Its not so simple. Depends on the Thermal target for the CPUs.
Microarchitecture that Apple designed at this very moment is more robust, and more complex than anything Intel offers. In single threaded applications, clock for clock Apple CPU will be faster than latest and greatest offering from Intel: "Skylake".
The problem comes with complexity of multithreaded applications, and thermal designs. A10 can have at best 5W TDP, for 2C/2T design. Apple can expand it indefinitely, however, that would require designing particular internal connection between cores, to not get diminishing returns from multi-core CPU design. Also key factor here is that Intel CPUs would scale better with clocks, on higher TDPs.
For Apple, the only way they could use AX chips in their computers is when Intel or AMD is in a situation where they cannot use high-amperage power circuit in their CPU design(they are for example reduced by power governor to reduce the power consumption of the CPUs), and are allowed for example to build only 25W designs, to control the energy grids.
Again: clock for clock, Apple chips are currently faster in Single core performance than anything Intel has. It is more and more apparent that actually shifting from x86 CPU vendors, to Apple custom designed hardware can be extremely beneficial in some cases. So far, it works only in ultra mobile world, where Apple is just years ahead of any competition. Years. But it can, eventually, scale in the world of desktop. Not because Desktop shrinked to ARM hardware. But because the ARM hardware evolved so much that is at least as good as what desktop can offer.