Microsoft KB Archive/30601

= Real-Number Data in IEEE Format =

Article ID: 30601

Article Last Modified on 11/17/2003

-

APPLIES TO


 * Microsoft Macro Assembler 5.1 Standard Edition
 * Microsoft Macro Assembler 6.0 Standard Edition

-



This article was previously published under Q30601



SUMMARY
By default, the Microsoft Macro Assembler Versions 5.1 (MASM) and later assemble real-number data in IEEE format. The previous versions of the assemblers used the Microsoft binary format by default.



MORE INFORMATION
The following is the IEEE format for encoding 4- and 8-byte real numbers:   Short Real Number 31     23      15      7       0    _________________________________   |_|_______|_______|_______|_______|   ^ ^       ^                       ^   | |       |                       |   | Exponent|                       | |        |Mantissa ->| |Sign The parts of the real numbers are described below:
 * 1) Sign (0 for positive or 1 for negative) is in the upper bit of the first byte.
 * 2) Exponent is in the next bits in sequence (8 bits for a short real number or 11 bits for a long real number).
 * 3) All except the first-set bit of Mantissa are in the remaining bits of the variable (13 bits for short real numbers, and 52 bits for long real numbers). Because the first significant bit is known to be set, it does not need to be stored.

Additional query words: kbinf 5.10 6.00

Keywords: KB30601

-

[mailto:TECHNET@MICROSOFT.COM Send feedback to Microsoft]

© Microsoft Corporation. All rights reserved.