Is there any significant performance difference between Parallel Studio XE and gcc?
If there is a significant difference is it possible to get a Parallel Studio XE license like the http://dap.xeonphi.com/ when i got the computer through a different vendor?
It would be nice to be able to point you at SPEC results where people have run gcc and icc on identical hardware. However, there are very few cases where people have done that. It seems that almost everyone uses Intel compilers when submitting SPEC results on their hardware. Of course, that may tell you all you need to know!
https://software.intel.com/en-us/qualify-for-free-software explains all the ways you can qualify for free Intel software.
While I fully agree with James' remark about seeing a SPEC result comparison between ICC and gcc , I have to add that with the latest GCC version the difference in performance between ICC and gcc is mostly minimal. Of course, there will be examples where ICC outperforms gcc , but on average the code generated by GCC 6 is quite well-tuned for the CPU type specified.
What the Intel Parallel Studio does give you is a whole suite of nice debugging and profiling tools, such as Vector Adviser, VTune etc etc: I have yet to find an open source alternative that comes close to these tools.