Fixed-point dsp can do floating-point operations, but the efficiency is too low, and the machine time and memory capacity are often one or two orders of magnitude more than Q format.

The application of DSP is mainly used in the fields that need high-speed operation. Of course, most of the data will be in Q format.

More importantly, the complexity of the program is greatly improved after using the floating-point operation subroutine. Therefore, if it is used in complex multi-process control software, the possibility of ill-considered bugs will be greatly increased.