[SOLVED] New rig sudden rebooting problem

[Exemplis]

HI!
Week ago Iv build new PC.
Ryzen 3600
Arctic Freezer 34 esport DUO
MSI Mortar Max
Crucial Balisix Sport LT 2x8Gb 3200
Saphire Pulse RX 5700
Seasonic Focus Gold 550W

First time everything was alright, Im playing little bit with ram, pushing into 3600mhz 16-16-18. Playing with GPU tweaking, flashing XT bios, but after few tests I’m decided to reroll it back due high vrm temperature.
Playing last few days normally. But suddenly I get rebooting my PC, without any BSOD or any errors. I checked and found out that before rebooting, the headset (USB) appear to have some cracking popping noise. Seconds before rebooting, they increase. More intense game cause faster rebooting. It’s totally not an overheating because I check it in hwinfo, CPU max t 71c, GPU around 80-85 (It’s pretty cold for navi). I don’t oc’d CPU or GPU.
CPU passes stress tests without any problem, GPU under FurMark is ok too. Mem test running 30mins without any errors. But in game system is unstable, 5-10mins in R6 siege causes rebooting, Civ6 after 40-60mins.
I’m trying to change GPU for gtx1080 and its working fine after few hours play, after this I’m putting RX5700 back and... its working fine.. and Im playing rest of the day. On the next day problem returns.
Check one more time my observations:

1. In usb headsets (or monitor speakers) some popping noises. It appears after few minutes playing and amplified, before seconds to reboot sound is totally crap.
2. After rebooting I have problem with USB devices. Headset or keyboard or WiFi adapter or both of them doesn’t work. Only after reconnecting its working again.

I have a warranty for all pieces but I don’t know what exactly is the problem. Can you have some suggestions?

 

[bmockeg]

If your DRAM is not running stock then that is definitely your problem. You have to stress test when you change values to determine if it is stable. 80-85c isn’t cold for a GPU. Not sure what kind of case air flow you have but it looks pretty poor. My 1080ti never topped 58c while maxed out. Also your CPU temp is fine if you were stress testing but seems high also.

 

[zhyf89th]

First, 30min of memtest is too short. Second, have you manually set SOC voltage to 1.1v or lower? Because this voltage scales with memory frequency, and above 1.1v, it only brings instability. Use some software to get the sensor voltage readings. I have similar problem as you that my system reboots suddenly when I run Prime95 and some intense gaming.

After you set the voltage correct (1.05-1.1 recommended, if you haven't done so), you can check with prime95, 30min probably would be enough to verify this. Also do more memory testing, memory error usually won't cause sudden reboot but BSOD.

Note: I have no problem with USB device, so the cause may be different, but worth a try.

 

[Exemplis]

I found a problem, this was dram, turn it to stock XMP profile and all working fine