https://svn.boost.org/trac10/ticket/12199 <p> If the difference of exponents of two cpp_bin_float values is exactly equal to # of bits in significand then result of addition is not rounded correctly. </p> <p> Platform: Mingw, gcc 5.3.0, x64. All optimization levels. Library version: both 1.60.0 and the newest files from github </p> en-us /htdocs/site/boost.png https://svn.boost.org/trac10/ticket/12199 Trac 1.4.3 Michael Shatz Thu, 12 May 2016 23:51:40 GMT https://svn.boost.org/trac10/ticket/12199 https://svn.boost.org/trac10/ticket/12199 <ul> <li><strong>attachment</strong> → <span class="trac-field-new">add_test.cpp</span> </li> </ul> Ticket John Maddock Mon, 16 May 2016 11:45:28 GMT https://svn.boost.org/trac10/ticket/12199#comment:1 https://svn.boost.org/trac10/ticket/12199#comment:1 <ul> <li><strong>status</strong> <span class="trac-field-old">new</span> → <span class="trac-field-new">closed</span> </li> <li><strong>resolution</strong> → <span class="trac-field-new">fixed</span> </li> </ul> <p> Reproduced in master, but not current develop - suspect it was fixed in previous commits for the other issues. Added some better tests for this: <a class="ext-link" href="https://github.com/boostorg/multiprecision/commit/0397d8bcd8d9139737b22c98880c85e19c821b99"><span class="icon">​</span>https://github.com/boostorg/multiprecision/commit/0397d8bcd8d9139737b22c98880c85e19c821b99</a> </p> Ticket