LLVM 2.9 Clang Performance On Intel's Sandy Bridge

Written by Michael Larabel in Software on 18 February 2011 at 04:12 AM EST. Page 3 of 7. 9 Comments.

Switching over to C-Ray, the favorite multi-threaded ray-tracing test, the LLVM Clang 2.9 SVN code was the fastest over any GCC solution but by a slim margin.

The single-threaded POV-Ray 3.6.1 ray-tracing benchmark was faster on LLVM Clang than the stock GCC 4.5.2/4.6.0, but the corei7-avx optimized GCC 4.6 development snapshot was even faster than the Low-Level Virtual Machine output.

LLVM Clang 2.9 SVN 2011-02-04 had lost out to all of the GCC compilers when looking at the Himeno performance under the Core i7 Sandy Bridge CPU.


Related Articles