ΣSigmaStudio Русский

Log2


This module computes the log to base 2 of the incoming data.

There are 2 versions of Log2 module on ADAU144x processors.

  1. Log2 (Linear Interpolation)
  2. Log2 Accurate (Binary Algorithm)

ADAU145x, ADSP-214xx and ADSP-SC5xx/215xx processors support a single version of the Log2 implementation. Log2

Log2 (Linear Interpolation)

This Log2 algorithm computes the integral part of Log2 accurately. Fractional part of the Log is calculated using Linear interpolation. This algorithm takes very less cycles and less accurate.

Input Pins

NameFormat [int/dec] - [control/audio]Function Description
Pin 0: Input Datadecimal - audioInput data to be converted to Log base 2

Output Pins

NameFormat [int/dec] - [control/audio]Function Description
Pin 0: Log valuedecimal - controllog2(input)

Log2 Accurate (Binary Algorithm)

This module uses a binary algorithm to compute the Log2. This is more accurate and takes more cycle. Number of iterations in the algorithm can be configured. Accuracy of the algorithm increases when the number of iterations is increased.

Input Pins

NameFormat [int/dec] - [control/audio]Function Description
Pin 0: Input Datadecimal - audioInput data to be converted to Log base 2

Output Pins

NameFormat [int/dec] - [control/audio]Function Description
Pin 0: Log valuedecimal - controllog2(input)

Configuration

Right Click on the module to configure the number of iterations.

Configuration

Log2 on ADSP-213xx, ADSP-214xx, ADSP-215xx and ADSP-SC5xx processors

Input Pins

NameFormat [int/dec] - [control/audio]Function Description
Pin 0: Input Datadecimal(ADAU145x) - audio
float(ADSP-214xx/ADSP-215xx) - audio
Input data to be converted to Log base 2

Output Pins

NameFormat [int/dec] - [control/audio]Function Description
Pin 0: Log valuedecimal(ADAU145x) - control
float(ADSP-214xx/ADSP-215xx) - control
log2(input)

Grow Algorithm

The module supports Add functionality. Growth is not supported.

GUI Controls

None

DSP Parameter Information

None

Algorithm Description

log2(x) is implemented as (ln(x) / ln(2)) using the change of base formula. ln(x) is implemented using the taylor series expansion.

Supported IC's

1. ADSP-213xx
2. ADSP-214xx
3. ADSP-SC5xx
4. ADSP-215xx