It seems that a lot of GPU problems revolve around specific versions of drivers. Though AMD has their own support structure, you can often learn from information reported by others who fold.
New to F@H, After some wrangling with finding opencl for amd I have had some minor success in getting GPU loads running. For a whopping 5 or so minutes. If I restart the FAHclient GPU load will go up to 100% for around 5mins or so then stop, for good, no spikes later on or anything. Nothing in the log either. Any suggestions?
FYI I'm using GPU-Z watch load on the GPU. And I have power slider at full, while I'm working.
Have you modified any of FAH's settings? When you post the log, we really need the second page from the beginning to understand about your system. Where is the slider set? Have you configured FAH to run on IDLE? (etc.)
Is there enough cooling air to keep your hardware from overheating? Most modern hardware will protect itself by shutting down if it's getting too hot.
I have the same issue. I get a few gpu WU every day, but after a few minutes it stops. The logs show a bad WU error. Next time that happens i will post the log.
I am running a rig with multiple RX580s ... and can confirm, they will overheat after 5~10 minutes without increasing the airflow. The stock fan settings allow things to get too hot before it reacts. You can override your fan settings using a multitude of tools. I'm on Ubuntu, so my solution is certainly different. Find an AMD, or third party, GPU fan controller. Here is one example from AMD: https://www.amd.com/en/support/kb/faq/dh-020
One thing to keep in mind with temperatures on graphics cards: The temperature readings and limits are typically focused on the GPU itself, while the memory may be your actual heat source. So if you are seeing steady 75C GPU temps and think you have ~20C margin, you may have less margin on the memory chips.
Once you have your GPU fan operating a bit better, make sure you have enough air moving through your case. Check your case fans and air filters.
BJMcGee wrote:One thing to keep in mind with temperatures on graphics cards: The temperature readings and limits are typically focused on the GPU itself, while the memory may be your actual heat source. So if you are seeing steady 75C GPU temps and think you have ~20C margin, you may have less margin on the memory chips.
One note about memory and folding, the RAM speed is not as important for the speed of folding calculations on a GPU. So reducing the memory clock will have little impact on folding speed, and can reduce the heat and temperature created.
iMac 2.8 i7 12 GB smp8, Mac Pro 2.8 quad 12 GB smp6
MacBook Pro 2.9 i7 8 GB smp3