Z270X Serious Voltage Issue

xcomponent

Member
Hello!
I've registered here as a last resort in trying to find out what's going on with HWiNFO64 on my GB Z270X Gaming 5 motherboard.
Here is the problem. After starting up HWiNFO, the program starts to show voltage drops within the IT8686E controller measurements.
Basically ALL the voltages, measured by the IT8686E are affected. (12v, 5v, 3.3, DRAM, etc - all of them but not at the same time)
This happens on random time intervals (might be 3,5,10 minutes) but it DOES always happen, no matter what.
I can assure you, I've tried everything I can possibly can harware-wise and also within the BIOS settings, C-states, you name it.
I've changed the PSU, the videocard, the fans and the pump configuration in case of some sort of a shortage - no difference.
Now, what got me to the conclusion that this is a driver issue is the following:
I've ran Gigabyte's System Information Viewer in order to record the voltages, then in the background, I left  HWiNFO64 running as well.
When HWiNFO "drops", GB SIV registers it. I'm attaching a screenshot.

I've made countless tests and these "drops" do not happen when HWiNFO is not running in the background. 
It's like it blocks the IT8686E and GB SIV reads results close to zero when it's running. 
I'm with i7-7700k, z270x Gaming 5, Windows 10 x64.
Can anyone else with the same board try it out?
Here is the screenshot:
 

Attachments

  • both.png
    both.png
    109 KB · Views: 5
This seems to be a collision between both tools, I assume the GIGABYTE tool doesn't used synchronization mechanisms which are supported by other tools like HWiNFO to play nice together.
Do those drops happen when only HWiNFO is running and no other system monitoring tool? Also does it happen when running SIV together with other tools like AIDA64 ?
 
Martin said:
This seems to be a collision between both tools, I assume the GIGABYTE tool doesn't used synchronization mechanisms which are supported by other tools like HWiNFO to play nice together.
Do those drops happen when only HWiNFO is running and no other system monitoring tool? Also does it happen when running SIV together with other tools like AIDA64 ?

Yes, it happens when only HWiNFO is running, it also happens with HWiNFO + SIV, but it doesn't happen when only SIV is running.
I've switched the CPU, I'll let you know. I don't see a solid reasoning, but I decided to try with a different CPU than 7700k
to eliminate a possible CPU issue. 
I haven't tried AIDA64, might do that as well. At this point the issue is present only when HWiNFO is running.
And most probably it's a reading issue, otherwise the system would be highly unstable with drops like that and I don't
see any physical effect, no freezing, errors or anything like that.
 
P.S. Not a CPU issue, this is with a G3900, the difference here is that the thermals are also wrong now (again - IT8686E). That didn't happen with the 7700k.
Here:
 

Attachments

  • Untitled.png
    Untitled.png
    62.2 KB · Views: 6
Must be some kind of readout error. Wondering if it happens with others like AIDA64.
 
Martin said:
Must be some kind of readout error. Wondering if it happens with others like AIDA64.

I'm now runing AIDA64 and HWiNFO to see if a drop will be registered in AIDA as well. What I've noticed is the issue appears faster (on a shorter intervals) on IDLE.
This is just a hint if someone is willing to reproduce it. Since the G3900 idles (full idle) much rarely than the 7700k due to some W10 black magic in the background and being only 2 cores,
it takes awhile but it does appear eventually, I'll post a screenshot when I'm done  with AIDA.
 
You might also capture a HWiNFO Debug File when running alone and the problem appears. I could then check more details what's going on there...
 

Attachments

  • HWiNFO64.rar
    1.4 MB · Views: 1
I have checked that data and I don't see anything suspicious there.
But.. I have had several reports from similar GIGABYTE boards, especially those that use the IT8792E chip. This chip seems to be problematic and reading data from it can sometimes cause several problems including a system crash. Since this problem can be reproduced on Linux as well (where a completely different software is running), this seems to be a hardware issue.
 
Martin said:
I have checked that data and I don't see anything suspicious there.
But.. I have had several reports from similar GIGABYTE boards, especially those that use the IT8792E chip. This chip seems to be problematic and reading data from it can sometimes cause several problems including a system crash. Since this problem can be reproduced on Linux as well (where a completely different software is running), this seems to be a hardware issue.

It might be, all I'm saying is I can't reproduce the issue with GB SIV or AIDA64. It only happens with HWiNFO64 alone or when HWiNFO64 is running as well.
 
Back
Top