I am trying to do some troubleshooting here with an EVGA FTW 3080 ti. No matter what other settings i’m putting in the card is throwing a GPU temps are lost error on Trex as soon as it goes over 1300 on the mem clock.
Just bought this 2 days ago and put new pads on it. Not sure if it needs a bios update or what, but any help would be appreciated.
what are the memory temps like at 1300 on the mem? id wager its thermal throttling.
I don’t think it is thermal throttling GPU #3 is the problem child I cant raise over 1300 on the mem clock. What i find strange is that the hash rate is so high with such a low clock at a low voltage. Maybe EVGA messed with the clock settings on the FTW3 Ultras or something.
Run nvidia-smi -q to see if it’s throttling for any reason. Your mem temps look fine.
No throttling at all that I can see.
PCI
Bus : 0x06
Device : 0x00
Domain : 0x0000
Device Id : 0x220810DE
Bus Id : 00000000:06:00.0
Sub System Id : 0x39673842
GPU Link Info
PCIe Generation
Max : 1
Current : 1
Link Width
Max : 16x
Current : 1x
Bridge Chip
Type : N/A
Firmware : N/A
Replays Since Reset : 0
Replay Number Rollovers : 0
Tx Throughput : 9000 KB/s
Rx Throughput : 26000 KB/s
Fan Speed : 95 %
Performance State : P2
Clocks Throttle Reasons
Idle : Not Active
Applications Clocks Setting : Not Active
SW Power Cap : Not Active
HW Slowdown : Not Active
HW Thermal Slowdown : Not Active
HW Power Brake Slowdown : Not Active
Sync Boost : Not Active
SW Thermal Slowdown : Not Active
Display Clock Setting : Not Active
That all looks good. Have you ruled out any riser/cable issues by swapping them around?
could this be an issue
the GPU link info for my fully function 3080 ti
GPU Link Info
PCIe Generation
Max : 2
Current : 2
vs the problem child.
GPU Link Info
PCIe Generation
Max : 1
Current : 1Looks like there are some errors with pci but from a different card
root@ThirdRig:/# nvidia-smi dmon -c 1 -s e
gpu sbecc dbecc pci
Idx errs errs errs
0 - - 0
1 - - 9
2 - - 0
3 - - 0
4 - - 3869
5 - - 0
6 - - 0Gen 1 is all that’s needed for full hashrate, so that won’t be your issue.
Well that is the extent of my ideas to try and get the same performance out of the card. Still low 80s is not terrible. I dont get why trying to go over 9800 MHz on the clock is causing it to die. Might have to take it off the rig and run it in windows to see if the same thing happens.
I appreciate the help Keaton.
Did you swap the cards to see if they behave the same way in the other slot/risers/cables etc?
Yep swapped the riser out with a new one in a new slot and no changes.
wouldnt hurt to try another bios and see if it makes it better or worse
Further update on this, I have since tried flashing the bios from 94.02.71.80.86 to the earlier varsion 94.02.71.80.74 and switched over to lolminerv1.47. Better hashrates with their unlocks. but still cant overclock beyond +1100mhz without constant crashing. Gave it it’s own 850w power supply too wondering if maybe I was running into something with the efficiency, but nope that didn’t change anything.
If you’re still looking to try other bios’, my ftw3 came with 94.02.71.00.B0 and has been working great without any hiccups. It could just be that you got a card that doesn’t like overclocking too. But I’d give that bios a try at least.
Did you manage to solve this issue? I’m running into the same problem with a gygabyte 3080 ti that won’t go over +1125mhz on mem without instant crash
This topic was automatically closed 416 days after the last reply. New replies are no longer allowed.