GPU slot issue

sekinosekino Member Posts: 12
edited August 2018 in Mining
I have 6 GPUs Aorus 570 4GB and Asrock 81 PRO BTC motherboard, 1200W Evga PSU...
I have SLOTS 0-5 and 6 GPUs installed in rig, but what ever I put in SLOT 1 (longest slot near to CPU and GPU with HDMI OUT) I have few seconds after start showing temperature on that slot between 85-90C (real temp measured with infrared thermometer is like any other GPU in rig around 50C) and hashrate starts falling from 29 MHs to 12 MHs and after few minutes to 0. I tried to switch couple GPUs and risers and I have always same result. When I put any card in SLOT 1 it will do what I described above. If I put card to any other slot GPU works like a charm. SO IT'S NOT GPU or RISERS problem...

Tried on both Windows and Ethos. Tried different overclocking underclocking. Nothing helps...

Any idea what causes this issue? Motherboard? Any settings to change?

Thank you!

Comments

  • lablettlablett Member Posts: 331 ✭✭
    I think this could be related to the GPU being used to drive the monitor maybe. Enable the on-board GPU and plug the monitor into the on-board HDMI slot. If you do not have a monitor connected then connect a dummy plug.
  • sekinosekino Member Posts: 12
    Hi all, problem solved, so I figured out that SLOT 0 is not first closest (PCIE 1x) to the CPU but it's second one (PCIE 16x). So SLOT 1 is first one closest to the CPU (PCIE 1x). I was trying to switch wrong card for others in SLOT 1, but nothing changed..

    So SLOTS in ETHOS are numbered in this way starting closest to the CPU: 1-0-2-3-4-5-...

    Issue was with overheating of GPU
Sign In or Register to comment.