Ryzen 3600 hwinfo 6.20, CPU CCD2 (Tdie) 206 degree?

edhunter

Member
Hi,
I wanted to report some strange readings on ryzen 3600 cpu.
In the attached screenshot - cpu ccd2 sensor reports 206.9 degrees. I dont believe that this is possible.

Also another issues during stress test (not in this screen shot) Tdie sometime dipped to -10 degrees. Also voltage spike to about 2v i think at the same moment. Also not correct readings I believe.
 

Attachments

Martin

HWiNFO Author
Staff member
The CCD2 temperature is a bug in HWiNFO and shall be fixed in the next (Beta) build of HWiNFO.
Which Tdie value dropped to -10 C, was it the same CCD2 Tdie, or the CPU (Tctl/Tdie)? It would be great if you could capture this glitch in a screenshot.
And Happy New Year !
 

Zach

Active Member
Happy New Year!

Hi,
I wanted to report some strange readings on ryzen 3600 cpu.
In the attached screenshot - cpu ccd2 sensor reports 206.9 degrees. I dont believe that this is possible.

Also another issues during stress test (not in this screen shot) Tdie sometime dipped to -10 degrees. Also voltage spike to about 2v i think at the same moment. Also not correct readings I believe.
Once I had the same issue in my 3600. Apparently this was caused by excessive DRAM and Infinity Fabric OC like I see you have (1900MHz). The fact is that wasnt an all time issue but at random boot up times. When IF and ram backed down to *1800MHz this never occurred again.

There are lot of different instabilities besides BSODs or obvious app crashes and errors. You have to keep an eye on them and associate the findings.

*EDIT: typo (1600)
 
Last edited:

edhunter

Member
Happy new year Martin :)
Thank you for the great HWiNFO tool - it's overclockers best friend :)

About the -10 C, I think it was CPU (Tctl/Tdie), but I am not so sure ... if it happen again will take screenshot.
Actually I might have it in the logs since I was making some stress tests and logged the sensors data for some of them. Tomorrow will try to find it.
 

edhunter

Member
I found it in one of the logs - it was the CPU CCD3 (Tdie). But how this is possible. As far as I know 3600 has only one chiplet.
In my other logs CCD3 (Tdie) reading does not exists.
 

Attachments

Martin

HWiNFO Author
Staff member
The non-existing CCD2/3 reporting is a bug in HWiNFO, that will be fixed in the next (Beta) build.
But if the CPU (Tctl/Tdie) would report an invalid value, or Thermal Throttling would be asserted when temperatures are OK, that would be a different issue in the CPU firmware.
 

edhunter

Member
The Stilt, it's MSI b450 tomahawk max board using bios 351 beta (not listed on msi site anymore, but it's the one before the final v35)
 

edhunter

Member
Happy New Year!



Once I had the same issue in my 3600. Apparently this was caused by excessive DRAM and Infinity Fabric OC like I see you have (1900MHz). The fact is that wasnt an all time issue but at random boot up times. When IF and ram backed down to *1800MHz this never occurred again.

There are lot of different instabilities besides BSODs or obvious app crashes and errors. You have to keep an eye on them and associate the findings.

*EDIT: typo (1600)
Yes you are correct, I have IF oc to 1900mhz, and ram to 3800 with tight timings. I has been conducting stress tests to find out if it was stable. I had to relax some of the secondary timings a bit to get it stable.
 
Top