I don't see why a RISC-V core can't be built to run as fast or as efficiently as an x86 on the high end. With the C extension for compressed (16-bit) instructions, the dynamic and static code size is very close to x86 and usually better than armv8 over a slew of practical benchmarks.

https://riscv.org/wp-content/uploads/2016/07/Tue1130celio-fu...