I haven't tried MICs personally, but AFAIK you need vectorization to match current Pascal generation GPU performance with Knight's landing -> which is where my

I haven't tried MICs personally, but AFAIK you need vectorization to match current Pascal generation GPU performance with Knight's landing -> which is where my comment applies. I don't doubt that you can get good speedups when this applies to your code already, but if you start from naive CPU code you'll have a lot of work to do, which IMO is similar to the work needed to port to GPGPU.