Bitcoin Forum

Hi all.

Sorry if this has been asked previously, have searched here and Googled it, but, no luck.

I have a mining rig with an Asus Z270A and 4 Asus Rog Strix 1070, it has been working for around 3 months or so mining Zcash, dual Ether+Sia and Sia Coins.

A few days ago one card stoped working, I realized a few hours late and after restart it started to make some kind of "coil whine" related with the aura colours, brighter colours increased the sound, so disabled the aura, it made some weird sound, but... it's ok.

It worked again for the day and almost all night, when I weak up I got a black screen and no mining.

After 12 hours of tests have tried almost everything:

- Tried with 4 gpus, 3 gpus, 2 gpus, 1 gpus.
- Removed drivers and installed from different sources.
- Reinstalled windows.
- Changed connections.
- Removed risers and plug directly to mobo.
- Reset and upgrade bios.

And several combinations with tweaks and miners, no way, the miner (claymore and ewbf) start to mine, after a few minutes all the fans goes to 100% (or more ?) and blank screen, so I have to restart.

Finally tried a different miner, standalone Sia Coin miner and... it worked, for half an hour it run smooth, so, seems that VRam allocation is the trouble.

Tried again and set (again) the Vram to default, middle of the bar using the Asus tweak 2 software and then realized that the maximum Memory MHz that I was able to set was round to 8.400, but now, it let me to pull memory over 10.000 MHz !!!, same for CPU, over 2.000 MHz.

I droped it to minimum and it last for a long time, made a new test restarting and MHz levels appeared at the right place, but miners crashed (except siacoin).

After so many hours Im getting mad and absolutely frustrated, no idea what to test, just have 2 clues:
- It does not crash with sia coin mining but crash with Ether and Zcash.
- The tweaker displayed impossible MHz ranges.
- Every time it crash, fans are like jet engine.

Really appreciate any help or suggestion.
Thank you in advance.

-----

SOLVED.

It was the power, some kind of failure into one of the external connections with the cable that made a really strange failure that only appeared when intense VRam was used. I RMA it and after a week it runs smooth again.

Thanks all for your support.

Start from scratch. Do a clean install of windows 10. if it still does it at that point then you definitely have a hardware issue.

Thanks for answering.

Did it, new install, same issue, but... can not track where is the failure to RMA.

Dont think that 4 gpu fails or 4 risers, but it doesnt seem to be a mobo fail as it crash with gpu's vram.

I suspect power supply.

Thanks, it's a 1000 Corsair gold.

The 3 months it worked I had it with reduced power, all 4 gpu at 600 W.

Did you test 1 gpu at a time without risers in a pci slot to verify correct operation of each gpu?

Hi, yes, fans becomes a jet in a few minutes and blank screen.

It never happends when fans are a 0%, it appears when fans start to work and may take 30 seconds or several minutes.

Thanks.