Surely this can be attributed to Blender's HIP code just being suboptimal because nobody really cares about it. By extension nobody cares about it because performance is suboptimal.
It's AMDs job to break that circle.
Surely this can be attributed to Blender's HIP code just being suboptimal because nobody really cares about it. By extension nobody cares about it because performance is suboptimal.
It's AMDs job to break that circle.