NBMiner
NBMiner copied to clipboard
NebuTech
RTX 3060TI - GPU shows FAN 0 while locks hashrate at 19 Mhs
Hello everyone,
From times to times the NBMiner locks the GPU hashrate on 19 mhs and show 0 fan. RTX 3060 TI with Hynix memory. Occours with any overclock even at low overclocks as 1300 Core Clock and 1500 Memory Clocks.
Runing on HiveOS, Driver 510.68.02.
Thanks!
Hi there Seen this in a few other topics as well I am experiencing this too. Issues happens on 2 of my 3060TIs , 1 3070TI and happened once on an 3080TI Note that the 2 3060TIs are Palit , and 3070TI and 3080TIs are Gainword (So essentially all the same model. Since Palit and Gainword are the same company) And It's totally random... No set time .. Sometimes it goes for 10 mins , sometimes at the very beginning it just shows a 0 Fan speed and that GPU doesn't work Sometimes It just stops hashing, sometimes it locks to really low numbers and keeps hashing (And getting shares!)
There is certainly something going on here...
Bare in mind in the same Rig , I also have a 2060S that is Palit but as we know it's Full Hashrate and it has 0 issues. Never had
Also it's note worthy the problem isn't solely with NBMiner... with other miners I also run into different types of problems with these cards... For example latest T-Rex had issues with my 3060TIs.. But not LOL Miner!
It's the second GPU that crashed in this mining session...
IT also locked at 1/3 speed...
Also the HiveOS Hashrate Watchdog does not work on NbMiner so it doesn't reboot
Same problem here - random cards(3060/3060ti): fan 0%, nvidia-smi ERR! on fan, hash rate low, mem util. 0%. Nvidia-settings: fan: unsupported. Restart only helps. Sometimes it happen after 5 minutes and later.
same problem( crash on GPU, then show GPU freq 0 mhz, hashrate only 1/3, up/down memory clock no impact to this issue. awaiting next release..
Same problem here - random cards(3060/3060ti): fan 0%, nvidia-smi ERR! on fan, hash rate low, mem util. 0%. Nvidia-settings: fan: unsupported. Restart only helps. Sometimes it happen after 5 minutes and later.
what driver?
Same problem here - random cards(3060/3060ti): fan 0%, nvidia-smi ERR! on fan, hash rate low, mem util. 0%. Nvidia-settings: fan: unsupported. Restart only helps. Sometimes it happen after 5 minutes and later.
what driver?
510.60.02
Same issue here, have a mix of 3060's and 3060ti's and one of them will randomly drop to 18-19 mh/s w/ 0% fan and only a reboot fixes it until it randomly happens again...
510.60.02 driver NBMiner 4.1
Bare in mind in the same Rig , I also have a 2060S that is Palit but as we know it's Full Hashrate and it has 0 issues. Never had
Also` it's note worthy the problem isn't solely with NBMiner... with other miners I also run into different types of problems with these cards... For example latest T-Rex had issues with my 3060TIs.. But not LOL Miner!
Scratch that, It happened on my 3080 EVGA as well... So it's not brand related...
I now am testing all the GPUs with a 200Mhz lower memory clock as per NBMiner suggestion (400 lower in HiveOS as you all know) Will post the results
I now am testing all the GPUs with a 200Mhz lower memory clock as per NBMiner suggestion (400 lower in HiveOS as you all know) Will post the results
Well, Scratch that as well As soon as the system came up, a new GPU (3070TI MSI Ventus 3x) lost the fan and locked at 1/3 hashrate... It is defo an NB Miner problem
Edit: Everybody please be aware this is fairly new, just released... AND it's a BIG change... so it will take time for this to be fully optimal and operational.
What is your OC on 3070TI ? I was getting same error every 30 minutes.
I was on 2400 memory for 3060TI and now on 1900 and have been runing for an hour an 15 minutes with out problems.
Anyway definitly NB miner team will have to work on.
Not sure on windows.... I had this going on i have been ok for about 4 hours so far on all the rigs try driver: 510.68.02 Use command: nvidia-update-driver https://us.download.nvidia.com/XFree86/Linux-x86_64/510.60.02/NVIDIA-Linux-x86_64-510.60.02.run
Seemed to work for me i tried clock and all that stuff nothing seemed to get me past an hour but this here! GL
Not sure on windows.... I had this going on i have been ok for about 4 hours so far on all the rigs try driver: 510.68.02 Use command: nvidia-update-driver https://us.download.nvidia.com/XFree86/Linux-x86_64/510.60.02/NVIDIA-Linux-x86_64-510.60.02.run
Seemed to work for me i tried clock and all that stuff nothing seemed to get me past an hour but this here! GL
i had this same driver from beggining and problem is exists. On start for first few hours works ok, now every few minutes hanging.
I confirm it's a random problem with Hive's last version and NV drivers to 510.60.02
Maybe someone tested driver version up down? Looks like more is using 510.60.02. I'm testing now 510.68.02... first result - completely device error (not about temperature).
same here after some time fan drop to 0 and mh drop to 17.
3060 & 3060 Ti
how long for you until it drop to 17?
Tried 200Mhz lower clocks on all cards.. as you can see cards dropped to Fan =0 one by one over time
It's un-usable at the moment
same here after some time fan drop to 0 and mh drop to 17. 3060 & 3060 Ti
how long for you until it drop to 17?
It's absolutely random... Sometimes 2 minutes , sometimes 2 hours...
Maybe someone tested driver version up down? Looks like more is using 510.60.02. I'm testing now 510.68.02... first result - completely device error (not about temperature).
no changes, still problem even on different drivers.
Using 12x 3060 Tis Galax Hynix memory. I have decreased Mem Clock from 1860 to 1000 on those cards that suddenly can't be detected and showing 19 mh/s. It doesn't seems to be the cards issues... Happening on a random basis from 30-45 min uptime on my end.
HiveOS 0.6.217@220428 Driver 510.06.02
Reboot is needed to resolve.
Update- Apparently another rig running 3060 Ti Samsung memories windows 10 pro n, drivers 512.15 Uptime 21 hours no errors nor Hashrate drop (stucked)
same here after some time fan drop to 0 and mh drop to 17. 3060 & 3060 Ti
how long for you until it drop to 17?
sometimes 2 3 minutes somtimes 30 mins....
someone has tested on rigs without 1/4 pcie splitters or m2 to pci x1 adapter, all my lhr rigs have splitters and m2 adapters so im thinking that can be the problem
My English is bad. I tried EVERYTHING. nothing helps, I think the problem is in the miner
someone has tested on rigs without 1/4 pcie splitters or m2 to pci x1 adapter, all my lhr rigs have splitters and m2 adapters so im thinking that can be the problem
It's not the splitter... I have no splitter Tried on x1 , x16 ... nothing works
someone has tested on rigs without 1/4 pcie splitters or m2 to pci x1 adapter, all my lhr rigs have splitters and m2 adapters so im thinking that can be the problem
It's not the splitter... I have no splitter Tried on x1 , x16 ... nothing works
try to lower the memory clock with 300, my rigs 10 x 3060 in hiveos have uptime about 2h 3h and fans on 100% give a try, i think this is the solution for the moment.