For the Function Approximation library element, use the list of configurable parameters and default values is presented below.
| Name | Type | Default | Description |
|---|---|---|---|
| DATA_TYPE | typename | cint16 | Data type. |
| COARSE_BITS | unsigned | 8 | Number of bits used for coarse lookup. Determines the number of linear sections in the lookup table. |
| FINE_BITS | unsigned | 7 | Number of bits used for interpolation between lookup sections. |
| DOMAIN_MODE | unsigned | 0 | Describes normalized input, x, domain. |
| FUNC_CHOICE | unsigned | 0 | Sets which utility function is used to create lookup table values. Each utility function provides approximations for a specific function. The following values and functions are enumerated for FUNC_CHOICE: 0 - SQRT_FUNC 1 - INVSQRT_FUNC 2 - LOG_FUNC 3 - EXP_FUNC 4 - INV_FUNC |
| SHIFT | unsigned | 0 | See Common Configuration Parameters. |
| WINDOW_VSIZE | unsigned | 6 | See Common Configuration Parameters. |
| ROUND_MODE | unsigned | 0 | See Common Configuration Parameters. |
| SAT_MODE | unsigned | 1 | See Common Configuration Parameters. |
| NITER | unsigned | 4 | See Common Configuration Parameters. |
| STIM_TYPE | unsigned | 0 | This parameter drives the input data stimulus differently from other library elements, as a different input generation script is used. STIM_TYPE = 0 provides random data in the selected DOMAIN_MODE. STIM_TYPE = 1 provides incrementing data across the entire domain for the selected DOMAIN_MODE. |