: ‘State-of-the-art in heterogeneous computing’. Andre R. Brodtkorb, Christopher Dyken, Trond R. Hagen, Jon M. Hjelmervik and Olaf O. Storaasli. Scientific Programming 18 (2010) 1–33
Depending on the application, FPGAs can use approximately 1/20th the power of a microprocessor to do the same task and 1/3 the power of a GPU for the same computing horsepower.
Arrayware applies FPGAs to achieve significant power reduction resulting in longer battery life.
Running temperature can also be reduced which can assist with reliability. It also can lead to smaller, lighter housings and heatsinks.