In the context of this test of MSVC vs Cygwin, sure, MSVC is faster. However, in the context of "all its competitors", Intel still holds that title. We use the Intel compilers on Windows/ia32 and Linux/ia32 for our product just for that boost in speed. It's not as dramatic, but in our business, performance is a marketing mantra, so we'll take every little bit we can get.
Intel promises a minimum of 4% speed boost. Of course, I'm not entirely sure what that really means - then again, I'm just a techie, not a marketing guru.
| [reply] [Watch: Dir/Any] |
In compiler technology nowadays, it's really to do with how clever you can arrange your code to take advantage of the CPU caching, super scalar pipelining, etc. I actually believe Intel will have a 4% speed boost because of 'undocumented features'. :-D
| [reply] [Watch: Dir/Any] |