Soldato
Can't find an MSI board I like.
-
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
- Thread starter Jamin280672
- Start date
If you have all the parts, build it. I order everything at the same time and build as soon as it lands. If it dies, send it back. Just keep your old PC as a backup. Most people don’t have issues and don’t post, the ones with issues post everywhere so it looks like everything is exploding.
Associate
That could simply be a reflection of sales numbers. Asus sell the most, so of course you'd see more occurrences on their boards.
Looking for anything in particular? I had my eyes on this but it wasn't available at the time. https://uk.msi.com/Motherboard/MAG-X670E-TOMAHAWK-WIFI
Caporegime
Oh yeah... he has a video coming on that.
Caporegime
@Silent_Scone Its really quite simple, its two fold.
Asus was pushing far too much voltage in to the SoC and then failed to trip the over current protection when the CPU started shorting from excessive voltage. That resulted in a runaway thermal overload which eventually melted the CPU.
Asus was pushing far too much voltage in to the SoC and then failed to trip the over current protection when the CPU started shorting from excessive voltage. That resulted in a runaway thermal overload which eventually melted the CPU.
Last edited:
Could be or someone messed up thinking it be fine as it was fine on the 5800x3d.
anyhow, have asus tuf gaming b650 since day 1 from release, been fine with my 7600x.
ordered a 7800x3d today for the BIG upgrade I wanted for a long time as I skip the 5800x3d and went for am5.
Recently went for 32gb and use buildzoids timings and its been a swiss clock experience as far.
Be fun with the 7800x3d
Caporegime
The fact that Asus thought it was ok to slam the the SoC with 1.45v is quite scary, don't you think? They can't even get the basics right.
The default is 1.05v. With the 5800X from what i understood the maximum was 1.2v and preferably less than that. i have it at 1.17v with 3200MT/s XMP overclocked to 3800MT/s.
You ask any, at least; Ryzen 5000 owner who has spent some time researching overlcocking their RAM that ^^^^^ is what they will tell you.
The default is 1.05v. With the 5800X from what i understood the maximum was 1.2v and preferably less than that. i have it at 1.17v with 3200MT/s XMP overclocked to 3800MT/s.
You ask any, at least; Ryzen 5000 owner who has spent some time researching overlcocking their RAM that ^^^^^ is what they will tell you.
Last edited:
Well they messed up with the x3d cpus for whatever reason.
The 1.45v if accurate likely is a human error mistake along the lines
Gigabytes F10d bios for the X670E Aorus Master is still 1.3v with EXPO, I will be manually setting mine to 1.2v when I build mine.
Caporegime
Yeah that's too high ^^^^ but its not 1.45v.
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.
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.
Last edited:
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 AsusYeah 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.
Last edited:
Associate
I'm looking at the same board, but with a 7700. Did you get C30 RAM, and did it hit that rating fine ?
Caporegime
Yep. I think Asus should get ire for their SoC reporting being way off what it was actually supplying and their OCP being bust. Gigabyte should get ire because it's bios seems to have a mind of its own with the SoC as proven by that video and buildzoids testing.Then Gigtabyte should get the same level of criticism.
I don't really understand the defence of Asus about them being more popular causing more reports really. While that might be a small factor I don't think that suddenly absolves them of the fact there is evidence that they were one of the worst for overvolting the SoC when GN, Amd and even the board partners themselves have said high SoC was likely the cause.
I think both issues are quite worrying really and I don't think we need to defend either manufacturer.
Soldato
No disrespect, but nothing in that statement is objectively correct. In fact, it shows why the way the GN saga was documented has some large holes due to the focal points.
1) Other vendors have experienced failures with less voltage. Whilst we can target ASUS for excessive voltage, it doesn't explain all cases. In one case, the CPU was post-flash and not even applying a memory overclock or even on an ASUS board, so targeting ASUS alone doesn't make sense. What can be easily ascertained is that the CPU dies with far less than 1.45v applied to VSOC. Moreover, if VSOC is being fed for I/O the transceivers see the voltage. This was covered briefly by Wendell in the video. My belief is that failure can occur at much less than this, and it has, as per the case mentioned above.
2) The video doesn't document how the failure occurred, only that it did. The cooler appears to have been removed which would expose PROCHOT not operating correctly, and at this point, the CPU is already dead. The subsequent OCP discovery is accelerated by GN removing both the VRM cooling array and CPU cooler.
3) Excessive live current, not voltage.
4) OCP is there to protect the VRM, not the CPU. If you intentionally boot up a dead CPU, it's still dead. The fact the trip point is set high is another story in itself and can only be drawn up to an oversight. However, if assessing power demands on a CPU-by-CPU basis, you can't look for a CPU register if the chip is dead. Powering it on after it has died and removimng all cooling solutions (something that nobody sitting at home would do) only conflate a different issue, including the fact that AMD's PROCHOT does not appear to be working correctly in all cases.
There's a lot to digest, so it's not really that surprising folks are getting confused.
Last edited:
Soldato
I looked at that, but I don't like it when too many of the m.2 bays are where the GPU will be thus restricting airflow. Would also be nice if the last slot had a m.2 heatsink on it. Also, isn't it a bit low on VRM phases?
The Caron is just so expensive.
Caporegime
Are you just plucking acronyms out of the air as something to pin it on?
WTF does Throttle stop (PROCHOT) have to do with this? The CPU's cores are not overheating, they do not need throttling, its not what causes the SoC to short. You're just making #### up now.
Last edited:
It's impossible to burn a CPU. Did you put the thermal paste correctly? Did you overclock your CPU with wrong voltage?AMD Ryzen 7 7800X3D CPU Burns Up, Kills Itself & The ASUS X670 Motherboard
AMD Ryzen 7 7800X3D CPU Burns Up, Kills Itself & The ASUS X670 Motherboard
There are reports of AMD's Ryzen 7 7800X3D CPUs burning up and killing the ASUS X670 motherboard with it from the AMD subreddit.wccftech.com
Caporegime
Yeah... that WCCF framing..... "The CPU is burning and causing the death of the motherboard"
Last edited: