The new Apple A8 SoC unveiled last night is the most complicated chipset from Apple ever. Architectural improvements lend this dual-core processor about a 20% increase in performance over the Apple A7, which was their first 64-bit chipset introduced last year.

The Apple A8 is actually small, with a die size of 89mm^2. This is despite carrying nearly twice as many transistors as the A7 SoC (~2 Billion vs. 1 Billion). However, thanks to the new 20nm manufacturing process, Apple is able to squeeze a lot more transistors into a tighter space.

Details regarding the GPU weren't revealed last night. For all we know, it could be a more powerful member of the PowerVR family (could be the PowerVR GX6650), which has been serving Apple SoCs since time unknown. What we do know is that the new GPU in the A8 SoC will end up being nearly 50% faster than the GPU in the A7 SoC.

Lastly, Apple claimed that the new A8 SoC is 50% more efficient than its predecessor. At this moment, we have no way of confirming any of Apple's claims, but the results are likely to be somewhere close.