@jamilbk
On your print you're still using 1 thread per gpu.
Change your xmr-stak-amd to 8 threads, 2 threads per gpu. something like this:
"gpu_thread_num" : 8,
"gpu_threads_conf" : [
{ "index" : 0, "intensity" : 2016, "worksize" : 8, "affine_to_cpu" : false },
{ "index" : 0, "intensity" : 1800, "worksize" : 8, "affine_to_cpu" : false },
{ "index" : 1, "intensity" : 2016, "worksize" : 8, "affine_to_cpu" : false },
{ "index" : 1, "intensity" : 1800, "worksize" : 8, "affine_to_cpu" : false },
{ "index" : 2, "intensity" : 2016, "worksize" : 8, "affine_to_cpu" : false },
{ "index" : 2, "intensity" : 1800, "worksize" : 8, "affine_to_cpu" : false },
{ "index" : 3, "intensity" : 2016, "worksize" : 8, "affine_to_cpu" : false },
{ "index" : 3, "intensity" : 1800, "worksize" : 8, "affine_to_cpu" : false },
],
Let us know what this change added to the H/s rate
If I'm not mistaken, intensity directly correlates to Vram usage. Since Vega FE has 16GB, he might be able to further raise intensity, or run extra threads as long as he's not core limited.