DSP Builder for Intel® FPGAs (Advanced Blockset): Handbook

ID 683337
Date 7/15/2024
Public
Document Table of Contents

10.1. DSP Builder Floating-Point Data Type Features

  • Variable precision. Nine floating-point precisions including half-precision, single-precision, and double-precision data.
  • High throughput. Architecture adapts to fMAX.
  • Bit-accurate simulation.
  • Efficient DSP block usage in all device families .
  • Standards compliant. Fundamental components (add, sub, mult, and div) are IEEE compliant. Elementary mathematical functions are OpenCL compliant.
  • Configurable. Trade off logic utilization against accuracy. Significant DSP block reduction using faithful rounding.