X870 will not post after switching itself off

Jerome said:
Just to provide the (hopefully) final update that AMD finally delivered my replacement 9800X3D on 28th Jan, almost two months after they received the RMA.

Thankfully it's in the past and I can move on and enjoy this build in its entirety! Thanks again to all.
Click to expand...
So all working with the new 9800X3D then?
 
Jerome said:
Just to provide the (hopefully) final update that AMD finally delivered my replacement 9800X3D on 28th Jan, almost two months after they received the RMA.

Thankfully it's in the past and I can move on and enjoy this build in its entirety! Thanks again to all.
Click to expand...

update bios if you haven't already, lets not do this again haha
 
ED209 said:
So all working with the new 9800X3D then?
Click to expand...

Very well indeed thank you.

WARburton said:
update bios if you haven't already, lets not do this again haha
Click to expand...

Did that first thing upon swapping processors back (the latest stable, non-beta version).

When it failed originally it was running with the latest BIOS (at the time) and was within 24 hours of adjusting my RAM to run at 6000Mhz (although the build itself was only 6 days old so it may or may not have been a contributing factor). It's now running at 6000Mhz for the last ~48hrs and no issues to report. I haven't enabled EXPO and I'm honestly put off by the idea of doing any further adjustments at this stage. I ran Cinebench yesterday to give it a thorough run around and no issues to report.
 
use HWinfo and just monitor your voltages for a little time. just incase you board is dumping too much in... what you set in bios is not always what you get

but also good am happy all is good, another PC Master race broth saved :cry: :cry: :cry:
 
Last edited:
freddie64 said:
You sure EXPO is not enabled? Thought it had to be to run 6000Mhz.

Glad you are up and running though. :)
Click to expand...
6000 has a JEDEC spec so it is possible, think Crucial sell them with a super high latency (C48?).

Would expect the odd BSOD or worst case: silent data corruption to occur if it is running tighter than JEDEC timings at stock voltage (1.1v).
 
Thanks for the information everyone.

The only option I changed in the BIOS was changing the RAM from 4800Mhz to 6000Mhz, no other settings were touched.

While looking at the voltages in HWinfo I thought I'd look at the RAM and would I be correct in saying the timings are running at CL50 instead of CL30?

The RAM itself consists of two CMK64GX5M2B6000Z30 sticks which should run at CL30 if I'm not mistaken?

I could go ahead and enable EXPO if that's advised. I've attached a few screenshots including two of the voltages (RAM/CPU all under idle for what it's worth).

s77m2Dh.png


S6eEjJt.png


NwGSj2O.png
 
Jerome said:
While looking at the voltages in HWinfo I thought I'd look at the RAM and would I be correct in saying the timings are running at CL50 instead of CL30?

The RAM itself consists of two CMK64GX5M2B6000Z30 sticks which should run at CL30 if I'm not mistaken?
Click to expand...
It will only run at CL30 if you enable EXPO.

If you don't plan to enable EXPO, I'd leave it be, else it is likely to be unstable.
 
I went ahead and reset the speed to [Auto] and enabled EXPO. I figured if there's a new hardware fault it would appear at some point anyway and I won't be venturing into manual overclocking at any point.

I ran a few 5 minute OCCT stability tests on CPU+RAM/CPU/RAM with no issues.

Timings now appear correct and voltage stayed at a maximum of 1.19V on the CPU during the stress tests. CPU reached a maximum of 83C during the CPU extreme stress test and it idles at roughly 43C.
 
Last edited:
I'm afraid to say it's happened again.

Went out and came back to an unresponsive PC - upon reboot it won't post. Same colour fault as before (CPU/RAM).

I'll start troubleshooting and resetting the BIOS, but I have another processor (9600X this time as I got rid of the 7600X) being delivered tomorrow.

Seriously unbelievable.

Is it possible to have two dud processors in a row that die after a week's usage? Do I have another piece of hardware which causes the 9800X3D alone to commit suicide? The 7600X worked fine for 6 weeks on exactly the same settings. I'm honestly beyond dismayed.
 
Last edited:
Joxeon said:
Is the ram in A2B2 slots?
Click to expand...

Yeah they are.

No configuration changes since the original 9800X3D died (where I tried a different PSU, RAM, and even a different motherboard prior to RMAing the first CPU). Used a 7600X in place for 6-7 weeks with the same settings and no issues whatsoever. Second 9800X3D has died after roughly the same amount of time (~1 week).

Just filled out the RMA form with AMD and dreading the next long wait.

I read a few posts over on a different site about people suspecting ASRock boards to be killing 9800X3Ds but despite there being a couple of dozen samples, out of many thousands using this board/CPU combo it's hard for it to be conclusive.

Two 9800X3Ds dying in a row has to be some serious bad luck on my side though.
 
Yes I was running the latest (non-beta) bios. I upgraded to the beta-version after the system stopped posting again (just to rule it out).
 
If anyone has recommendations for a replacement motherboard that would be much appreciated.

I tested out an Asus X870 board back in November and detested it (it cost more than the ASRock and didn't even have any lights to indicate specific post issues).

Ideally I'd like to continue with an X870/WiFi combo although I'm flexible. Budget friendly if possible as I have no idea if I'd be able to get a refund on this ASRock board due to it technically being unproven that it's the culprit and that I bought it almost 3 months ago - therefore I'm unfortunately 200 down already.

At this stage I'd rather replace the motherboard as well as the CPU because I really don't want to face this a third time. Either the motherboard is murdering this CPU model specifically or I've received two duds in a row (which I'm lead to believe is more rare than obtaining a 5090FE back on launch day)
 
Last edited:
After switching to a 7800X3D which arrived today I can confirm the system works fine again.

Therefore it can be concluded that it's another dead 9800X3D.

My intention is to RMA it once more and I'm considering switching to a B650E board prior to its arrival to rule out any issues with this particular motherboard. Considering the previous 7600X worked fine for 7 weeks in the same configuration and if this 7800X3D also works fine, it's hard to blame the board, however. Two faulty 9800X3Ds in a row though is surely reaching beyond improbable.
 
Last edited:
Jerome said:
After switching to a 7800X3D which arrived today I can confirm the system works fine again.

Therefore it can be concluded that it's another dead 9800X3D.

My intention is to RMA it once more and I'm considering switching to a B650E board prior to its arrival to rule out any issues with this particular motherboard. Considering the previous 7600X worked fine for 7 weeks in the same configuration and if this 7800X3D also works fine, it's hard to blame the board, however. Two faulty 9800X3Ds in a row though is surely reaching beyond improbable.
Click to expand...
Again @jigger probably as a good insight on this, hes the local expert on CPU's dyeing

there was a high fail rate on the first run of 7800x3d's(i got one), maybe you are that unlucky
my board was a B650e so its not just the X board's, what i wouild say is if the 7800x3d works keep it, the differance in the 2 chips is no that big.
 
Last edited:
You must log in or register to reply here.
Back
Top Bottom