Despite Moore's law stating that the speed of processors doubles every 18 months, in fact, engineers are currently working with inadequate tools. For example the lithographic process for the "writing" of the circuits inside the chips uses beams of light too large to allow reaching the level of miniaturization required by a 20 GHz chip. An alternative would be the EUV (Extreme Ultraviolet) rays , a technology still under development. A second factor that causes problems with energy consumption, which in turn leads to disturbance signals. A solution could be the use of aluminum oxide, tantalum or titanium instead of silicon oxide. But does anyone really need a 20 GHz processor? Maybe not for the moment, but such fast processors could open new perspectives to the world of computers, for example processes of image recognition or "natural" dictation systems able to translate texts in dozens of different languages in real time.