Rawr · 17 May 2023 at 22:24

rare said:
Doesn’t seem that big a deal to me but I can see how it could be perceived.

You say everyone affected, but how many dead chips are there? Hardly any, it seems. Literally a handful globally.

If anything the amount being small makes their reaction look even more petty since it really shouldn't have cost them much, but maybe that's just me.

I dunno, I think Asus brought this on themselves with their reaction as I've said before. If they'd have come out ahead of it and said they'd fix it and cover anyone who did happen to unfortunately get burned I don't think it would have blown up at all. Overall it just feels like such an own goal.

Iron Man said:
I have not been following this but what’s the outcome? Is the Asus tax now not just money but a dead cpu?

I mean realistically it's mostly fixed now as far as any of us is aware so even on an Asus board you're probably unlikely to see it. Even before you were probably unlikely to see it as it feels like more of a silicon lottery fail than anything else.

That all said some of us aren't that pleased with how Asus reacted so they've taken a reputation hit either way.

rare · 17 May 2023 at 22:48

Iron Man said:
I have not been following this but what’s the outcome? Is the Asus tax now not just money but a dead cpu?

It’s fine, and it’s not just ASUS boards despite what some of the big YouTube channels have positioned it as. There is more to this than they have portrayed, especially considering the board manufacturers may well have been only following AMD’s own specification limitations.

Sc00p007 · 17 May 2023 at 22:52

Gossip. Small things being blown out of proportion by half truths and conclusions made without all the facts.

Dreadi · 17 May 2023 at 23:31

I posted this elsewhere but it's definitely not just Asus and it seems gigabyte are still pushing SoC beyond recommended levels in their newest, safe voltages, bios .

Higher EXPO and SoC voltages before and after the BIOS update on a Gigabyte Aorus Master X670E? | igor´sLAB

Recently, there has been and continues to be a lot of uproar about ASUS motherboards maintaining excessively high SoC voltages, which appear to be the problem with the Ryzen 7000X3D failures. However…

www.igorslab.de

Silent_Scone · 18 May 2023 at 04:48

Sc00p007 said:
Gossip. Small things being blown out of proportion by half truths and conclusions made without all the facts.

https://www.reddit.com/r/Amd/comments/13kdz5i/饼哥超有材2304subtitlevoltage_accuracy_which_vsoc_is/

[餅哥超有料#2304](字幕)Voltage Accuracy_which Vsoc is accurate?(電壓怎麼量比較準?)

#AMD_X3D #Vsoc*使用配備:ROG CROSSHAIR X670E HERO、BIOS:1415AMD 7950X3DGskill F5-6000J3038F16GX2、 DDR5-6000 EXPO*在正確的電壓量測點上量測電壓就會有一點點微小的誤差量(+/-V)，量測點離負載末端愈遠則誤差量愈大。...

youtu.be

SVI3 SOC voltage can also be seen from within Ryzen Master if not wanting to use HWInfo.

Sc00p007 · 18 May 2023 at 09:18

Silent_Scone said:
https://www.reddit.com/r/Amd/comments/13kdz5i/饼哥超有材2304subtitlevoltage_accuracy_which_vsoc_is/

[餅哥超有料#2304](字幕)Voltage Accuracy_which Vsoc is accurate?(電壓怎麼量比較準?)

#AMD_X3D #Vsoc*使用配備:ROG CROSSHAIR X670E HERO、BIOS:1415AMD 7950X3DGskill F5-6000J3038F16GX2、 DDR5-6000 EXPO*在正確的電壓量測點上量測電壓就會有一點點微小的誤差量(+/-V)，量測點離負載末端愈遠則誤差量愈大。...

youtu.be

SVI3 SOC voltage can also be seen from within Ryzen Master if not wanting to use HWInfo.

When i say blown out of proportion, I don't mean there isn't any truth in the rumours.
It's well known in the enthusiast scene about under reporting voltage etc which I don't agree with.
Out of proportion is when I'm reading posts all over the Internet of people scared to build their machines etc because they must think it's likely to blow up. Sort of like the nvidia 12 pin. Yes a small number of users had a melted connector, mainly due to not ensuring the connector is fully inserted. The Internet blew up with speculation of house fires and people calling for recalls etc. Again some truth, but blown out of proportion.

Dreadi · 18 May 2023 at 09:48

Sc00p007 said:
When i say blown out of proportion, I don't mean there isn't any truth in the rumours.
It's well known in the enthusiast scene about under reporting voltage etc which I don't agree with.
Out of proportion is when I'm reading posts all over the Internet of people scared to build their machines etc because they must think it's likely to blow up. Sort of like the nvidia 12 pin. Yes a small number of users had a melted connector, mainly due to not ensuring the connector is fully inserted. The Internet blew up with speculation of house fires and people calling for recalls etc. Again some truth, but blown out of proportion.

And people don't seem to realise its not just Asus! Wasn't the very first one on a gigabyte board?
Sure, Asus deserve the backlash with how the went about things, but it's all brands across more then one chipset and some are still pushing beyond recommended limits despite*fixing* the issue. It's really been blown out of proportion.
I built my system last week, updates bios to lastest non beta, turned on expo then manually tightened the ram up and set voltages and it's been solid and I've not been worried about it blowing up

xrob · 18 May 2023 at 09:51

Sc00p007 said:
When i say blown out of proportion, I don't mean there isn't any truth in the rumours.
It's well known in the enthusiast scene about under reporting voltage etc which I don't agree with.
Out of proportion is when I'm reading posts all over the Internet of people scared to build their machines etc because they must think it's likely to blow up. Sort of like the nvidia 12 pin. Yes a small number of users had a melted connector, mainly due to not ensuring the connector is fully inserted. The Internet blew up with speculation of house fires and people calling for recalls etc. Again some truth, but blown out of proportion.

definitely, its kind of cringe these tech tubers like GN and J2C cancelling ASUS, feels like they're just doing it for clicks, especially J2C

Silent_Scone · 18 May 2023 at 09:52

Sc00p007 said:
When i say blown out of proportion, I don't mean there isn't any truth in the rumours.
It's well known in the enthusiast scene about under reporting voltage etc which I don't agree with.
Out of proportion is when I'm reading posts all over the Internet of people scared to build their machines etc because they must think it's likely to blow up. Sort of like the nvidia 12 pin. Yes a small number of users had a melted connector, mainly due to not ensuring the connector is fully inserted. The Internet blew up with speculation of house fires and people calling for recalls etc. Again some truth, but blown out of proportion.

I was agreeing with you. The link is evidence of that and explains why all other readings besides SVI3 (AMD's own controller and telemetry on die) are the most accurate voltage sensor. All other sensors will read higher depending on where the SIO is tapped. The point is, the latest firmware works as intended and the voltages aren't being exceeded as some outlets are claiming.

str · 18 May 2023 at 09:54

An important aspect of this situation is that it is evident Asus customer support have on many occasions legally enforced the beta bios disclaimer as warranty void and the people affected by that dumb decision get stonewalled into reluctantly accepting it as a matter of fact.

Evidently Asus customer support lied to these customers, and here in the UK, it's a criminal offence to lie on behalf of a company. I don't think Asus staff should be prosecuted, it's more of a dumb overreach that has become the industry norm, and the responsible people higher up also have the dumb duty to not have a moral compass meaning they turn a blind eye until of course they have no other choice.

xrob · 18 May 2023 at 09:54

Dreadi said:
And people don't seem to realise its not just Asus! Wasn't the very first one on a gigabyte board?
Sure, Asus deserve the backlash with how the went about things, but it's all brands across more then one chipset and some are still pushing beyond recommended limits despite*fixing* the issue. It's really been blown out of proportion.
I built my system last week, updates bios to lastest non beta, turned on expo then manually tightened the ram up and set voltages and it's been solid and I've not been worried about it blowing up

yeh i built mine before all this hoo-ha, my MSI board did exactly the same thing with SOC voltages on EXPO

eddiew · 18 May 2023 at 10:20

Troezar said:
From what I've seen you'll be fine unless you are running all four DIMMS and large amounts of RAM.

2x32 G.Skill 6000 c30. I guess fingers crossed.

To be fair, I checked the new QVL and there is still one other set of G.Skills with those specs and a slightly different model number - which probably means they're the same thing with/without LEDs on top.

Rroff · 18 May 2023 at 10:30

str said:
An important aspect of this situation is that it is evident Asus customer support have on many occasions legally enforced the beta bios disclaimer as warranty void and the people affected by that dumb decision get stonewalled into reluctantly accepting it as a matter of fact.

Evidently Asus customer support lied to these customers, and here in the UK, it's a criminal offence to lie on behalf of a company. I don't think Asus staff should be prosecuted, it's more of a dumb overreach that has become the industry norm, and the responsible people higher up also have the dumb duty to not have a moral compass meaning they turn a blind eye until of course they have no other choice.

Funny thing is I got called crazy for making comments about this kind of stuff earlier in the thread :cry:

Some quite weird posts in these threads here and elsewhere when it comes to trying to deflect from Asus, I didn't realise they had such a fanboy following and/or using shills to that extent.

Dave2150 · 18 May 2023 at 10:32

Brizzles said:
Interesting point raised here, the board vendors were ' told ' that 1.4 SOC was fine by AGESA firmware ( look at the last line ).

So AMD were/are at fault for a lot of the problems

AMD allowing too high SOC via FM (high enough to kill CPU's) is why we've seen failures on Gigabyte, Asrock, MSI and ASUS (I.E., not just one vendor)

rare · 18 May 2023 at 10:34

xrob said:
definitely, its kind of cringe these tech tubers like GN and J2C cancelling ASUS, feels like they're just doing it for clicks, especially J2C

100%, Jays video was literally empty except for some ramblings about a couple of apex boards. I found the whole thing embarrassing. He seems to always jump on a bandwagon.

Rawr · 18 May 2023 at 10:34

I mean as mentioned time and time again on the previous pages no one even argued it was just Asus. They just got the main focus because they actively telling people they weren't covered until they walked it back. I don't think it's hard to see why people are annoyed and why they got heat for it when others didn't. It feels like this conversation is just going round in am endless circle, lol.

Dave2150 · 18 May 2023 at 10:37

rare said:
ASUS have unfairly taken a lot of the brunt of this IMO, despite them probably following AMD’s own specifications. Although I think GN did a good investigation, the problem with these kinds of things is they don’t have access to all the information and have probably made quite a few assumptions. Now half the internet is ‘ASUS=bad’.

All the while Gigabyte is standing in the corner trying not to move so no one can see it.

I think ASUS could have done a better job with the warranty text in the BIOS release, but again that’s copy and paste. I’ve seen it on beta BIOS releases since…forever. Glad they removed it though.

ASUS have priced out many customers that were used to buying them over the years. A Maximus Hero board used to cost £180-200 (paid around £200 for my Maximus Hero VIII). Now you need to pay £650 (on release) for the current gen Maximus Hero.

This has earned ASUS much hate and loathing, though it's worth pointing that many still buy at this price (as there are so many of them in the wild) as they still like the product and feel it's worth it.

At the end of the day, best everyone just buys whichever motherboard they prefer. In this situation, all the motherboard vendors killed CPU's due to AMD's SOC voltage being allowed to be set too high, which is an AMD failure.

Dave2150 · 18 May 2023 at 10:38

xrob said:
definitely, its kind of cringe these tech tubers like GN and J2C cancelling ASUS, feels like they're just doing it for clicks, especially J2C

Oh they 100% are - their revenue is down due to no big new hardware launches, so they must generate hype/clicks another way.