Patent · US Expired

Mechanism for floating point to integer conversion with RGB bias multiply

US6061782A · kind A · utility

18Cited by
4References
32Claims
0Family size

Assignee

Inventors

Key dates

Filing dateMar 26, 1998
Grant dateMay 9, 2000
Priority date
Expiry dateMar 26, 2018

Classification

  • Technology area (CPC G)Physics
  • CPC primaryG06F9/30025
  • WIPO fieldComputer technology
  • WIPO sectorElectrical engineering

Abstract

An apparatus and method for performing a floating point multiply (by a fixed graphics constant), and converting the product of the multiply into an integer, within a single operation is provided. The apparatus includes detection logic to determine whether a special multiply/convert operation is specified, and if so, floating point conversion logic to adjust a bias constant prior to conversion of the floating point number to an integer. More specifically, if the multiply/convert operation specified relates to calculation of graphic points for display, execution of the multiply convert operation effectively multiplies a specified floating point number by a graphics constant, as part of subtracting the exponent bias.

Source: USPTO / EPO open patent data. Objective bibliographic and citation counts.