This post doesn't mention the strategic dilemma Intel has with GPUs. GPUs are fast because they have fast local memory, the core count is secondary, and some Intel chips have 22x16 vector processing elements clocked 2x higher than NVIDIA ( around 3k CUDA core equivalent ).
Why doesn't Intel make a fast local memory GPU style device. Well, they did (Xeon Phi) but if they give it a reasonable price point it would cannibalize their existing x86 market, which is more than the GPU market.
Why doesn't Intel make a fast local memory GPU style device. Well, they did (Xeon Phi) but if they give it a reasonable price point it would cannibalize their existing x86 market, which is more than the GPU market.