The Fermat Number Transform (FNT) provides error-free computation of convolution and correlation, as well as low implementation cost. However, despite having been known for more than two decades, the FNT has not gained widespread acceptance by industry. It is the aim of this contribution to demonstrate that with the present maturity of ASIC technology and logic-synthesis tools the FNT can be a cost-effective alternative for a large class of practical filtering and correlation tasks. In the sequel we report on the design of a macro-pipelined ASIC for computation of the FNT and give figures for area expenditure for a selection of structures for FNT-based convolvers and correlators.