Im realy frustrated and i need your help

[stonedgz]

Hello i have assemble my first mining rig

Motherboard: gigabyte H110 D3a
ram :4gb ddr4
Hard disk: 120 ssd
Psu: corsair 750wat
2 rx570 red devils
Windows10

I flashed them and overclocked, they worked fine for a day and then they crashed i restarted the miner and it worked fine for 1more day then they start crashing

I reflashed to ORIGINAL bios,kept on crashing

Im realy frustrated and i dnt know what to do im scared its hardware
Then ill be fucked. And yeah my hdmi keeps on disconnecting...

HELP ME PLZZZZ

[gotminer]

Check the event log.  What kind of errors do you have when the rig is crashing?  What is your power draw?  What is your swap file size?  Upgrade to 8GB ram.

[Set Ready Go]

sounds like its just OC to much hence the crashings but as you said you reflashed with original bios,

If you have any new risers at hand you could try switch the risers out.

Other then that you should disconnect one of the cards and just hook up one of the cards in the x16 slot, no risers.
run it and see if it will crash in 48 hours, if not shit down.

Rince repeat with second card.

If second card crashes well then you probably have to rma that card.

[stonedgz]

Check the event log.  What kind of errors do you have when the rig is crashing?  What is your power draw?  What is your swap file size?  Upgrade to 8GB ram.

most of the errors im getting display driver amdkmdap stopped responding and has successfully recovered 

sometimes im getting this the previous system shutdown at.... was unexpected and its around the time that the rig stopped  kernel-boot when im forcing it to shutdown coz it freeze and i cant do anything

i got this error bugcheck : unable to produce minidump file from the ful dump file i dnt know what is this !!


and this one too distrbutedCOM  the server(bla bla bla ) did not register with DCOM within the required timeout

and service control manager :  the Dhcp service depends of the AFD service which failed to start because of the following error:
a device attached tothe system is not functioning.



DCOM got error "1084" attempting to start the service dps wit arguments "unavailable "in order to run the service ( blah blah blah)

im sorry its too long coz i couldnt figure out what kind of error i might need to know   
my power draw is 110 gpuz card only and its not stable sometimes goes down and up
i couldnt find the swapfile but if oyu mean my virtual memory i set up it to 16gb 

[stonedgz]

sounds like its just OC to much hence the crashings but as you said you reflashed with original bios,

If you have any new risers at hand you could try switch the risers out.

Other then that you should disconnect one of the cards and just hook up one of the cards in the x16 slot, no risers.
run it and see if it will crash in 48 hours, if not shit down.

Rince repeat with second card.

If second card crashes well then you probably have to rma that card.

Thx for the advice ill try if nothing worked now its been an hour mining with 1 card no crashing no nothing 1200,1950  -50
Ill wait till tomorrow and see

[stonedgz]

Check the event log.  What kind of errors do you have when the rig is crashing?  What is your power draw?  What is your swap file size?  Upgrade to 8GB ram.

so i started 1 card and after few hours it freeze again and i got this error from the event log
dwrninit : the desktop windos manager process has exited (process exit code blah blah )restart count 1 primary display device id radeon rx 570 series !!!!

and 1 more error  ATIlerecord


help

[stonedgz]

HELP me

anyone 

[coinut]

Did you test both cards one at a time installed directly into the motherboard?

[stonedgz]

Did you test both cards one at a time installed directly into the motherboard?

you want me to connect 1 of the cards directly into the motherboard and try it ??

im doing that now let see what will happend

[coinut]

yes only one GPU at a time and test it on stock clocks, connected directly to the x16 slot in your mobo

[stonedgz]

yes only one GPU at a time and test it on stock clocks, connected directly to the x16 slot in your mobo

plugged 1 card directly into the motherboard its doing 21.900 mh\s stock setting ill wait and see if its gonna crash

ill wait for a while and update you guys thx for the help

[coinut]

if it crashes, change the GPU with your other card and retest under the same conditions, stock settings use the same pcie connector(s) from your PSU, you want nothing else to change other than changing the GPU.

[coinut]

what is the model of PSU that you have ?

[stonedgz]

what is the model of PSU that you have ?

no crashes untill now but im gonna keep it for few hours, yesterday it stayed stable for 3 hours then it crashed

i have corsair cx750 W

[Undefined31415]

If I had to guess, your issue was probably aggressive overclocking. Do you remember what your settings were?

[yhhy]

Have you checked what memory the cards have (samsung, hynix?), and have the correct bios flash?

[Bitsaurus]

I'm assuming that the "-50" that you put in settings is -50mv voltage to the core? Try getting rid of that. Most cards can handle that but occasionally one card can't and it will cause the whole rig to go down.

Also, you can leave Hardwareinfo64 running and look to see if you're getting memory errors - if you are back off on the memory overclocks.

[stonedgz]

If I had to guess, your issue was probably aggressive overclocking. Do you remember what your settings were?

1200, 1950 -50 mv 60% fan
Hynix memory

[stonedgz]

Have you checked what memory the cards have (samsung, hynix?), and have the correct bios flash?

Yeah i checked and did bios flash and i reflashed it bk to original now the rig mining on stock setting getting 21 3 hours runing no crashes untill now ill wait untill im bk from work and keep you guys updated

[stonedgz]

I'm assuming that the "-50" that you put in settings is -50mv voltage to the core? Try getting rid of that. Most cards can handle that but occasionally one card can't and it will cause the whole rig to go down.

Also, you can leave Hardwareinfo64 running and look to see if you're getting memory errors - if you are back off on the memory overclocks.

Now im running 1 card directly to the motherboard stock setting lets see if its gonna crash...
Question will that happen if i was running 2 gpu??