summaryrefslogtreecommitdiffstats
path: root/Doc/library/math.rst
diff options
context:
space:
mode:
authorMark Dickinson <dickinsm@gmail.com>2008-08-01 09:13:07 (GMT)
committerMark Dickinson <dickinsm@gmail.com>2008-08-01 09:13:07 (GMT)
commitdadcd1fd50b02697efcc81cba0bd5b3b5083e69f (patch)
tree65741f30aa32163d8b8dadea357b27255dd01985 /Doc/library/math.rst
parent52597be2a15130ced487240159392a076d1c7cca (diff)
downloadcpython-dadcd1fd50b02697efcc81cba0bd5b3b5083e69f.zip
cpython-dadcd1fd50b02697efcc81cba0bd5b3b5083e69f.tar.gz
cpython-dadcd1fd50b02697efcc81cba0bd5b3b5083e69f.tar.bz2
Tone down math.fsum warning.
Diffstat (limited to 'Doc/library/math.rst')
-rw-r--r--Doc/library/math.rst25
1 files changed, 2 insertions, 23 deletions
diff --git a/Doc/library/math.rst b/Doc/library/math.rst
index 6f8d3c8..be3b70b 100644
--- a/Doc/library/math.rst
+++ b/Doc/library/math.rst
@@ -92,29 +92,8 @@ Number-theoretic and representation functions:
.. note::
- On platforms where arithmetic results are not correctly rounded,
- :func:`fsum` may occasionally produce incorrect results; these
- results should be no less accurate than those from the builtin
- :func:`sum` function, but nevertheless may have arbitrarily
- large relative error.
-
- In particular, this affects some older Intel hardware (for
- example Pentium and earlier x86 processors) that makes use of
- 'extended precision' floating-point registers with 64 bits of
- precision instead of the 53 bits of precision provided by a C
- double. Arithmetic operations using these registers may be
- doubly rounded (rounded first to 64 bits, and then rerounded to
- 53 bits), leading to incorrectly rounded results. To test
- whether your machine is one of those affected, try the following
- at a Python prompt::
-
- >>> 1e16 + 2.9999
- 10000000000000002.0
-
- Machines subject to the double-rounding problem described above
- are likely to print ``10000000000000004.0`` instead of
- ``10000000000000002.0``.
-
+ The accuracy of fsum() may be impaired on builds that use
+ extended precision addition and then double-round the results.
.. versionadded:: 2.6