I do have the v3 and same issue with my 3060ti…
Please Help
one gpu gainward 3060ti stable 44-45mh (1400 1800 145) for 4months
now having problem only stable 38mh ( 0 0 135).
every time i change OC, it will crash, even with lower number .
i do update /downgrade hive os and nvidia, but i cant change it like before without crash.
im very confused, becouse if hardware problem why it stable with (0 0 134).
im realy need help from the expert
one gpu gainward 3060ti stable 44-45mh (1400 1800 145) for 4months
now having problem only stable 38mh ( 0 0 135).
every time i change OC, it will crash, even with lower number .
i do update /downgrade hive os and nvidia, but i cant change it like before without crash.
im very confused, becouse if hardware problem why it stable with (0 0 134).
im realy need help from the expert
Set core at 1500, and memory as high as is stable. If it crashes, reduce mem but leave the core. Repeat until stable.
1 Like
Hello Keaton
Please, advise me setting for Colorful 2060 super. I 've tried to configure core (-502; 1035; 1050; 1100) and mem (1500; 1900; 2000; 2100) but it cannot pass 37mh/s. Thanks!
Or can someone advise me how to configure Colorful model? Thanks!
Start no core clock, no power limit, find the highest stable memory clock that gets you the highest hashrate, then find the lowest locked core clock that maintains that full hashrate.
Hi!
I will try as you advised
Thanks!
Hi Keaton
I spent 120 minutes doing as you advised but it couldn’t reach 37mh (core clock from 0 → 100, 500 after -100 → -200, …502).
My motherboard info:
Values above 500 are treated as locked core clocks, you don’t want a core offset there (1-499 or any negative number). What number did you end up at on the colorful?
Hi Keaton!
First, I set core = 0, PL =0, mem = 2100 and reboot machine.
Second, when miner star the stable shown 29.2Mh, I began for core from 100, It drop 26mh and when I set core 500 It were 25mh.
I 've set core -100 until 502 and still keeping mem 2100, the best stable was 36.3mh but the temperature is very high and power get 120w.
So, I adjusted as before (picture that you saw)
Try setting locked core to 1000 then memory to whotewer is stable and set power limit to max wich is 210 W in your case, i had one card whish behave similary till i set it to higher power limit 5 below maximum, if i get lower has rate drops.
1 Like
Hi Mozgala
Please show me your stable
Thanks!
i run it at 1400 memory gives like 41,5 MHS
So set core 1000, mem 1400, PL 210
1 Like
Hi!
I set as your instruction. it’ s increasing
Thanks
HI Mozgala
After 30 minutes setting as you advised, I have a stable for Colorful 2060 super.
And now, Have I need to decrease PL from 200w to 125w?
Please tell me
Thanks!
you can leave it that way, power draw is limited by locked core not PL itself, i think it has something to do with power tables in card bios so on higher power it is able to get more juce to chip or memory and with lower PL it cuts power deliver to some card parts resulting in lower has rate then optimal for that card
You can try and decrease core in increments of 15 to check if it will keep that hashrate at lower core, it should lower power draw as well.
Each card is diferent i have some 2060S runing at 930 core some at 1005 some at 960 each card from each manufacturer and batch is diferent, silicon lotery sometimes you get beter chip sometimes worse
1 Like
Hi Mozgala.
Thank you so much!
Hey, this happened to me after a recent update too! I’ve looked far and wide. checked my risers, cables, OS etc. it turned out that somehow the “enable 4g encoding” was turned off in my bios for some reason.
best of luck!
Z270P-D3-CF Gigabyte Technology Co., Ltd. (F7 11/08/2017)
“MALFUNCTION” ！ bios 选择 Gan2 选择 pcle gpu显示，不用HD Graphics 610 · Gigabyte核显，我的问题是这样解决的，完美工作 6 gpu。 msi 3060ti lhr 2x！
This topic was automatically closed 416 days after the last reply. New replies are no longer allowed.