Among integrated circuits, field programmable gate arrays (FPGAs) may be the most spectacular benefactors of the steady progress of very large scale integration (VLSI) technology in the last two decades: current FPGA chips offer up to one million programmable gates. A close look at recent families of mainstream FPGAs, however, shows that the amount of silicon dedicated to routing increases much faster than that devoted to the gates themselves. The main reason for this is the fact that the routing needs to be reconfigurable. A simple quantitative analysis shows that this trend, if pursued, will lead to a widening gap between FPGA performance and full-cust- om VLSI performance. Some prospective solutions to this problem are exposed and discussed.