Descripción

This mode avoids traps by clamping floating point value to a reasonable range when performing arithmetic operations.