More
referral
Increase your income with Hive. Invite your friends and earn real cryptocurrency!

Adding 2nd 6600XT crashes rig?

Hey gang,

I’m setting up a small rig with some 6600 XTs and unfortunately it never seems to want to accept more than 1 card at a time.

Each card individually will boot up and mine fine, but if I add a 2nd card, it will sometimes boot for 10-20 minutes, then crash, and eventually send my rig offline. This happens even if I clear all OC and just run stock.

I’m on a Z390-A pro motherboard, 8Gb ram, and the latest version of HiveOS. PCI-E is set to Gen2 (tried on Gen1 also) and above 4G is enabled.

I’ve tried with lolMiner, TeamRedMiner, Pheonix miner, and NBminer with same results. (Device 1 detected as crashed, or Gpu1 detected dead)

I’ve also noticed that if I don’t turn on auto fans, the 2nd gpus fans will not run during mining, only at start up.

I’ve switched risers, tried directly on board, switched psu cables, put the risers on individual psu cables also, but nothing seems to want to work. I also tested every PCI-E slot individually and they are all working.

Has anyone got a multi-6600XT rig running in HiveOS? Did I just lose the silicon lottery?

I’m on version 0.6-210@210918 (which was the latest until today)

Cards:
XFX QICK 6600xt
AsRock Challenger 6600XT
MSI Mech 6600XT (RMAed because it wouldn’t even run solo)

PSU: EVGA 1000W G+

I’ve also noticed sometimes (but not always) the 2nd card will generate a “GPU driver error, no temps” error, and the detailed message is “amdgpu: Msg issuing pre-check failed and SMU may be not in the right state!”. I also see some “amdgpu - Failed to read reg” errors when this happens

Bios settings:

Settings\Advanced\PCIe/PCI Sub-system Settings
PEG0 - Max Link Speed: Gen2
PCI Latency Timer: 32 PCI Bus Cl…
Above 4G memory/Crypto Currenty Mining: Enabled
Re-Size BAR Support: Disabled

Settings\Advanced\Power Management Setup
ErP Ready: Disabled
Restore after AC Power Loss: Power On
System Power Fault Protection: Disabled
USB Standby Power at S4/S5: Disabled

Settings\Advanced\Windows OS Configuration
Windows 10 WHQL Support: CSM
MSI Fast Boot: Disabled
Fast Boot: Disabled

Settings\Advanced\ACPI Settings
Power LED: Blinking
CPU Over Temperature Alert: Auto

Settings\Advanced\Integrated Peripherals
Onboard LAN Controller: Enabled
LAN Option ROM: Disabled
Network Stack: Disabled
Sata Mode: AHCI Mode
Sata1 Hot Plug: Disabled

Sata6 Hot Plug: Disabled

HD Audio Controller: Disabled

Settings\Advanced\Integrated Graphics Configuration
Initiate Graphic Adapter: PEG
IGD Multi-Monitor: Disabled

Settings\Advanced\Intel ® Thunderbolt
Discrete Thunderbolt Support: Disabled

Settings\Advanced\USB Configuration
USB Controller: Enabled
XHCI Hand-off: Enabled
Legacy USB Support: Enabled

Settings\Advanced\Super IO Configuration\Serial(COM) Port 0 Configuration
Serial(COM) Port0: Enabled
Serial(COM) Port0 Settings: Auto

Settings\Advanced\Super IO Configuration\Parallel(LPT) Port Configuration
Parallel(LPT) Port: Enabled
Parallel(LPT) Port Settings: Auto
Device Mode: STD Printer Mode

Settings\Advanced\Wake Up Event Setup
Wake Up Event By: BIOS
: Disabled

Settings\Boot
Full Screen Logo Display: Enabled
GO2BIOS: Disabled
Bootup Numlock State: On
Info Block Effect: Unlock
POST Beep: Disabled
Boot Mode Select: Legacy + UEFI

Update:

On the latest version (0.6-211@211107) I’m now getting much more stable mining with the 6600 XTs.

Ended up have 1 bad riser that I swapped as well, and suspect a 2nd one (one gpu will consistently hang after about 7 hours), so I’ve ordered some replacement ones to get that one swapped as well.

Definitely seems to have fixed the driver issues though

This topic was automatically closed 416 days after the last reply. New replies are no longer allowed.