Revision as of 12:49, 27 March 2025 edit Vincent Lefèvre (talk \| contribs) Extended confirmed users 5,208 edits →Internal representation: For "Half", no need to say "IEEE 754-2008" as the current standard applies (BTW, nothing was said for quad, formally binary128, which has also been introduced in 2008). ← Previous edit		Revision as of 12:52, 27 March 2025 edit undo Vincent Lefèvre (talk \| contribs) Extended confirmed users 5,208 edits →Internal representation: avoid useless redirections Next edit →
Line 193: !Total \|- \|[[Half -precision floating-point format\|Half]] \|1 \|5 Line 202: \|~3.3 \|- \|[[Single -precision floating-point format\|Single]] \|1 \|8 Line 211: \|~7.2 \|- \|[[Double -precision floating-point format\|Double]] \|1 \|11 Line 229: \|~19.2 \|- \|[[~~Quad~~ Quadruple-precision floating-point format\|Quad]] \|1 \|15

Floating-point arithmetic: Difference between revisions