• Competitor rules

    Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.

AMD Ryzen 7 7800X3D CPU Burns Up

Soldato
OP
Joined
12 Feb 2014
Posts
2,864
Location
Somewhere Only We Know
Yeah let's see them along with info. Maybe my google is broken.

I equally struggled to find a scenario where the manufacturer and partners had to run around and mitigate the issue. Maybe you can help me find that also.
Yup, your google must be broken, either that you you just simply couldnt be bothered to look, came straight up for me:


If you click on all instead of images you'll get loads of threads about it.
 
Last edited:
Soldato
OP
Joined
12 Feb 2014
Posts
2,864
Location
Somewhere Only We Know
Not sure if its been mentioned, but 1.0.0.7 AGESA is full of bugs and anyone who has already flashed it they are recommending you to roll back to you previous bios, there will be another new update by the end of may with all the fixes in it and better memory support, it will either be 1.0.0.7a or 1.0.0.9.
 
Soldato
OP
Joined
12 Feb 2014
Posts
2,864
Location
Somewhere Only We Know
My machine will stay off. Just picked up a MAC so no real loss.

Still using mine, ive just manually set all the voltages excpet vCore, even set them in the AMD Overclocking menu which I normally dont touch, I didnt pay over £1000 for a machine to gather dust, ive got a hoover for that, just stick it in eco mode or reduce the wattage to 65watts, my 7900 performs amazing for a 65watt part, if you cooling is sufficient then there shouldnt be a problem.

heres the link:

 
Last edited:
Soldato
OP
Joined
12 Feb 2014
Posts
2,864
Location
Somewhere Only We Know
I have not noticed any bugs and the CPU-SOC voltage has always been between 1.22 - 1.28. I don’t overclock other than clicking EXPO so will not notice most issues, just disable [wifi, blutooth, iGPU, ASUS rootkit installer] and set ECO 65.
Nope me neither, ive always got HWInfo open in the background at the moment, ive not seen SOC move off what ive manually set it to no matter what im doing, im not using EXPo though, ive manually set everything to do with RAM (Speed, Timings and Voltages)

4j8ql1L.jpg
 
Last edited:
Soldato
OP
Joined
12 Feb 2014
Posts
2,864
Location
Somewhere Only We Know
Its probably on auto, I have not changed it.

Unless you hard set it in both the AMD Overclocking menu and you board advanced voltages menu then its going to bounce around like that, but shouldnt go over 1.3v, its the thermal triggers that haven't been implemented yet, so what was happening was too much SOC was killing peoples CPU but even after they died the board continued to whack a tonne of current into the CPU's cooking them to +200oC, but as they were already dead, it didnt have anywhere to go, so the chip swelled up and burned the socket, the thermal triggers either failed or were set too high, in simple terms.
 
Last edited:
Soldato
OP
Joined
12 Feb 2014
Posts
2,864
Location
Somewhere Only We Know
Thanks for that info, i have not done that and just checked HW Monitor and its gone from 1.35 to the 1.25 any way. But i will certainly go into that and do it any way
I dont normally mess with that menu myself, but Buildzoid did a recent video where he was testing the SOC voltage in different menus with a multimeter attached, granted it was on a Gigabyte board, but as it was with the new AGESA, im going to guess its the same with all of them at the moment, and the only way he could get it to stick was setting it in both menu's, others like me think one setting should override the other, but for some reason, thats not working properly either.

 
Last edited:
Soldato
OP
Joined
12 Feb 2014
Posts
2,864
Location
Somewhere Only We Know
I've no clue if this is Soc voltage related but since setting it to 1.2v on my new build ive encountered 3 bsod (memory management error) in a week.
So ive increased my RAM voltage to 1.35v and Soc to 1.25v and we'll see how we get on.

Windows memory diagnostic is reporting no errors.
Are you using memory content restore - enabled in your bios ? if you are you also have to set power down enabled to enabled, if I disable both, I get a mega slow post, if I enable MCR but disable Power down then I get BSOD's, crashes etc.
 
Last edited:
Soldato
OP
Joined
12 Feb 2014
Posts
2,864
Location
Somewhere Only We Know
ASUS now have a new beta bios 1412 which contains the AGESA 1.0.0.7 Patch A from AMD which is supposed to fix all the bugs found in AGESA 1.0.0.7

 
Last edited:
Soldato
OP
Joined
12 Feb 2014
Posts
2,864
Location
Somewhere Only We Know
GN said that Asus in the biggest culprit , wait for part 3 and you`ll understand. Those `multiple vendors` are single or low number cases and require stupid steps and disabling protections to replicate; with Asus boards they just go bang out of the box, because asus doing what asus always has done. Asus boards overvolting isnt a new thing.
Ummm no, Buildzoid did a video with a Gigabyte motherboard not sticking to set voltages, and GN's first video which also contained a Gigabyte motherboard show'd the board (no matter what you set the voltage too) spiking 1.45v SOC into the CPU at random, its at 1min 8 seconds of the video, happens right there in the beginning on the GN video, so you dont even have to watch the whole thing, so its far from ASUS alone, no vid's have shown how ASRock and MSI boards behave yet, but one of those destroyed chips came from an ASRock board.

3iqwxkd.jpg
 
Last edited:
Soldato
OP
Joined
12 Feb 2014
Posts
2,864
Location
Somewhere Only We Know
Yeah that's too high ^^^^ but its not 1.45v.


The X3D are more sensitive to over-voltage as the added L3 cache is stacked on top of the CPU die, power and signalling is through microscopic ball sockets that tunnel through the substrate of both the cache and CPU.
That's the difference between the X3D and none X3D chips, the only reason we know about this is because it was blowing those X3D chips, that doesn't mean the none X3D chips aren't being killed more slowly

There is another reason this was so dramatic, the motherboards over current protection failed to detect the problem and trip, so it kept feeding it power after the CPU had shorted, we all know what happens when a circuit in your house that shorts fails to trip the fuse, it heats up and melts, in your fuse box look for the main flick switch, there is a red or yellow button under it, that is the short circuit protection test button, press it.

Its a combination of the motherboard malfunctioning and having the SoC voltage set WAY too high.

6M2zmFW.jpg
Lol, that was a gigabyte motherboard in the picture I posted, which was a screenshot from the first GN video, so again, it's really not just Asus
 
Last edited:
Soldato
OP
Joined
12 Feb 2014
Posts
2,864
Location
Somewhere Only We Know
Something's changed then as they have the most expensive B650E listed on overclockers and the most expensive B650 too, by a margin in both cases. Their X670E boards start at just £500...


Even the Gigabyte and ASRock issues?


Yeah, I'm having the exact same issue. I've got a Asus Strix X670E-F and can't find anything comparable in that sort of price range.
Ive been using an ASUS X670e Gene for months, and I love it, never had a problem with it, its the only X670E board that I liked that came fully loaded with PCI-e 5 and USB 4.
 
Soldato
OP
Joined
12 Feb 2014
Posts
2,864
Location
Somewhere Only We Know
Ive noticed ASUS have actually removed the disclaimer from all the beta bioses on their site now, so maybe they are starting to listen, funny though, the latest beta (1412) still isnt there, its not on the UK site either, possibly trying to get this one right before releasing the finished product.

4ZrcER7.jpg
 
Last edited:
Soldato
OP
Joined
12 Feb 2014
Posts
2,864
Location
Somewhere Only We Know
Maybe the text but I bet they'll still try and use it as an excuse to reject RMAs, etc.
In the UK they dont handle the RMA's themselves, I contacted them about another issue I had with my previous board that had, after filling out the form stating it had a beta bios on it, I spoke with the UK RMA centre who stated they dont handle components, only things like Laptops and monitors, and told me to send it back to where I purchased it from for a replacement, so I did, and got it replaced, never had a problem with them, it then goes back to the supplier / distributer who then return it to ASUS.

Had they of argued it, id of just flashed a non beta onto it and put the request in again.
 
Last edited:
Soldato
OP
Joined
12 Feb 2014
Posts
2,864
Location
Somewhere Only We Know
Unless it has changed recently they use LetMeRepair or something like that in the UK since around 2007 or so.

And for anyone doubting what I was saying above look up LetMeRepair on any consumer review site :s

EDIT: Not sure if LetMeRepair handle all Asus RMAs - I can't remember now which company handled (initially) the Asus "8800GT" (Actually an 8800GS which had somehow been mounted on an 8800GT board and packed as an 8800GT - man that was a mess trying to make them understand) I tried to RMA but I don't think that was LetMeRepair but fortunately Overclockers eventually stepped in and replaced the GPU with a brand new one from another brand.
never heard of them, ASUS used to have offices in Milton Keynes, literally just around the corner from the Gigabyte RMA place too, (ive dropped faulty products into both of them in the past), they also have a place in Hemel Hempstead who dont deal with components, they shut the place in Milton Keynes down (who did deal with components) so now you have to send it back if its faulty to where you brought it from, whether that be Overclockers or some other retailer.
 
Soldato
OP
Joined
12 Feb 2014
Posts
2,864
Location
Somewhere Only We Know
Really thats amazing, i assume you have turned somthing on to enable this perhaps ?
My PC boots to the windows lockscreen in 6 seconds, takes 3 seconds for me to type in my password, then startup programs take about 20 seconds, thats using a WD 850X nvme drive, with an ASUS x670e gene, 7900 non x and 32gb of RAM sort of expo enabled, but timings, speed, voltage etc entered manually, ive tried removing the 10 second windows explorer delay, but it doesnt seem to work on windows 11
 
Last edited:
Soldato
OP
Joined
12 Feb 2014
Posts
2,864
Location
Somewhere Only We Know
As fast as the hardware gets we never seem to make much ground on start-up. To be fair though I remember being able to go and make a cup of tea while waiting for the PC to boot :cry: I think we'll need some sort of persistent memory before we see it get much faster.
TBH, I remember being able to go make a whole roast dinner whilst waiting for a game to load on my first ever PC lol......Sinclair ZX Spektrum.......only to come back and find a syntax error at the end.......arrrrgggghhhh, but hey, the colours were nice to watch and hypnotising :cry:

For those who dont know what im talking about
 
Last edited:
Soldato
OP
Joined
12 Feb 2014
Posts
2,864
Location
Somewhere Only We Know
Lol, I still get PTSD if I hear that Spectrum loading sound. Waiting forever for it to load up only for it to error :mad: Used to infuriate me playing with volume settings :cry: Still got it up the loft somewhere.
You cant beat a bit of Horace goes skiing lol, they dont makes games like that anymore......(mumbles under my breath) "Thank god for that"
 
Back
Top Bottom