Revision as of 14:38, 18 August 2020 edit Steel1943 (talk \| contribs) Autopatrolled, Extended confirmed users, Page movers, Pending changes reviewers, Rollbackers, Template editors 228,888 edits Update ← Previous edit		Revision as of 11:32, 20 September 2020 edit undo Guy Macon (talk \| contribs) Extended confirmed users, File movers, New page reviewers, Pending changes reviewers, Rollbackers 59,299 edits →Fastfoat16?: new section Next edit →
Line 75: ::::::The effect on the algorithms is various. Some algorithms (such as Malcolm's algorithm) are actually based on the rounding errors in order to work correctly. There is no short answer. Correct rounding is nowadays required in implementations of the FP basic operations; as long as this requirement is followed, the implementer has the choice of the hardware design. Cancellation is just the effect of subtracting two numbers that are close to each other; in this case, the subtraction operation itself is exact (assuming the same precision for all variables), and the normalization does not introduce any error. [[User:Vincent Lefèvre\|Vincent Lefèvre]] ([[User talk:Vincent Lefèvre\|talk]]) 20:13, 16 August 2019 (UTC) == Fastfoat16? == https://www.analog.com/media/en/technical-documentation/application-notes/EE.185.Rev.4.08.07.pdf Is this a separate floating point format or another name for an existing format? --[[User:Guy Macon\|Guy Macon]] ([[User talk:Guy Macon\|talk]]) 11:32, 20 September 2020 (UTC)

Talk:Floating-point arithmetic: Difference between revisions