557277 - Number.MIN_VALUE is looser than in ES3 due to denormal issues

Status: RESOLVED WONTFIX

(Tamarin Graveyard :: Documentation, defect, P4)

Description

[Lars T Hansen]

See bug #555805.  IEEE arithmetic or no, we don't want to state unequivocally that MIN_VALUE is the smallest IEEE 64-bit denormal, rather we want to state what Stan suggests:

"I think we want the following two things to be true:

  Min value is not zero:  Number.MIN_VALUE > 0

and

  All smaller values (eg MIN_VALUE/2) are zero: Number.MIN_VALUE/2 == 0"

Currently our on-line documentation states this, which is too specific:

"The smallest representable non-negative, non-zero, number (double-precision IEEE-754). This number is approximately 5e-324."

Comment 1

[Stan Switzer]

How about something like "The smallest representable non-negative, (double-precision
IEEE-754) number that does not compare equal to zero. This number usually is approximately 5e-324 but can be larger on systems that do not support denormalized numbers."

Comment 2

[Sylvestre Ledru [:Sylvestre]]

Tamarin is a dead project now. Mass WONTFIX.

Comment 3

[Sylvestre Ledru [:Sylvestre]]

Tamarin isn't maintained anymore. WONTFIX remaining bugs.