Talk:Single precision

From Wikipedia, the free encyclopedia

(empty)

[edit] Exponent/mantissa ranges are inconsistent with other sources

The C++ standard library, as an example, will report the max and min exponents as (-125,+128) and assume that the mantissa is in the range (0.5,1). This article assumes that the exponent is between (-126,127) and that the mantissa is in the range (1,2). Although either method gets the same answer, the one given in the article does not match standard usage. The article should make this clear. —Preceding unsigned comment added by 71.104.122.200 (talk) 03:12, 13 February 2008 (UTC)

The values in the article are those given in the IEEE 754 standard (1985), which the article calls out in the first paragraph. That is the standard for floating-point arithmetic. What clarification(s) would you like to see? mfc (talk) 20:36, 22 February 2008 (UTC)