Now, Fintronic has announced the availability of variable-precision (VP) fixed-point and IEEE 754/854 floating-point objects, functions, and math operators, all in pure Verilog syntax. Fixed-point values can have integer and fractional components of any size, while floating-point values can have exponent and mantissa components of any size. Of particular interest is the fact that the math operators +, -, *, **, and / can be applied to any combination of the following VP and Verilog operands and results formats:

• Variable precision (VP)
  • Arbitrary-precision fixed-point
  • Arbitrary-precision floating-point
• Verilog
  • Integers
  • Reals
  • Registers
  • Supported constants

Furthermore, all trigonometric and hyperbolic (direct and inverse) functions can be called using any format and precision. Also, power, logarithm, and square root operations are available for any format. Accurate rounding and special-condition reporting are supported, while sub-expression width size and format inference can be automatic or user-controlled.

Simulation and language expert Alex Zamfirescu says: "These data types and math functions provide designers with the ability to work at the Matlab level in Verilog." For more information, please contact Dr. Alec Stanculescu at alec@fintronic.com.