248 x87 Floating-Point Optimizations Chapter 10
25112 Rev. 3.06 September 2005
Software Optimization Guide for AMD64 Processors
10.7 Avoiding Extended-Precision Data
Optimization
Store floating-point data in single-precision or double-precision format.
Application
This optimization applies to:
• 32-bit software
• 64-bit software
Rationale
Loading and storing extended-precision data is significantly slower than storing single- or double-
precision data.