Showing results for 
Search instead for 
Did you mean: 


Journeyman III

Why does my 3900x ryzen suddenly go out after a series of beeps?

I recently invested my savings in upgrading my i5 6600k to a 3900x ryzen, the computer is composed of the 3900x processor, a Gigabyt X570 ultra motherboard, power supply Corsair source HX650, x2 Corsair Vengeance LPX DDR4 3200 PC4-25600 16GB 2x8gb CL16, Toshiba XG5 NVMe 1TB SSD M.2 SATA 3, 2x Samsung Evo 960 (256/500) and GTX 1070 ti MSI Armor. The motherboard bios are up to date as are all drivers, Windows updates and the problem happens when unexpectedly the pc emits two series of 3 or 4 beeps followed to shut down completely as if the current had been cut off. It happens either while playing or with the computer at rest without a set time value. The memories have been fully tested several times by various programs including Windows and pan the tests correctly without any error. The voltages of the power supply are correct and stable, the temperatures of all the equipment as well.I have also tried to pass prime 95 and does not give errors with tests of 15/20 minutes.

The processor under charge in video games does not exceed 65º Celsius, the memories were not stable with its XMP profile, I tried other configurations but gave me blue screen capture so finally I stopped them from stock, still the problem persists, I’m starting to regret having opted for AMD..

3 Replies
Adept III


What does BIOS beep sound mean?

1.AWARD BIOS 1short beep :System normal 2 short beep :CMOS Error 1 long beep and 1 short beep:Memory error 1 long beep and 2 short beep:Graphic card error 1 long beep and 3 short beep :AGP error 1 long beep and 9 short beep :Memory Error Continuous long beep :Memory not correctly installed Continuous short beep :Power supply unit failed 2.
AMI BIOS 1 short beep :Memory Error 2 short beep:Memory parity check error. 3 short beep :basic memory 64K address check error 4 short beep :Real Time Clock malfunction . 5 short beep :CPU error 6 short beep:Keyboard error 7 short beep :CPU interruption error 8 short beep:Graphic card error 9 short beep :Memory error 10 short beep :CMOS error 11 short beep :CPU cache memory malfunction
Try running on just 1 stick of ram, try a different memory channel (consult manual for which lane to use), re-seat GPU and check 6 pin  and 8 pin power cables correctly plugged in (use 2 separate 6/8 pin power cable instead of the split end type cable). in Bios set PCI-e to Gen 3. (also check PCIe x 16 Lane dedication it may be set to your NVMe drives).
If you could removed any M2 drives from system that are not needed to operate Windows (only keep Windows Boot Drive) and test with a bare bones system.
PCIEX16 Bifurcation Allows you to determine how the bandwidth of the PCIEX16 slot is divided. Options: Auto, PCIE 2x8, PCIE 1x8/2x4, PCIE 4x4. (Default: Auto)

Are you comfortable going into the bios and checking what your vcore voltage is?  I've seen videos where default bios settings are set entirely too high and you could be getting an overvoltage shutdown, just a suggestion..  I'd also try one stick of ram at the default 2133mhz as mentioned above...  I'd also pull out the other hard drive that doesnt have the os on it just for testing.. 

make sure you have the amd chipset drivers installed, and check the ryzen balanced powerplan in the windows power settings..


Like others have said above, you could try just using one stick and so some stress testing- reboot if it passes then manually dial in your memory kits settings and voltage as stated in your memory kits manufacturer specs then redo the tests. If it passes again, redo the tests but with two sticks and be sure memory placement is at the reccomended slots for dual channel as stated on your motherboards manual.

Would recommend you add using aida64 for full system stability tests. I've had new builds that I Overclocked before that had random App CTD or system reboots while using blender even though it had already passed 3hours of prime/mem tests. It would only fail after 2 hours of full system stability stress test from aida64, and it will usually tell you what test it failed at.

If all else fails and your still getting errors even after testing with a new memory kits, you could try testing with an older bios as your last resort,  one way you could do so without having to reflash is by using your bBios since updating your bios only updates your mBios (unless your at bBios and updated that one instead) so your bBios should still be using an older one.

Gigabyte boards usually have dual bios (mBios and bBios, led indicator located near the PCI-e x16 slot tells you what bios your currently at) and have a physical switch usually located at the bottom part of your board (check your board manual).

Be warned though that switching bios takes time while it does memory training at first boot, and may do a couple of reboots before its done. May also loose display signal even when windows is clearly running normaly, in which case you may have to reset your PC using the power switch before it will show your display normally again.