So.. i am back with a new Ethereum Minig Rig tutorial, this time i will be creating with the new GPU Sapphire Nitro+ Radeon RX 480 8gb OC. YAY! and also with a bit different frame.. finally. I this post i will have less content and more pictures, if you really need a step by step tutorial with good explanation on how to create your first rig check here.
As for the frame i will only add a gallery and you will have to follow the steps, measurements will be found below. More detailed tutorial for building the mining frame can be found here.
Ethereum Mining Rig Frame / Case V2.
41cm (16inch) wide
60cm (23.5inch) long
30cm (11inch) tall
Unbox the motherboard carefully and keep the red material under it. The remove the black plastic protection from CPU slot.
Installing the CPU, Cooler & RAM
Once unboxed, get the cooler and remove thermal paste from it. Install the CPU and then apply thermal paste to it, final step is to install the cooler.
Frame, PSU and Motherboard
Now that we are all set with the motherboard and the frame let’s unbox the PSU. Then place the Motherboard in the center of the frame and the PSU in the left back corner.
Prepare the following cables:
24 PIN ATX
And also do not forget about the keyboard 🙂
Once you have all the cables, begin to install them in PSU and motherboard.
(in motherboard you MUST have 1x Power Switch cable, 1x CPU cable, 1x 24 PIN ATX cable and one MOLEX cable that will give more power to PCI-e slots CHECK THIS IMAGE )
I highly recommend to install the GPUs one by one and test them, follow the next steps to make sure that everything will work fine.
DO NOT install the SSD/ethOS yet
Plug in the PSU, make sure you also plug in the watt usage monitor (if you have one)
Install 1 GPU ONLY , turn on the PSU, power up the rig and make sure that it is booting correctly.
Now install the SSD and turn on the rig (video tutorial: how to write ethOS image on SSD), wait for full boot and an error with WRONG DRIVER text will be displayed in terminal will be displayed (like in this image). You will have to access the local.conf file, type in terminal “nano local.conf” and below “globalfan 85” add “globaldriver amdgpu”, save the file by pressing CTR+O and them hit ENTER,exit with CTRL + X and reboot. Now the mining should start with no problem for your 1 GPU (23.xx mh/s)
Success boot and mining for the 1st gpu? Great! Now install the second GPU and follow the same steps till you have all 6 GPU installed and working (in this way is very easy to find if a USB riser or Molex is not functioning correctly or the video card has a problem)
Rig is not booting with one GPU ?
Double check cables (from psu and motherboard)
Make sure CPU cooler is good placed and is not moving
Make sure that the RAM fits good in the slot, if still doesn’t boot change the slot
Check the jumpers for power switch if this is placed correctly
ethOS is not booting ?
Black screen? Shutdown and start it again
Is booting but the mining process is not starting? Shutdown and start it again
All GPU installed but one keeps crashing or one GPU is always crashing, ethOS version is 1.1.1? No worries, new version will be released (1.1.2) and that will fix this problems, also more features for RX 4xx series will be available)
Mining process not starting at all ? type “helpme” in terminal, copy the generated link and send it to mirc chat #ethereumOS
Everything is running perfect!? Great! Let’s configure ethOS
I had no problems, everything went nice a smooth from the beginning to the end, no riser/molex/sata/etc problem. Only that a card keeps crashing but like i said this will be fixed with the new ethOS release 1.1.2. I am too lazy to make this rig on windows, ethOS is much better, simple setup, easy configuration etc no headache on finding the right clocks and mems.. The screenshot of Watt usage meter is for 5 GPUs only, i forgot to take a photo with all 6 :/ but is somewhere around 900w.
OK, so we are all good, first of all we will need to configure ethos. Follow this steps
Update ethOS to the latest version
This can be done by typing the command sudo ethos-update && sleep 5 && sudo reboot in the black window. Now hit the enter button and do not interrupt the process of update.
Set ethOS to the single rig mode
Run this command to make the remote file completely blank echo -n "" > /home/ethos/remote.conf
Changing ethOS password
a) In terminal type passwd and you will see “Changing password for ethos.”
b) Type default/current password which is “live”
c) Then type your preferred password twice and that’s it, we secured the rig.
Run the command nano local.conf and the default settings will be listed like in the image below.
maxgputemp 85 leave it default for now. This number means that this is the maximum temperature your GPU should reach when the system automatically stops the miner.
stratumproxy enabled this is a default command and should not modify it.
Using arrows go down to proxywallet and change the address with yours.
On proxypool you will see it in 2 lines, proxypool1 and proxypool2, it is because the second link is as a backup and if proxypool1 has connection problems or is down for any other reason then ethOS will connect to proxywallet2. If you are located in EU change proxywallet1 to this one: eu1.ethermine.org:4444
Save new modifications by pressing CTRL+O then ENTER. CTRL+X to exit local.conf file.
Here is my configuration file
update: –cl-global-work 8192 should be changed to –cl-global-work 16384
Here it is, up an running. Currently i am waiting for the new ethOS version 1.1.2 (now using 1.1.1). The problem now is that i can only run with sotck settings, so no overclock till the 1.1.2 version with more support for rx 4xx series.
Hash power (stock settings): 140 mh/s
Power consumption (no overclock): 915w
Max GPU temp (near to open window, 40%-50% global fan): 55-60 °C
I will continue to update this post when the new ethOS version released, with overclock, hashing power, power consumption and much more, don’t forget to subscribe.
To receive the latest updates follow me on social media! 🙂
So now we are able to overclock our cards, here is what i did.
I am having serious problems, one card keeps crashing, not sure why but i am still testing and checking to find the best core mem for this, so if one or two cards keep failing try to change usb risers recheck all the cables, maybe v 1.1.2 is not that stable yet but we can do a lot more things than previous version so maybe we can find the best core mem for our cards (for now).
First of all try to add “globalmem 2150” in local. conf (type “nano local.conf” in terminal). Core settings should remain default (do not add them). Leave it to run like for about 1-2 hours and if everything goes good increase the mem to “2225” and leave it like this for a while (significant mh/s will be visible).
Thank you for reading. As always, your comments, suggestions and questions are welcome.