...
NUM05-EX0: Denormalized numbers are acceptable when competent numerical analysis demonstrates that the computed values will meet all accuracy and behavioral requirements that are appropriate to the application. Note that "competent numerical analysis" generally requires a specialized professional numerical analyst; lesser levels of rigor fail to qualify for this exception.
Risk Assessment
Floating-point numbers are an approximation; denormalized floating-point numbers are a less precise approximation. Use of denormalized numbers can cause unexpected loss of precision, possibly leading to incorrect or unexpected results. Although the severity stated below for violations of this rule is low, applications that require accurate results should consider the severity of this violation to be high.
...
<ac:structured-macro ac:name="unmigrated-wiki-markup" ac:schema-version="1" ac:macro-id="6f8443d893a0389d-0b6c702a-426a49c0-914aa8dd-4504e87c68acbaffcda0b783"><ac:plain-text-body><![CDATA[ | [[IEEE 754 | AA. Bibliography#IEEE 754 2006]] |
| ]]></ac:plain-text-body></ac:structured-macro> |
<ac:structured-macro ac:name="unmigrated-wiki-markup" ac:schema-version="1" ac:macro-id="c7194e32d64e34a1-9ac1bec3-4d444ed2-8b778b4d-6d0fc140b7dfff3bc47a0410"><ac:plain-text-body><![CDATA[ | [[Bryant 2003 | AA. Bibliography#Bryant 03]] | Computer Systems: A Programmer's Perspective. Section 2.4 Floating Point | ]]></ac:plain-text-body></ac:structured-macro> |
...