Open mrobleno opened 3 years ago
Adding 8GB of ram fixed the issue and now have all 10 GPUs up and running on the one rig. I will try and add two more to max it out at 12 later this week.
Some how after restarting the miner my virtual memory available was decreased. Not sure how this happened but the result was that one of my 10 GPUs would not load the DAG file. I ended up installing a spare harddrive and setting all of that to virtual memory which then allowed me to download the DAG file.
I am currently running 10x RX580s on a B250 with 8GB ram w/ two 1000w power supplies. (But with issues)
I can run 6x RX580 8gb with no issues. Once I go over 7 GPUs i seem to start running into issues. Right now I have 10x RX580s installed on my one rig. Two of those GPUs are having issues when it comes to mining. Under the device manager you can see all 10x GPUs and showing no issues. Load up Adrenaline and all boards are fine. If I make no overclock changes the boards, will start up with the miner fine, but will overheat quickly. I can only make changes to specific GPUs prior to loading the miner or I seem to consistently crash the whole system and need to restart. To stop the overheating, I make overclock changes to the boards via Adrenaline after the miner has loaded up and completed the autotuning. I know which GPUs to now make the overclocks to and start the miner. I now consistently get 9 out of 10 GPUs working and 1 of the 9 GPUs is only doing ~1-2 MH to the pool but showing 30.5 MH otherwise. The dead GPU is showing DAG errors or is stating that its DEAD during start up. I'm running 8GB of RAM and ordered up another 8GB to see if that will help. I have the virtual memory up to 72gb to assist in that. I would love to get some insight on how to fix these issues as I seem to have tried a ton of things and watched just about everything out there to try and fix it.