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

Problem in my farm with GPU

Good I hope you can help me, I have 3 GPUs I am starting to form my farm, 1 rx580, 1 rx470 and unarx570 I move it with a corsair 750 source, normal motherboard for 4 gpu.

I have everything configured mining rvn with kawpow, the stan flash cards but when I want to put more core or mem to raise the hash more and be able to reach about 13 the miner works for approximately 40 minutes and restarts every so often, I put the cards at 12MH and there if it remains stable it does not restart, lower the watts consumption but it did not work, someone could help me please

What could be happening?

what do u mean by this…? you lower power limit? you lower the core voltage (vdd) ?

If correct, with vdd lower the power consumption that caused the temperature to drop, but it keeps restarting, it means that the restarts are not due to heat, I played with the values ​​at all.

if u set too low vdd, Core is not getting enoght power and gpu becomes unstable that will cause Crash. Sa. e thing happens with Core and Mem, when u reach the gpu limit, it will Crash. with those rx cards 850 should be safe vdd value. For 580 Mem 2150max Mem 1170. fot 570 Mem 1050 Core 1120. can say about 4xx cards. these are not max values just good safe values and these are good with eth, i havent mined raven, so its bossible that u need different settings for gpu… but for reference these should be a good start. defo not much higher values for stable mining. good luck

how much do you think should put the vdd?

my experience with 580 is that most of them becomes unstable below 840…depends on the card naturally,i would keep all 5xx vdd min 840, there is not too much difference in consumption if u go lower…with the expence of stability. 800 is way too low on yours 570´s imho, but that´s just my experience, its not a fact…could be that its working with some cards…

I changed everything to 840, look at these pig graphics, I can’t keep the rig from collapsing, I’m even thinking about weighing windows to see, the production is fatal.

give it a try with all, Mem 1000 Core 1000 dpm1 mdpm 1 aggressive under volting and fan 70 and see what happens. No power limits with 5xx cards needed.

And to this do I apply it too?

Radeon RX 470 4096 MB · Sapphire

sorry, dont know settings for that one, i didnt notice there was 4gb card.these values are just to test can u start up your rig at all, should be good enought for that 480 allso…check some setting for that one and go way below for testing purposes so u know is it even possible to boot up or is the issue other than your gpu settings.

ok working at the moment.

At the moment the equipment did not restart anymore, it is working stable, I would say rather that the line does not move from the charts, and now?

it restarted at 2 hours

But it did not turn off, only the miner was restarted, because the problem is always that the computer restarted completely, what should I do now?

image

ad some vdd (850) to core and experiment wirh mem and core values…one value change at the time so you know what fails and what works. I belive u can put 2050 mem and 1050 core to all cards, those are “normal” values for 5xx gpus.
For 570 u should be able to go mem 2050 vdd 840 core 1120 or even 1150.
For 580 me 2150 vdd 840 core 1170-90 is max reasonable value.
so… just go up with core/mem values until u find the limits, the card dies…then back up a bit . and last thing, add ref value. Research the right values to 470.

and the dpm and mdpm? do I leave 1 and 1 the same?

yes both states, dpm/mdpm = 1

It loads me crazy, look at the values I’m just setting the first 2, and it restarts after an hour, and the second is super low and still nothing

Epale, you know doing tests with the miner, I discovered that one of the card is the one that sends me to restart the miner, I remove it and the equipment lasts 7 hours or more, I do not know if it is any configuration, it is the 470 card