|
|
Article: Fixed or floating? A pointed question in DSPs. (digital signal processing)
- Article from:
- EDN
- Article date:
- August 3, 1995
- Author:
CopyrightCOPYRIGHT 1995 Reed Business Information, Inc. (US). This material is published under license from the publisher through the Gale Group, Farmington Hills, Michigan. All inquiries regarding rights should be directed to the Gale Group. (Hide copyright information)
|
Designers considering DSP applications must first decide whether to commit their designs to fixed- or floating-point DSPs. The answer may seem to be a trade-off between cost and performance. However, designers must carefully consider a variety of factors before choosing a DSP.
The most fundamental difference between fixed- and floating-point DSPs is the numeric format. The hardware of fixed-point DSPs uses integer arithmetic. Floating-point DSPs perform integer, or real, arithmetic. Normally, floating-point DSP formats are 32 data bits wide, in which 24 bits form the mantissa and 8 bits make up the exponent. For example, the chip would store the decimal value 194 ...