9800X3D Failures/Deaths in the UK?

[jigger]

Agreed it's not just ASRock but I think we are in the realms of acceptable failure rates on other boards, especially now nodes are so small and we have all the high-κ dielectrics etc in silicon to mitigate all the issues at this size, I imagine intel, Qualcomm etc must have similar rates. Also a huge amount of X3D chips sold for obvious reasons.

Be nice to get an end to this from the ASrock side though as its a constant fear for me.

It’s hard to say what the issue is. Some people are pointing at Lotes/Foxconn sockets as the issue.

It’s seems the boards have been built with whatever components where available to ASRock at the time. If that meant the lowest priced components available at the time then ASRock was maybe asking for trouble.

 

[Rroff]

I doubt the sockets are the issue though there isn't enough information to be sure, as failures consistent with this issue are happening on other boards just at a far lower rate and there is a high correlation with failures and the CPU batch and/or selling date which likely indicates similar batches, though that doesn't entirely eliminate a fitting issue which is exacerbated on some boards.

 

[jigger]

Both manufacturers have had issues traced back to them in the past. LGA sockets are tricky to get right as there is very tolerance. It’s impossible to know what is going on. Decent reports seem few and far between and failures seem pretty random.

 

[welshrat]

I doubt the sockets are the issue though there isn't enough information to be sure, as failures consistent with this issue are happening on other boards just at a far lower rate and there is a high correlation with failures and the CPU batch and/or selling date which likely indicates similar batches, though that doesn't entirely eliminate a fitting issue which is exacerbated on some boards.

Yeah it is quite possibly a perfect storm of on the edge processors and some rather loose BIOS settings. One thing I found interesting in the comments in reddit is someone pointed out that one of the failures was at VDDIO but gamer nexus replied saying that the burn marks and explosions on the chips were all over the place and never just in once place that could help isolate the issue, so maybe it is a socket issue.

Honestly I have no idea but it would be nice to know. I am already wondering if all stays good my side whether I will upgrade the board when I inevitably upgrade to the single CCD 12 core X3D chip that is apparently coming in zen 6.

Hopefully a less troubled chipset will arrive after X870 with less of this stupid lane sharing which is the reason I went Asrock in the first instance.

 

[welshrat]

interestingly at 11:30 he mentions micro arcing which brings back node sizes and issues with something in the CCD possibly with electric current not correctly isolated. Again though who bloody knows

 

[jigger]

Yeah it is quite possibly a perfect storm of on the edge processors and some rather loose BIOS settings. One thing I found interesting in the comments in reddit is someone pointed out that one of the failures was at VDDIO but gamer nexus replied saying that the burn marks and explosions on the chips were all over the place and never just in once place that could help isolate the issue, so maybe it is a socket issue.

Honestly I have no idea but it would be nice to know. I am already wondering if all stays good my side whether I will upgrade the board when I inevitably upgrade to the single CCD 12 core X3D chip that is apparently coming in zen 6.

Hopefully a less troubled chipset will arrive after X870 with less of this stupid lane sharing which is the reason I went Asrock in the first instance.

If it makes you feel any better I’m still daily using the OG Asus ROG X670 murder board, with 7800X3D and release BIOS. I’m still willing it to spectacularly die, on the daily… The thing just keeps chugging away and sipping power.

 

[jigger]

interestingly at 11:30 he mentions micro arcing which brings back node sizes and issues with something in the CCD possibly with electric current not correctly isolated. Again though who bloody knows

What he says in regards to micro arching makes next to no sense. Once a short occurs current will flow until it’s stopped. Either by reaching ground or something like a trace/pin has been obliterated. If there was some kind of intermittent fault causing a short that can bridge an air gap it would either cause instability in signals and crash the system or increase until something fails.

The cause would be very obvious.

 

[welshrat]

What he says in regards to micro arching makes next to no sense. Once a short occurs current will flow until it’s stopped. Either by reaching ground or something like a trace/pin has been obliterated. If there was some kind of intermittent fault causing a short that can bridge an air gap it would either cause instability in signals and crash the system or increase until something fails.

The cause would be very obvious.

Maybe but following the Reddit thread a lot of people start having instability a few weeks prior. Honestly though I think we are all clutching at straws

 

[jigger]

Maybe but following the Reddit thread a lot of people start having instability a few weeks prior. Honestly though I think we are all clutching at straws

You would be able to easily measure and identify the cause of instability. The fault would be pretty obvious and repeatable. Voltage and current reading would be all over the place.

 

[deviation]

If it makes you feel any better I’m still daily using the OG Asus ROG X670 murder board, with 7800X3D and release BIOS. I’m still willing it to spectacularly die, on the daily… The thing just keeps chugging away and sipping power.

 

[G J]

interestingly at 11:30 he mentions micro arcing which brings back node sizes and issues with something in the CCD possibly with electric current not correctly isolated. Again though who bloody knows

At the start of the video he mentions that he was approached by Asrock to sign an NDA which raises the question that some must have signed this NDA which could explain the reason why a lot of the hardware space has been quiet.
Seems theres no problem what Asrock is doing but when it was Nvidia...