AMD AOCC 1.1 Shows Compiler Improvements vs. GCC vs. Clang
Traditionally the Clang compiler has been much slower than recent releases of GCC when running the C-Ray multi-threaded ray-tracer. Thus it's exciting to see now that thanks to AMD's optimizations, AOCC 1.1 is running between GCC 7.2 and GCC 8.0.1 performance levels.
Running some benchmarks of the Bullet Physics Engine were mixed, but AOCC 1.1 did have strong showings in a number of the sub-tests.