[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
[lmi-commits] [lmi] master c50cc9b1: Fix defect introduced 20220412T1520
From: |
Greg Chicares |
Subject: |
[lmi-commits] [lmi] master c50cc9b1: Fix defect introduced 20220412T1520Z: wrong equation |
Date: |
Thu, 21 Apr 2022 13:47:12 -0400 (EDT) |
branch: master
commit c50cc9b13df1bcd830e32f1191b68f0a29ddf5f9
Author: Gregory W. Chicares <gchicares@sbcglobal.net>
Commit: Gregory W. Chicares <gchicares@sbcglobal.net>
Fix defect introduced 20220412T1520Z: wrong equation
Commit a6ecf8916528f of 20220412T1520Z incorrectly gave the (imaginary)
roots of f'(x) where the (nonexistent) roots of f(x) are wanted. Thanks
to Vadim Zeitlin for finding the mistake.
---
toms_748.html | 15 +++++++--------
1 file changed, 7 insertions(+), 8 deletions(-)
diff --git a/toms_748.html b/toms_748.html
index 6ddd4b67..6797642f 100644
--- a/toms_748.html
+++ b/toms_748.html
@@ -266,15 +266,14 @@ all algorithms give correct solutions.
</blockquote>
<p>
-According to
-<a
href="https://www.wolframalpha.com/input?i=derivative+of+xe^(-x^-2)">Wolfram</a>,
-the roots of
+The subexpression
<i>xe<sup><small>−x<sup><small>−2</small></sup></small></sup></i>
-are ±<i>i</i>⋅√2; but
-the actual problem defines the value as zero at <code>x=0</code>, so
-the correct solution is <code>x=0</code>. Due to inevitable
-underflow, no iterative algorithm that uses customary
-floating-point numbers can find that solution unless by
+has
+<a href="https://www.wolframalpha.com/input?i=xe^(-x^-2)">no roots</a>,
+but the actual problem defines the function as zero at <code>x=0</code>,
+so the correct solution is <code>x=0</code>. Due to inevitable
+underflow, no iterative algorithm that uses customary floating-point
+numbers can find that solution unless by
chance<a href="#fn4" name="fr4" title="">[4]</a>.
</p>
[Prev in Thread] |
Current Thread |
[Next in Thread] |
- [lmi-commits] [lmi] master c50cc9b1: Fix defect introduced 20220412T1520Z: wrong equation,
Greg Chicares <=