Hi Marco, on Windows I'm using Visual C++ 2013 and on Mac clang/Apple LLVM version 7.3.0 . As far as I know neither has any special treatment for 6th and 7th generation Intel i7 CPUs, or at least not that I would be able to use without having a special version that would only run just on those CPUs and not on anything else. It would require a lot of extra work to have a special version of MoI for every different kind of CPU so targeting specific CPU models is not really practical for me.

I have seen that just the general switch from x87 floating point code to x64 SSE floating point code seems to give a speed boost but I don't have any numbers on that handy to give to you. Really I'm not sure if the improvement is from using SSE floating point or if it's the additional registers available. It is actually possible in some cases for conversion to 64-bit to actually reduce performance due to doubling the size of data that's being moved around but it doesn't seem like that is going to be the case here. The only thing I'm a little worried about is the 64-bit QtWebKit doesn't have as high of a JavaScript JIT engine enabled for it, I'm not sure yet if that will impact script performance very much or not.

- Michael