Still is a CPU benchmark as long as the rest of the components are the same.
And since overclocking the CPU can make a difference for Intel Core 2nd & 3rd generation processors with a single GTX 680, with the Titan or an SLI/CF config ( I hate them, but since people use those, putting those to the test as well would be nice ) the gap could be wider.
I have seen reviews that could be used to compare a GTX 680 with an FX-8150 and a 2600K, but can't be sure about the results ( validity, consistency, testing routines, configs, etc ).
I'd really like to do it myself, but can't be arsed to spend more money on tech, especially one I don't need and has nothing to offer me at all for the time being at least ( perhaps somebody might lend me an AMD FX platform to do those tests, we'll see ).