-
Hey, guest user. Hope you're enjoying NeoGAF! Have you considered registering for an account? Come join us and add your take to the daily discourse.
Multiple Reports of AMD Ryzen 7800X3D CPU V-Cache burning itself and the motherboardš„
- Thread starter Leonidas
- Start date
baphomet
Member
Thanks for the heads up. Newegg had no problems giving me a key.
Even though it apparently runs like shit, I'll still take it for free.
lmimmfn
Member
Lol, yeah, seems like a junk port but I'm sure it will be fixed eventually and free is free
RadioactiveLobster
Member
lmimmfn
Member
looks like Steve managed to end the life of an 7800x3d, video will be interesting.
GHG
Gold Member
Yeh just saw it on my feed. Curious to see what the exact cause is but it seems like Asus are almost definitely involved.
Last edited:
RadioactiveLobster
Member
OverHeat
Ā« generous god Ā»
Damn!!! Should have gone Intel this round
Buggy Loop
Member
Can't wait to see that video from GN.
Motherfucking tech jesus will unveil everything
Motherfucking tech jesus will unveil everything
Rentahamster
Rodent Whores
Incompetence on many levels. Stay tuned to find out from whom!
oldergamer
Member
Fuck I just bought a 7000 and a 590e mobo. Should I not use it? Or not use the bios?
oldergamer
Member
People need to stop downplaying. And already acknowledged there is an issue
baphomet
Member
Acknowledged what? That running too much power into something will fuck it up? No shit, you can screw up literally any CPU and motherboard by running well above the rated power directly into it.
The amount of idiots not knowing how to plug their 4090's in was orders of magnitude more than this.
So you bought a CPU and a motherboard that aren't compatible with each other?
RadioactiveLobster
Member
Buggy Loop
Member
When tech Jesus says Ā« AM5 is a mess Ā»
You know youāre in for something
Glad I made the last AM4 upgrade with the 5800x3d rather than upgrade the whole platform.
This is a fuckup from AMD and board vendors.
Agent_4Seven
Tears of Nintendo
Meanwhile at intel's HQ:
LiquidMetal14
hide your water-based mammals
Hopefully ASUS puts in fixes soon.
Luckily mine has been solid.
Luckily mine has been solid.
GHG
Gold Member
It appears to not just be an Asus problem unfortunately.
LaughingStock
Member
I ordered a 7800X3D yesterday. Now I'm worried.
winjer
Gold Member
Don't be. Just don't forget to update your BIOS.
Celcius
Ā°Temp. member
I was just about to return to AMD for the first time since Athlon 64.
Nope, AM5 looks like a mess at the moment. This exploding issue, people complaining about Ethernet issues with the 2.5g Intel Ethernet included, people complaining about coil whine from the asus x670e boards, long post times, and moreā¦
Iāll just keep what I have for now and I guess wait for pcie 5.0 drives from either WD or Samsung before I do a platform upgrade.
Nope, AM5 looks like a mess at the moment. This exploding issue, people complaining about Ethernet issues with the 2.5g Intel Ethernet included, people complaining about coil whine from the asus x670e boards, long post times, and moreā¦
Iāll just keep what I have for now and I guess wait for pcie 5.0 drives from either WD or Samsung before I do a platform upgrade.
rnlval
Member
What's 590e?
rnlval
Member
My ROG X670E Hero's Intel 2.5G is fine. My network's QNAP switch has a 10G backbone and QNAP NAS has Realtek 2.5G.
No coil whine for my TUF X670E Plus Wifi (Realtek 2.5G NIC) and ROG X670E Hero (Intel 2.5G NIC) motherboards.
Last edited:
KuraiShidosha
Member
My 7950x3D is very likely affected by this to some degree. I left it on EXPO II for a month with an Asus B650E-F and it's been acting up ever since. Nice. Glad I spent like $1400 on this setup.
The crazy part is AMD isn't even a value option anymore. Their pricing is insane at MSRP for the past two years, they are basically trying to Nvidia the CPU space, except unlike Nvidia their new products are mediocre at best and their competitor is refusing to budge on pricing, so Intel ends up being significantly better value. AM5 sold like total shit at launch and it got significant price cuts to remain remotely competitive.
Honestly right now with 12th and 13th gen Intel is killing it, there's kind of no reason to buy AMD aside from maybe the 7800x3D, and even then apparently it looks like those chips are extremely shit bins.
simpatico
Member
It's a sad thing this is what they're choosing to do with their newfound improving reputation. Textbook example of shooting oneself in the foot.
Chiggs
Gold Member
StereoVsn
Gold Member
3D cache chips seem to be a lot more sensitive to voltage variations.
Chiggs
Gold Member
Totally. It's just that after the Nvidia PSU adapter thing, I'm not falling for the hype. Even GamersNexus listed this on their YouTube video:
Also from Steve himself @ 34 minutes and 25 seconds.
Zzzzzzzzzz
Last edited:
LaughingStock
Member
Just to be safe I'm picking a MSI mobo. A B650 MSI MAG Tomahawk Wi-Fi. Am I going to be safe, GAF?
Chiggs
Gold Member
Yes.
I really think people need to watch the conclusion of that Gamers Nexus video. Here's LITERALLY what Steve says (in giant font):
Last edited:
rnlval
Member
https://community.intel.com/t5/Processors/XMP-Warranty-void/m-p/1196241Legit thinking about returning Asus x670e board and 7800x3D, and going Intel despite much higher power draws. I just don't want to deal with this shit.
As pointed out, Asus bios is beta and voids your warranty as just one example.
From Maria_R_Intel, Moderator
Yes, Intel* XMP is considering overclocking, altering clock frequency or voltage may damage or reduce the useful life of the processor and other system components and may reduce system stability and performance. Product warranties may not apply if the processor is operated beyond its specifications.
You can check this information in our warranty terms and conditions "any Product which has been modified or operated outside of Intelās publicly available specifications, including where clock frequencies or voltages have been altered, or where the original identification markings have been removed, altered or obliterated. Intel assumes no responsibility that the Product, including if used with altered clock frequencies or voltages, will be fit for any particular purpose and will not cause any damage or injury. "
StereoVsn
Gold Member
That's great and all, but I have yet to see reports of Intel XMP frying CPUs. Or at least at scale.
rnlval
Member
https://au.pcmag.com/motherboards/9...lled-this-z690-motherboard-over-a-burn-hazard
Asus alerted consumers about a burn hazard with the ROG Maximus Z690 Hero in December, but now the US Consumer Product Safety Commission has issued an official recall order.
rnlval
Member
EXPO II with manual settings that is similar to
The BIOS version is before Ryzen Zen 4 3D era.
Last edited:
StereoVsn
Gold Member
Maybe you should state your point vs posting random images. There multiple reports all over the place about issues with AMD AM5 platform right now.
rnlval
Member
My 821 BIOS version is before Ryzen Zen 4 3D era.
I have the same motherboard as you and running a 7800X3D. Previously was running bios version 1405 with zero problems, but updated to 1409 out of caution and holy shit this bios is awful. EXPO 1 profile isnāt stable anymore (constant memory management BSOD), SOC voltage is now defaulting to 1.40 volts (I canāt remember what it was before but this seems much higher/unsafe), takes over a minute for the computer to boot/restart. Also, I was running -35 all core offset using curve optimizer, I canāt even enable PBO or CO now without immediate BSOD when booting into windows. No idea how Asus could fuck this up so bad.
---
Zen 4 X3D CPUs require newer BIOS for their operation and some X3D BIOS releases have SOC voltages set too high.
From https://rog-forum.asus.com/t5/amd-6...ge-images-901992=69731i851B97444452BC1E#M1006
ASUS Strix X670E with 0088 BIOS has "CPU VDDCR_SOC" at 1.350V while my "CPU VDDCR_SOC" ASUS TUF X670E BIOS 0821 is at 1.240V.
Last edited:
StereoVsn
Gold Member
AMD and Asus are issuing BIOS updates to prevent the burn outs. GamersNexus verified the issue. There are many reports of problems actively happening. It's your choice to ignore that but most reasonable folks wouldn't.And, of course, if that was AMD,Leonidas would have made a megathread about it.
Just like with that Nvidia adapter, right? Watch out world: people on the internet are saying things!
And on 4090 adapters, it's a shit design coupled with substandard adapters. Different animal altogether.
marquimvfs
Member
Wow! I've entered here assuming that the problem was probably some overclocker fuckery, but, no. Resuming for the folks who aren't able to watch a video with almost an hour, here's the catch:
-3D chips are very fragile, at least, more than the non 3D. They're the ones being affected by the problem.
-Some board vendors (specially ASUS) fucked up on protection and voltage settings, what leaded to some of most spectacular failures we are able to see in tha hardware scene (explosions are always a show, i like them).
-AMD is also to blame, because some (or most, if not all) of the motherboard manufacturers fuckery could be prevented if they were more active in quality control, or, at least, in providing directives for platform manufacturing.
The good news is that this is a very uncommon problem, looks like it won't be a widespread failure, and there is a way to prevent it, and, also, if you can't implement the measures to mitigate the failure (bios updates), you could also monitor the functioning of your system and see if you are being affected by the problem. And, AMD is replacing the fried CPUs for free (even paying the shipping) even if you were using EXPO.
Point for you,
Leonidas
Edit: Engrish
-3D chips are very fragile, at least, more than the non 3D. They're the ones being affected by the problem.
-Some board vendors (specially ASUS) fucked up on protection and voltage settings, what leaded to some of most spectacular failures we are able to see in tha hardware scene (explosions are always a show, i like them).
-AMD is also to blame, because some (or most, if not all) of the motherboard manufacturers fuckery could be prevented if they were more active in quality control, or, at least, in providing directives for platform manufacturing.
The good news is that this is a very uncommon problem, looks like it won't be a widespread failure, and there is a way to prevent it, and, also, if you can't implement the measures to mitigate the failure (bios updates), you could also monitor the functioning of your system and see if you are being affected by the problem. And, AMD is replacing the fried CPUs for free (even paying the shipping) even if you were using EXPO.
Point for you,
Leonidas
Edit: Engrish
Last edited:
Chiggs
Gold Member
Oh, you referenced Steve from Gamers Nexus? How convenient for me since I quoted him earlier...and from the very video you're referencing. Enjoy:
And for anyone else needing grief crisis counseling:
Last edited:
rnlval
Member
ASUS has progressively increased SoC voltages during the X3D era and there's an overshoot problem i.e. set to 1.35V and it overshoots towards 1.41V. Setting 1.24V with an overshoot problem would remain in the 1.28V and low 1.3xV range. If I have Ryzen X3D chips, the SoC voltage must be low and must factor in the overshoot problem.Wow! I've entered here assuming that the problem was probably some overclocker fuckery, but, no. Resuming for the folks who aren't able to watch a video with almost an hour, here's the catch:
-3D chips are very fragile, at least, more than the non 3D. They're the ones being affected by the problem.
-Some board vendors (specially ASUS) fucked up on protection and voltage settings, what leaded to some of most spectacular failures we are able to see in tha hardware scene (explosions are always a show, i like them).
-AMD is also to blame, because some (or most, if not all) of the motherboard manufacturers fuckery could be prevented if they were more active in quality control, or, at least, in providing directives for platform manufacturing.
The good news is that this is a very uncommon problem, looks like it won't be a widespread failure, and there is a way to prevent it, and, also, if you can't implement the measures to mitigate the failure (bios updates), you could also monitor the functioning of your system and see if you are being affected by the problem. And, AMD is replacing the fried CPUs for free (even paying the shipping) even if you were using EXPO.
Point for you,
Edit: Engrish
Last edited:
marquimvfs
Member
Yes, but, setting automatically to 1.35V is itself a problem, especially with X3D chips... that's one of the mistakes that AMD itself could actively prevent.
Last edited:
rnlval
Member
This is mostly an ASUS problem. AMD's CPU replacement program is the "get out of jail card" for ASUS's mistakes.
marquimvfs
Member
Now we can only hope that they refuse to replace one of the faulty boards an end up with a pretty big lawsuit up their asses.