realmin

Smallest positive normalized fixed-point value or quantized number

Syntax

x = realmin(a)

x = realmin(q)

Description

x = realmin(a) is the smallest positive normalized real-world value that can be represented in the data type of fi object a. Anything smaller than x underflows or is an IEEE^® “denormal” number.

x = realmin(q) is the smallest positive normal quantized number where q is a quantizer object. Anything smaller than x underflows or is an IEEE “denormal” number.

Examples

Smallest Real-World Value of fi Object

a = fi(0,true,6,3);
realmin(a)

ans = 

    0.1250

          DataTypeMode: Fixed-point: binary point scaling
            Signedness: Signed
            WordLength: 6
        FractionLength: 3

Smallest Quantized Number of quantizer Object

q = quantizer('float',[6 3]);
x = realmin(q)

x =

    0.2500

Input Arguments

a — Input array

scalar | vector | matrix | multidimensional array

Input array, specified as a fi scalar, vector, matrix, or multidimensional array.

Data Types: fi

q — Data type properties to use for quantization

quantizer object

Data type properties to use for quantization, specified as a quantizer object.

Algorithms

• If q is a floating-point quantizer object, $$x=2^{E_{min}}$$ where $$E_{min}=\mathrm{exponentmin}(q)$$ is the minimum exponent.

• If q is a signed or unsigned fixed-point quantizer object, $$x=2^{-f}=\epsilon$$ where f is the fraction length.

Extended Capabilities

C/C++ Code Generation
Generate C and C++ code using MATLAB® Coder™.

HDL Code Generation
Generate VHDL, Verilog and SystemVerilog code for FPGA and ASIC designs using HDL Coder™.

Version History

Introduced before R2006a