9820X VCore showing as 1.888v!? MSI X299 Tomahawk Artic

[OfficerDD]

Hi Guys

 

TLDR - Second 9820x died randonly after 3 weeks on a 4.2 @ 1.2v OC. Intel replaced on warrenty. Replaced installed but is showing stock vCore is running at 1.888v. Computer posts and runs cool but just seems way too high and makes me wonder if that killed my previous CPU.

 

Full story:

 

Recently purchased what i deemed as 'a good deal' on ebay. I game but do 3D modelling and 4k videe so X299 is a great fit:

• MSI Tomahawk Artic X299
• Intel i9 9820X
• 32gb 3000mhz DDR4 Corsair Vengance RGB
• Corsair H100i Cooler

Price paid for the bundle (made an offer) was about the same as a retail 9820X alone. It was all put into a Define R6 with my existing Gigabyte 1080 Gaming, New 850w Plat Seasonic and a variety of NVME and SSDs. I benched it on Firestrike and Cinebench and was a bit annoyed it came in slower than my 5960x running a 4.4 OC with 32gb 2133.

I did what i deem a mild OC to 4.2ghz @ 1.2v. It ran stable for maybe 4 weeks with no crashes or BSODs. One night after some gaming i slept the machine, next morning it turns on, lights up but no post. Noticed a 00 code on the board LED. I did some research and suspected dead mobo or CPU.. pretty dogged. Hauled it all out to make sure it wasn't a short, tested another PSU, tried a new MOBO (since returned to Amazon) and it failed to even see a CPU. On a whim I raised an RMA Warranty with Intel who, to their credit, replaced it with no questions asked. Very pleased given it was second hand ebay purchase so could easily be Tray CPU and hence no warrenty coverage at all. Lucky boy me thinks!

 

So I just rebuilt it, default (latest 7B05v2D) BIOS settings etc no XMP anything yet. Literally just stock. I was about to start an OC but notice UEFI showing VCore 1.888v? This seems massively high! CPUZ and Coretemp in Windows show VID at 1.074/1.102v (see attached).

 

Has anyone else seen this behaviour? should i be worried? Having read up there is very little on OC'ing the 9820x (understandable it was a CPU that fell between the high end gaming/low end workstation crack). I did read 1.35v on this chip would be 'safe' but 1.8v I've never seen.

 

All help greatly welcomed!

[Jurrunio]

the stupid bios is probably showing Vccin as Vcore. Check HWinfo64, it shows both.

[Vishera]

1.888v will kill any CPU in the market,not necessarily immediately.

Looks like a bug in the BIOS...

[lawliot]

Have you checked with hwinfo64?

[OfficerDD]

Thanks guys. Yep looks like that is exactly the issue. Looking at MSI Command Center it confirms it (try HWInfo too just to clarify).

 

Assuming it is just a bug I want a safe 4.4 OC ideally. Any suggestions. So little on OC'ing this CPU but i guess anything on the i9 X series will work. 1.2v seems sensible.

 

 

 

 

[OfficerDD]

All confirmed in HWInfo that it was a false alarm.

 

Managed to get the 9820x to 4.4 very easily at under 1.2v too. Interesting that I still can't quite get it to beat my old 5960x in Firestrike. Seems to be the Graphics and combined scores are significantly lower despite both builds using the same Gigabyte 1080 G1 GPU and the same overclock running on that. Physics score and Cinebench much better now.

Seems like the 9820x isn't playing nicely with the GPU for some reason. Also indicates how little difference between the excellent 5960x there has been in 4+ years!

Anyone heard of anything like this?