]> git.proxmox.com Git - ceph.git/blob - ceph/src/boost/libs/math/doc/html/math_toolkit/bessel/bessel_first.html
bump version to 12.2.2-pve1
[ceph.git] / ceph / src / boost / libs / math / doc / html / math_toolkit / bessel / bessel_first.html
1 <html>
2 <head>
3 <meta http-equiv="Content-Type" content="text/html; charset=US-ASCII">
4 <title>Bessel Functions of the First and Second Kinds</title>
5 <link rel="stylesheet" href="../../math.css" type="text/css">
6 <meta name="generator" content="DocBook XSL Stylesheets V1.77.1">
7 <link rel="home" href="../../index.html" title="Math Toolkit 2.5.1">
8 <link rel="up" href="../bessel.html" title="Bessel Functions">
9 <link rel="prev" href="bessel_over.html" title="Bessel Function Overview">
10 <link rel="next" href="bessel_root.html" title="Finding Zeros of Bessel Functions of the First and Second Kinds">
11 </head>
12 <body bgcolor="white" text="black" link="#0000FF" vlink="#840084" alink="#0000FF">
13 <table cellpadding="2" width="100%"><tr>
14 <td valign="top"><img alt="Boost C++ Libraries" width="277" height="86" src="../../../../../../boost.png"></td>
15 <td align="center"><a href="../../../../../../index.html">Home</a></td>
16 <td align="center"><a href="../../../../../../libs/libraries.htm">Libraries</a></td>
17 <td align="center"><a href="http://www.boost.org/users/people.html">People</a></td>
18 <td align="center"><a href="http://www.boost.org/users/faq.html">FAQ</a></td>
19 <td align="center"><a href="../../../../../../more/index.htm">More</a></td>
20 </tr></table>
21 <hr>
22 <div class="spirit-nav">
23 <a accesskey="p" href="bessel_over.html"><img src="../../../../../../doc/src/images/prev.png" alt="Prev"></a><a accesskey="u" href="../bessel.html"><img src="../../../../../../doc/src/images/up.png" alt="Up"></a><a accesskey="h" href="../../index.html"><img src="../../../../../../doc/src/images/home.png" alt="Home"></a><a accesskey="n" href="bessel_root.html"><img src="../../../../../../doc/src/images/next.png" alt="Next"></a>
24 </div>
25 <div class="section">
26 <div class="titlepage"><div><div><h3 class="title">
27 <a name="math_toolkit.bessel.bessel_first"></a><a class="link" href="bessel_first.html" title="Bessel Functions of the First and Second Kinds">Bessel Functions of
28 the First and Second Kinds</a>
29 </h3></div></div></div>
30 <h5>
31 <a name="math_toolkit.bessel.bessel_first.h0"></a>
32 <span class="phrase"><a name="math_toolkit.bessel.bessel_first.synopsis"></a></span><a class="link" href="bessel_first.html#math_toolkit.bessel.bessel_first.synopsis">Synopsis</a>
33 </h5>
34 <p>
35 <code class="computeroutput"><span class="preprocessor">#include</span> <span class="special">&lt;</span><span class="identifier">boost</span><span class="special">/</span><span class="identifier">math</span><span class="special">/</span><span class="identifier">special_functions</span><span class="special">/</span><span class="identifier">bessel</span><span class="special">.</span><span class="identifier">hpp</span><span class="special">&gt;</span></code>
36 </p>
37 <pre class="programlisting"><span class="keyword">template</span> <span class="special">&lt;</span><span class="keyword">class</span> <span class="identifier">T1</span><span class="special">,</span> <span class="keyword">class</span> <span class="identifier">T2</span><span class="special">&gt;</span>
38 <a class="link" href="../result_type.html" title="Calculation of the Type of the Result"><span class="emphasis"><em>calculated-result-type</em></span></a> <span class="identifier">cyl_bessel_j</span><span class="special">(</span><span class="identifier">T1</span> <span class="identifier">v</span><span class="special">,</span> <span class="identifier">T2</span> <span class="identifier">x</span><span class="special">);</span>
39
40 <span class="keyword">template</span> <span class="special">&lt;</span><span class="keyword">class</span> <span class="identifier">T1</span><span class="special">,</span> <span class="keyword">class</span> <span class="identifier">T2</span><span class="special">,</span> <span class="keyword">class</span> <a class="link" href="../../policy.html" title="Chapter&#160;15.&#160;Policies: Controlling Precision, Error Handling etc">Policy</a><span class="special">&gt;</span>
41 <a class="link" href="../result_type.html" title="Calculation of the Type of the Result"><span class="emphasis"><em>calculated-result-type</em></span></a> <span class="identifier">cyl_bessel_j</span><span class="special">(</span><span class="identifier">T1</span> <span class="identifier">v</span><span class="special">,</span> <span class="identifier">T2</span> <span class="identifier">x</span><span class="special">,</span> <span class="keyword">const</span> <a class="link" href="../../policy.html" title="Chapter&#160;15.&#160;Policies: Controlling Precision, Error Handling etc">Policy</a><span class="special">&amp;);</span>
42
43 <span class="keyword">template</span> <span class="special">&lt;</span><span class="keyword">class</span> <span class="identifier">T1</span><span class="special">,</span> <span class="keyword">class</span> <span class="identifier">T2</span><span class="special">&gt;</span>
44 <a class="link" href="../result_type.html" title="Calculation of the Type of the Result"><span class="emphasis"><em>calculated-result-type</em></span></a> <span class="identifier">cyl_neumann</span><span class="special">(</span><span class="identifier">T1</span> <span class="identifier">v</span><span class="special">,</span> <span class="identifier">T2</span> <span class="identifier">x</span><span class="special">);</span>
45
46 <span class="keyword">template</span> <span class="special">&lt;</span><span class="keyword">class</span> <span class="identifier">T1</span><span class="special">,</span> <span class="keyword">class</span> <span class="identifier">T2</span><span class="special">,</span> <span class="keyword">class</span> <a class="link" href="../../policy.html" title="Chapter&#160;15.&#160;Policies: Controlling Precision, Error Handling etc">Policy</a><span class="special">&gt;</span>
47 <a class="link" href="../result_type.html" title="Calculation of the Type of the Result"><span class="emphasis"><em>calculated-result-type</em></span></a> <span class="identifier">cyl_neumann</span><span class="special">(</span><span class="identifier">T1</span> <span class="identifier">v</span><span class="special">,</span> <span class="identifier">T2</span> <span class="identifier">x</span><span class="special">,</span> <span class="keyword">const</span> <a class="link" href="../../policy.html" title="Chapter&#160;15.&#160;Policies: Controlling Precision, Error Handling etc">Policy</a><span class="special">&amp;);</span>
48 </pre>
49 <h5>
50 <a name="math_toolkit.bessel.bessel_first.h1"></a>
51 <span class="phrase"><a name="math_toolkit.bessel.bessel_first.description"></a></span><a class="link" href="bessel_first.html#math_toolkit.bessel.bessel_first.description">Description</a>
52 </h5>
53 <p>
54 The functions <a class="link" href="bessel_first.html" title="Bessel Functions of the First and Second Kinds">cyl_bessel_j</a>
55 and <a class="link" href="bessel_first.html" title="Bessel Functions of the First and Second Kinds">cyl_neumann</a> return
56 the result of the Bessel functions of the first and second kinds respectively:
57 </p>
58 <p>
59 cyl_bessel_j(v, x) = J<sub>v</sub>(x)
60 </p>
61 <p>
62 cyl_neumann(v, x) = Y<sub>v</sub>(x) = N<sub>v</sub>(x)
63 </p>
64 <p>
65 where:
66 </p>
67 <p>
68 <span class="inlinemediaobject"><img src="../../../equations/bessel2.svg"></span>
69 </p>
70 <p>
71 <span class="inlinemediaobject"><img src="../../../equations/bessel3.svg"></span>
72 </p>
73 <p>
74 The return type of these functions is computed using the <a class="link" href="../result_type.html" title="Calculation of the Type of the Result"><span class="emphasis"><em>result
75 type calculation rules</em></span></a> when T1 and T2 are different types.
76 The functions are also optimised for the relatively common case that T1 is
77 an integer.
78 </p>
79 <p>
80 The final <a class="link" href="../../policy.html" title="Chapter&#160;15.&#160;Policies: Controlling Precision, Error Handling etc">Policy</a> argument is optional and can
81 be used to control the behaviour of the function: how it handles errors,
82 what level of precision to use etc. Refer to the <a class="link" href="../../policy.html" title="Chapter&#160;15.&#160;Policies: Controlling Precision, Error Handling etc">policy
83 documentation for more details</a>.
84 </p>
85 <p>
86 The functions return the result of <a class="link" href="../error_handling.html#math_toolkit.error_handling.domain_error">domain_error</a>
87 whenever the result is undefined or complex. For <a class="link" href="bessel_first.html" title="Bessel Functions of the First and Second Kinds">cyl_bessel_j</a>
88 this occurs when <code class="computeroutput"><span class="identifier">x</span> <span class="special">&lt;</span>
89 <span class="number">0</span></code> and v is not an integer, or when
90 <code class="computeroutput"><span class="identifier">x</span> <span class="special">==</span>
91 <span class="number">0</span></code> and <code class="computeroutput"><span class="identifier">v</span>
92 <span class="special">!=</span> <span class="number">0</span></code>.
93 For <a class="link" href="bessel_first.html" title="Bessel Functions of the First and Second Kinds">cyl_neumann</a> this
94 occurs when <code class="computeroutput"><span class="identifier">x</span> <span class="special">&lt;=</span>
95 <span class="number">0</span></code>.
96 </p>
97 <p>
98 The following graph illustrates the cyclic nature of J<sub>v</sub>:
99 </p>
100 <p>
101 <span class="inlinemediaobject"><img src="../../../graphs/cyl_bessel_j.svg" align="middle"></span>
102 </p>
103 <p>
104 The following graph shows the behaviour of Y<sub>v</sub>: this is also cyclic for large
105 <span class="emphasis"><em>x</em></span>, but tends to -&#8734; &#160; for small <span class="emphasis"><em>x</em></span>:
106 </p>
107 <p>
108 <span class="inlinemediaobject"><img src="../../../graphs/cyl_neumann.svg" align="middle"></span>
109 </p>
110 <h5>
111 <a name="math_toolkit.bessel.bessel_first.h2"></a>
112 <span class="phrase"><a name="math_toolkit.bessel.bessel_first.testing"></a></span><a class="link" href="bessel_first.html#math_toolkit.bessel.bessel_first.testing">Testing</a>
113 </h5>
114 <p>
115 There are two sets of test values: spot values calculated using <a href="http://functions.wolfram.com" target="_top">functions.wolfram.com</a>,
116 and a much larger set of tests computed using a simplified version of this
117 implementation (with all the special case handling removed).
118 </p>
119 <h5>
120 <a name="math_toolkit.bessel.bessel_first.h3"></a>
121 <span class="phrase"><a name="math_toolkit.bessel.bessel_first.accuracy"></a></span><a class="link" href="bessel_first.html#math_toolkit.bessel.bessel_first.accuracy">Accuracy</a>
122 </h5>
123 <p>
124 The following tables show how the accuracy of these functions varies on various
125 platforms, along with comparisons to other libraries. Note that the cyclic
126 nature of these functions means that they have an infinite number of irrational
127 roots: in general these functions have arbitrarily large <span class="emphasis"><em>relative</em></span>
128 errors when the arguments are sufficiently close to a root. Of course the
129 absolute error in such cases is always small. Note that only results for
130 the widest floating-point type on the system are given as narrower types
131 have <a class="link" href="../relative_error.html#math_toolkit.relative_error.zero_error">effectively zero
132 error</a>. All values are relative errors in units of epsilon. Most of
133 the gross errors exhibited by other libraries occur for very large arguments
134 - you will need to drill down into the actual program output if you need
135 more information on this.
136 </p>
137 <div class="table">
138 <a name="math_toolkit.bessel.bessel_first.table_cyl_bessel_j_integer_orders_"></a><p class="title"><b>Table&#160;6.40.&#160;Error rates for cyl_bessel_j (integer orders)</b></p>
139 <div class="table-contents"><table class="table" summary="Error rates for cyl_bessel_j (integer orders)">
140 <colgroup>
141 <col>
142 <col>
143 <col>
144 <col>
145 <col>
146 </colgroup>
147 <thead><tr>
148 <th>
149 </th>
150 <th>
151 <p>
152 Microsoft Visual C++ version 12.0<br> Win32<br> double
153 </p>
154 </th>
155 <th>
156 <p>
157 GNU C++ version 5.1.0<br> linux<br> long double
158 </p>
159 </th>
160 <th>
161 <p>
162 GNU C++ version 5.1.0<br> linux<br> double
163 </p>
164 </th>
165 <th>
166 <p>
167 Sun compiler version 0x5130<br> Sun Solaris<br> long double
168 </p>
169 </th>
170 </tr></thead>
171 <tbody>
172 <tr>
173 <td>
174 <p>
175 Bessel J0: Mathworld Data (Integer Version)
176 </p>
177 </td>
178 <td>
179 <p>
180 <span class="blue">Max = 2.52&#949; (Mean = 1.2&#949;)</span><br> <br>
181 (<span class="emphasis"><em>&lt;math.h&gt;:</em></span> Max = 1.89&#949; (Mean = 0.988&#949;))
182 </p>
183 </td>
184 <td>
185 <p>
186 <span class="blue">Max = 6.55&#949; (Mean = 2.89&#949;)</span><br> <br>
187 (<span class="emphasis"><em>&lt;tr1/cmath&gt;:</em></span> Max = 5.04&#949; (Mean = 1.78&#949;)
188 <a class="link" href="../logs_and_tables/logs.html#errors_GNU_C_version_5_1_0_linux_long_double_cyl_bessel_j_integer_orders___tr1_cmath__Bessel_J0_Mathworld_Data_Integer_Version_">And
189 other failures.</a>)
190 </p>
191 </td>
192 <td>
193 <p>
194 <span class="blue">Max = 0&#949; (Mean = 0&#949;)</span><br> <br> (<span class="emphasis"><em>GSL
195 1.16:</em></span> Max = 1.12&#949; (Mean = 0.488&#949;))<br> (<span class="emphasis"><em>Rmath
196 3.0.2:</em></span> <span class="red">Max = +INF&#949; (Mean = +INF&#949;) <a class="link" href="../logs_and_tables/logs.html#errors_GNU_C_version_5_1_0_linux_double_cyl_bessel_j_integer_orders__Rmath_3_0_2_Bessel_J0_Mathworld_Data_Integer_Version_">And
197 other failures.</a>)</span><br> (<span class="emphasis"><em>Cephes:</em></span>
198 Max = 1.12&#949; (Mean = 0.568&#949;))
199 </p>
200 </td>
201 <td>
202 <p>
203 <span class="blue">Max = 6.55&#949; (Mean = 2.86&#949;)</span>
204 </p>
205 </td>
206 </tr>
207 <tr>
208 <td>
209 <p>
210 Bessel J0: Mathworld Data (Tricky cases) (Integer Version)
211 </p>
212 </td>
213 <td>
214 <p>
215 <span class="blue">Max = 1e+007&#949; (Mean = 4.09e+006&#949;)</span><br>
216 <br> (<span class="emphasis"><em>&lt;math.h&gt;:</em></span> <span class="red">Max
217 = 2.54e+008&#949; (Mean = 1.04e+008&#949;))</span>
218 </p>
219 </td>
220 <td>
221 <p>
222 <span class="blue">Max = 1.63e+08&#949; (Mean = 6.67e+07&#949;)</span><br>
223 <br> (<span class="emphasis"><em>&lt;tr1/cmath&gt;:</em></span> Max = 4.79e+08&#949; (Mean
224 = 1.96e+08&#949;))
225 </p>
226 </td>
227 <td>
228 <p>
229 <span class="blue">Max = 7.98e+04&#949; (Mean = 3.26e+04&#949;)</span><br>
230 <br> (<span class="emphasis"><em>GSL 1.16:</em></span> Max = 1e+07&#949; (Mean = 4.11e+06&#949;))<br>
231 (<span class="emphasis"><em>Rmath 3.0.2:</em></span> Max = 1.04e+07&#949; (Mean = 4.29e+06&#949;))<br>
232 (<span class="emphasis"><em>Cephes:</em></span> <span class="red">Max = 2.54e+08&#949; (Mean
233 = 1.04e+08&#949;))</span>
234 </p>
235 </td>
236 <td>
237 <p>
238 <span class="blue">Max = 1.64e+08&#949; (Mean = 6.69e+07&#949;)</span>
239 </p>
240 </td>
241 </tr>
242 <tr>
243 <td>
244 <p>
245 Bessel J1: Mathworld Data (Integer Version)
246 </p>
247 </td>
248 <td>
249 <p>
250 <span class="blue">Max = 1.73&#949; (Mean = 0.976&#949;)</span><br> <br>
251 (<span class="emphasis"><em>&lt;math.h&gt;:</em></span> Max = 11.4&#949; (Mean = 4.15&#949;))
252 </p>
253 </td>
254 <td>
255 <p>
256 <span class="blue">Max = 2.66&#949; (Mean = 1.38&#949;)</span><br> <br>
257 (<span class="emphasis"><em>&lt;tr1/cmath&gt;:</em></span> Max = 6.1&#949; (Mean = 2.95&#949;)
258 <a class="link" href="../logs_and_tables/logs.html#errors_GNU_C_version_5_1_0_linux_long_double_cyl_bessel_j_integer_orders___tr1_cmath__Bessel_J1_Mathworld_Data_Integer_Version_">And
259 other failures.</a>)
260 </p>
261 </td>
262 <td>
263 <p>
264 <span class="blue">Max = 0&#949; (Mean = 0&#949;)</span><br> <br> (<span class="emphasis"><em>GSL
265 1.16:</em></span> Max = 1.89&#949; (Mean = 0.721&#949;))<br> (<span class="emphasis"><em>Rmath
266 3.0.2:</em></span> <span class="red">Max = +INF&#949; (Mean = +INF&#949;) <a class="link" href="../logs_and_tables/logs.html#errors_GNU_C_version_5_1_0_linux_double_cyl_bessel_j_integer_orders__Rmath_3_0_2_Bessel_J1_Mathworld_Data_Integer_Version_">And
267 other failures.</a>)</span><br> (<span class="emphasis"><em>Cephes:</em></span>
268 Max = 2.88&#949; (Mean = 1.12&#949;))
269 </p>
270 </td>
271 <td>
272 <p>
273 <span class="blue">Max = 1.44&#949; (Mean = 0.637&#949;)</span>
274 </p>
275 </td>
276 </tr>
277 <tr>
278 <td>
279 <p>
280 Bessel J1: Mathworld Data (tricky cases) (Integer Version)
281 </p>
282 </td>
283 <td>
284 <p>
285 <span class="blue">Max = 3.23e+004&#949; (Mean = 1.45e+004&#949;)</span><br>
286 <br> (<span class="emphasis"><em>&lt;math.h&gt;:</em></span> Max = 1.44e+007&#949; (Mean
287 = 6.5e+006&#949;))
288 </p>
289 </td>
290 <td>
291 <p>
292 <span class="blue">Max = 2.18e+05&#949; (Mean = 9.76e+04&#949;)</span><br>
293 <br> (<span class="emphasis"><em>&lt;tr1/cmath&gt;:</em></span> Max = 2.15e+06&#949; (Mean
294 = 1.58e+06&#949;))
295 </p>
296 </td>
297 <td>
298 <p>
299 <span class="blue">Max = 106&#949; (Mean = 47.5&#949;)</span><br> <br>
300 (<span class="emphasis"><em>GSL 1.16:</em></span> Max = 1.26e+06&#949; (Mean = 6.28e+05&#949;))<br>
301 (<span class="emphasis"><em>Rmath 3.0.2:</em></span> Max = 2.93e+06&#949; (Mean = 1.7e+06&#949;))<br>
302 (<span class="emphasis"><em>Cephes:</em></span> Max = 9.56e+05&#949; (Mean = 4.99e+05&#949;))
303 </p>
304 </td>
305 <td>
306 <p>
307 <span class="blue">Max = 2.18e+05&#949; (Mean = 9.76e+04&#949;)</span>
308 </p>
309 </td>
310 </tr>
311 <tr>
312 <td>
313 <p>
314 Bessel JN: Mathworld Data (Integer Version)
315 </p>
316 </td>
317 <td>
318 <p>
319 <span class="blue">Max = 14.7&#949; (Mean = 5.4&#949;)</span><br> <br>
320 (<span class="emphasis"><em>&lt;math.h&gt;:</em></span> <span class="red">Max =
321 +INF&#949; (Mean = +INF&#949;) <a class="link" href="../logs_and_tables/logs.html#errors_Microsoft_Visual_C_version_12_0_Win32_double_cyl_bessel_j_integer_orders___math_h__Bessel_JN_Mathworld_Data_Integer_Version_">And
322 other failures.</a>)</span>
323 </p>
324 </td>
325 <td>
326 <p>
327 <span class="blue">Max = 6.85&#949; (Mean = 3.41&#949;)</span><br> <br>
328 (<span class="emphasis"><em>&lt;tr1/cmath&gt;:</em></span> <span class="red">Max
329 = 2.13e+19&#949; (Mean = 5.16e+18&#949;) <a class="link" href="../logs_and_tables/logs.html#errors_GNU_C_version_5_1_0_linux_long_double_cyl_bessel_j_integer_orders___tr1_cmath__Bessel_JN_Mathworld_Data_Integer_Version_">And
330 other failures.</a>)</span>
331 </p>
332 </td>
333 <td>
334 <p>
335 <span class="blue">Max = 0&#949; (Mean = 0&#949;)</span><br> <br> (<span class="emphasis"><em>GSL
336 1.16:</em></span> Max = 6.9e+05&#949; (Mean = 2.53e+05&#949;) <a class="link" href="../logs_and_tables/logs.html#errors_GNU_C_version_5_1_0_linux_double_cyl_bessel_j_integer_orders__GSL_1_16_Bessel_JN_Mathworld_Data_Integer_Version_">And
337 other failures.</a>)<br> (<span class="emphasis"><em>Rmath 3.0.2:</em></span>
338 <span class="red">Max = +INF&#949; (Mean = +INF&#949;) <a class="link" href="../logs_and_tables/logs.html#errors_GNU_C_version_5_1_0_linux_double_cyl_bessel_j_integer_orders__Rmath_3_0_2_Bessel_JN_Mathworld_Data_Integer_Version_">And
339 other failures.</a>)</span><br> (<span class="emphasis"><em>Cephes:</em></span>
340 <span class="red">Max = +INF&#949; (Mean = +INF&#949;) <a class="link" href="../logs_and_tables/logs.html#errors_GNU_C_version_5_1_0_linux_double_cyl_bessel_j_integer_orders__Cephes_Bessel_JN_Mathworld_Data_Integer_Version_">And
341 other failures.</a>)</span>
342 </p>
343 </td>
344 <td>
345 <p>
346 <span class="blue">Max = 463&#949; (Mean = 112&#949;)</span>
347 </p>
348 </td>
349 </tr>
350 </tbody>
351 </table></div>
352 </div>
353 <br class="table-break"><div class="table">
354 <a name="math_toolkit.bessel.bessel_first.table_cyl_bessel_j"></a><p class="title"><b>Table&#160;6.41.&#160;Error rates for cyl_bessel_j</b></p>
355 <div class="table-contents"><table class="table" summary="Error rates for cyl_bessel_j">
356 <colgroup>
357 <col>
358 <col>
359 <col>
360 <col>
361 <col>
362 </colgroup>
363 <thead><tr>
364 <th>
365 </th>
366 <th>
367 <p>
368 Microsoft Visual C++ version 12.0<br> Win32<br> double
369 </p>
370 </th>
371 <th>
372 <p>
373 GNU C++ version 5.1.0<br> linux<br> long double
374 </p>
375 </th>
376 <th>
377 <p>
378 GNU C++ version 5.1.0<br> linux<br> double
379 </p>
380 </th>
381 <th>
382 <p>
383 Sun compiler version 0x5130<br> Sun Solaris<br> long double
384 </p>
385 </th>
386 </tr></thead>
387 <tbody>
388 <tr>
389 <td>
390 <p>
391 Bessel J0: Mathworld Data
392 </p>
393 </td>
394 <td>
395 <p>
396 <span class="blue">Max = 2.52&#949; (Mean = 1.2&#949;)</span>
397 </p>
398 </td>
399 <td>
400 <p>
401 <span class="blue">Max = 6.55&#949; (Mean = 2.89&#949;)</span><br> <br>
402 (<span class="emphasis"><em>&lt;tr1/cmath&gt;:</em></span> Max = 5.04&#949; (Mean = 1.78&#949;)
403 <a class="link" href="../logs_and_tables/logs.html#errors_GNU_C_version_5_1_0_linux_long_double_cyl_bessel_j__tr1_cmath__Bessel_J0_Mathworld_Data">And
404 other failures.</a>)
405 </p>
406 </td>
407 <td>
408 <p>
409 <span class="blue">Max = 0&#949; (Mean = 0&#949;)</span><br> <br> (<span class="emphasis"><em>GSL
410 1.16:</em></span> Max = 0.629&#949; (Mean = 0.223&#949;) <a class="link" href="../logs_and_tables/logs.html#errors_GNU_C_version_5_1_0_linux_double_cyl_bessel_j_GSL_1_16_Bessel_J0_Mathworld_Data">And
411 other failures.</a>)<br> (<span class="emphasis"><em>Rmath 3.0.2:</em></span>
412 <span class="red">Max = +INF&#949; (Mean = +INF&#949;) <a class="link" href="../logs_and_tables/logs.html#errors_GNU_C_version_5_1_0_linux_double_cyl_bessel_j_Rmath_3_0_2_Bessel_J0_Mathworld_Data">And
413 other failures.</a>)</span><br> (<span class="emphasis"><em>Cephes:</em></span>
414 Max = 1.12&#949; (Mean = 0.568&#949;))
415 </p>
416 </td>
417 <td>
418 <p>
419 <span class="blue">Max = 6.55&#949; (Mean = 2.86&#949;)</span>
420 </p>
421 </td>
422 </tr>
423 <tr>
424 <td>
425 <p>
426 Bessel J0: Mathworld Data (Tricky cases)
427 </p>
428 </td>
429 <td>
430 <p>
431 <span class="blue">Max = 1e+007&#949; (Mean = 4.09e+006&#949;)</span>
432 </p>
433 </td>
434 <td>
435 <p>
436 <span class="blue">Max = 1.63e+08&#949; (Mean = 6.67e+07&#949;)</span><br>
437 <br> (<span class="emphasis"><em>&lt;tr1/cmath&gt;:</em></span> Max = 4.79e+08&#949; (Mean
438 = 1.96e+08&#949;))
439 </p>
440 </td>
441 <td>
442 <p>
443 <span class="blue">Max = 7.98e+04&#949; (Mean = 3.26e+04&#949;)</span><br>
444 <br> (<span class="emphasis"><em>GSL 1.16:</em></span> Max = 6.5e+07&#949; (Mean = 2.66e+07&#949;))<br>
445 (<span class="emphasis"><em>Rmath 3.0.2:</em></span> Max = 1.04e+07&#949; (Mean = 4.29e+06&#949;))<br>
446 (<span class="emphasis"><em>Cephes:</em></span> <span class="red">Max = 2.54e+08&#949; (Mean
447 = 1.04e+08&#949;))</span>
448 </p>
449 </td>
450 <td>
451 <p>
452 <span class="blue">Max = 1.64e+08&#949; (Mean = 6.69e+07&#949;)</span>
453 </p>
454 </td>
455 </tr>
456 <tr>
457 <td>
458 <p>
459 Bessel J1: Mathworld Data
460 </p>
461 </td>
462 <td>
463 <p>
464 <span class="blue">Max = 1.73&#949; (Mean = 0.976&#949;)</span>
465 </p>
466 </td>
467 <td>
468 <p>
469 <span class="blue">Max = 2.66&#949; (Mean = 1.38&#949;)</span><br> <br>
470 (<span class="emphasis"><em>&lt;tr1/cmath&gt;:</em></span> Max = 6.1&#949; (Mean = 2.95&#949;)
471 <a class="link" href="../logs_and_tables/logs.html#errors_GNU_C_version_5_1_0_linux_long_double_cyl_bessel_j__tr1_cmath__Bessel_J1_Mathworld_Data">And
472 other failures.</a>)
473 </p>
474 </td>
475 <td>
476 <p>
477 <span class="blue">Max = 0&#949; (Mean = 0&#949;)</span><br> <br> (<span class="emphasis"><em>GSL
478 1.16:</em></span> Max = 6.62&#949; (Mean = 2.35&#949;) <a class="link" href="../logs_and_tables/logs.html#errors_GNU_C_version_5_1_0_linux_double_cyl_bessel_j_GSL_1_16_Bessel_J1_Mathworld_Data">And
479 other failures.</a>)<br> (<span class="emphasis"><em>Rmath 3.0.2:</em></span>
480 <span class="red">Max = +INF&#949; (Mean = +INF&#949;) <a class="link" href="../logs_and_tables/logs.html#errors_GNU_C_version_5_1_0_linux_double_cyl_bessel_j_Rmath_3_0_2_Bessel_J1_Mathworld_Data">And
481 other failures.</a>)</span><br> (<span class="emphasis"><em>Cephes:</em></span>
482 Max = 2.88&#949; (Mean = 1.12&#949;))
483 </p>
484 </td>
485 <td>
486 <p>
487 <span class="blue">Max = 1.44&#949; (Mean = 0.637&#949;)</span>
488 </p>
489 </td>
490 </tr>
491 <tr>
492 <td>
493 <p>
494 Bessel J1: Mathworld Data (tricky cases)
495 </p>
496 </td>
497 <td>
498 <p>
499 <span class="blue">Max = 3.23e+004&#949; (Mean = 1.45e+004&#949;)</span>
500 </p>
501 </td>
502 <td>
503 <p>
504 <span class="blue">Max = 2.18e+05&#949; (Mean = 9.76e+04&#949;)</span><br>
505 <br> (<span class="emphasis"><em>&lt;tr1/cmath&gt;:</em></span> Max = 2.15e+06&#949; (Mean
506 = 1.58e+06&#949;))
507 </p>
508 </td>
509 <td>
510 <p>
511 <span class="blue">Max = 106&#949; (Mean = 47.5&#949;)</span><br> <br>
512 (<span class="emphasis"><em>GSL 1.16:</em></span> Max = 8.75e+05&#949; (Mean = 5.32e+05&#949;))<br>
513 (<span class="emphasis"><em>Rmath 3.0.2:</em></span> Max = 2.93e+06&#949; (Mean = 1.7e+06&#949;))<br>
514 (<span class="emphasis"><em>Cephes:</em></span> Max = 9.56e+05&#949; (Mean = 4.99e+05&#949;))
515 </p>
516 </td>
517 <td>
518 <p>
519 <span class="blue">Max = 2.18e+05&#949; (Mean = 9.76e+04&#949;)</span>
520 </p>
521 </td>
522 </tr>
523 <tr>
524 <td>
525 <p>
526 Bessel JN: Mathworld Data
527 </p>
528 </td>
529 <td>
530 <p>
531 <span class="blue">Max = 14.7&#949; (Mean = 5.4&#949;)</span>
532 </p>
533 </td>
534 <td>
535 <p>
536 <span class="blue">Max = 6.85&#949; (Mean = 3.41&#949;)</span><br> <br>
537 (<span class="emphasis"><em>&lt;tr1/cmath&gt;:</em></span> <span class="red">Max
538 = 2.13e+19&#949; (Mean = 5.16e+18&#949;) <a class="link" href="../logs_and_tables/logs.html#errors_GNU_C_version_5_1_0_linux_long_double_cyl_bessel_j__tr1_cmath__Bessel_JN_Mathworld_Data">And
539 other failures.</a>)</span>
540 </p>
541 </td>
542 <td>
543 <p>
544 <span class="blue">Max = 0&#949; (Mean = 0&#949;)</span><br> <br> (<span class="emphasis"><em>GSL
545 1.16:</em></span> Max = 6.9e+05&#949; (Mean = 2.15e+05&#949;) <a class="link" href="../logs_and_tables/logs.html#errors_GNU_C_version_5_1_0_linux_double_cyl_bessel_j_GSL_1_16_Bessel_JN_Mathworld_Data">And
546 other failures.</a>)<br> (<span class="emphasis"><em>Rmath 3.0.2:</em></span>
547 <span class="red">Max = +INF&#949; (Mean = +INF&#949;) <a class="link" href="../logs_and_tables/logs.html#errors_GNU_C_version_5_1_0_linux_double_cyl_bessel_j_Rmath_3_0_2_Bessel_JN_Mathworld_Data">And
548 other failures.</a>)</span><br> (<span class="emphasis"><em>Cephes:</em></span>
549 Max = 5.53e+05&#949; (Mean = 1.9e+05&#949;))
550 </p>
551 </td>
552 <td>
553 <p>
554 <span class="blue">Max = 463&#949; (Mean = 112&#949;)</span>
555 </p>
556 </td>
557 </tr>
558 <tr>
559 <td>
560 <p>
561 Bessel J: Mathworld Data
562 </p>
563 </td>
564 <td>
565 <p>
566 <span class="blue">Max = 14.9&#949; (Mean = 3.82&#949;)</span>
567 </p>
568 </td>
569 <td>
570 <p>
571 <span class="blue">Max = 14.7&#949; (Mean = 4.05&#949;)</span><br> <br>
572 (<span class="emphasis"><em>&lt;tr1/cmath&gt;:</em></span> Max = 3.49e+05&#949; (Mean =
573 7.89e+04&#949;) <a class="link" href="../logs_and_tables/logs.html#errors_GNU_C_version_5_1_0_linux_long_double_cyl_bessel_j__tr1_cmath__Bessel_J_Mathworld_Data">And
574 other failures.</a>)
575 </p>
576 </td>
577 <td>
578 <p>
579 <span class="blue">Max = 10&#949; (Mean = 2.19&#949;)</span><br> <br>
580 (<span class="emphasis"><em>GSL 1.16:</em></span> Max = 2.39e+05&#949; (Mean = 5.24e+04&#949;)
581 <a class="link" href="../logs_and_tables/logs.html#errors_GNU_C_version_5_1_0_linux_double_cyl_bessel_j_GSL_1_16_Bessel_J_Mathworld_Data">And
582 other failures.</a>)<br> (<span class="emphasis"><em>Rmath 3.0.2:</em></span>
583 <span class="red">Max = +INF&#949; (Mean = +INF&#949;) <a class="link" href="../logs_and_tables/logs.html#errors_GNU_C_version_5_1_0_linux_double_cyl_bessel_j_Rmath_3_0_2_Bessel_J_Mathworld_Data">And
584 other failures.</a>)</span><br> (<span class="emphasis"><em>Cephes:</em></span>
585 Max = 5.47e+05&#949; (Mean = 1.3e+05&#949;))
586 </p>
587 </td>
588 <td>
589 <p>
590 <span class="blue">Max = 14.7&#949; (Mean = 4.12&#949;)</span>
591 </p>
592 </td>
593 </tr>
594 <tr>
595 <td>
596 <p>
597 Bessel J: Mathworld Data (large values)
598 </p>
599 </td>
600 <td>
601 <p>
602 <span class="blue">Max = 9.31&#949; (Mean = 5.52&#949;)</span>
603 </p>
604 </td>
605 <td>
606 <p>
607 <span class="blue">Max = 607&#949; (Mean = 305&#949;)</span><br> <br>
608 (<span class="emphasis"><em>&lt;tr1/cmath&gt;:</em></span> Max = 34.9&#949; (Mean = 17.4&#949;)
609 <a class="link" href="../logs_and_tables/logs.html#errors_GNU_C_version_5_1_0_linux_long_double_cyl_bessel_j__tr1_cmath__Bessel_J_Mathworld_Data_large_values_">And
610 other failures.</a>)
611 </p>
612 </td>
613 <td>
614 <p>
615 <span class="blue">Max = 0.536&#949; (Mean = 0.268&#949;)</span><br> <br>
616 (<span class="emphasis"><em>GSL 1.16:</em></span> Max = 4.91e+03&#949; (Mean = 2.46e+03&#949;)
617 <a class="link" href="../logs_and_tables/logs.html#errors_GNU_C_version_5_1_0_linux_double_cyl_bessel_j_GSL_1_16_Bessel_J_Mathworld_Data_large_values_">And
618 other failures.</a>)<br> (<span class="emphasis"><em>Rmath 3.0.2:</em></span>
619 Max = 35.9&#949; (Mean = 18.1&#949;))<br> (<span class="emphasis"><em>Cephes:</em></span> <span class="red">Max = +INF&#949; (Mean = +INF&#949;) <a class="link" href="../logs_and_tables/logs.html#errors_GNU_C_version_5_1_0_linux_double_cyl_bessel_j_Cephes_Bessel_J_Mathworld_Data_large_values_">And
620 other failures.</a>)</span>
621 </p>
622 </td>
623 <td>
624 <p>
625 <span class="blue">Max = 607&#949; (Mean = 305&#949;)</span>
626 </p>
627 </td>
628 </tr>
629 <tr>
630 <td>
631 <p>
632 Bessel JN: Random Data
633 </p>
634 </td>
635 <td>
636 <p>
637 <span class="blue">Max = 17.5&#949; (Mean = 1.46&#949;)</span>
638 </p>
639 </td>
640 <td>
641 <p>
642 <span class="blue">Max = 50.8&#949; (Mean = 4.15&#949;)</span><br> <br>
643 (<span class="emphasis"><em>&lt;tr1/cmath&gt;:</em></span> Max = 1.12e+03&#949; (Mean =
644 88.7&#949;))
645 </p>
646 </td>
647 <td>
648 <p>
649 <span class="blue">Max = 0&#949; (Mean = 0&#949;)</span><br> <br> (<span class="emphasis"><em>GSL
650 1.16:</em></span> Max = 75.7&#949; (Mean = 5.36&#949;))<br> (<span class="emphasis"><em>Rmath
651 3.0.2:</em></span> Max = 3.93&#949; (Mean = 1.22&#949;))<br> (<span class="emphasis"><em>Cephes:</em></span>
652 Max = 91.4&#949; (Mean = 6.47&#949;))
653 </p>
654 </td>
655 <td>
656 <p>
657 <span class="blue">Max = 99.6&#949; (Mean = 22&#949;)</span>
658 </p>
659 </td>
660 </tr>
661 <tr>
662 <td>
663 <p>
664 Bessel J: Random Data
665 </p>
666 </td>
667 <td>
668 <p>
669 <span class="blue">Max = 9.24&#949; (Mean = 1.36&#949;)</span>
670 </p>
671 </td>
672 <td>
673 <p>
674 <span class="blue">Max = 9.81&#949; (Mean = 1.59&#949;)</span><br> <br>
675 (<span class="emphasis"><em>&lt;tr1/cmath&gt;:</em></span> Max = 501&#949; (Mean = 52.3&#949;))
676 </p>
677 </td>
678 <td>
679 <p>
680 <span class="blue">Max = 0&#949; (Mean = 0&#949;)</span><br> <br> (<span class="emphasis"><em>GSL
681 1.16:</em></span> Max = 15.5&#949; (Mean = 3.33&#949;) <a class="link" href="../logs_and_tables/logs.html#errors_GNU_C_version_5_1_0_linux_double_cyl_bessel_j_GSL_1_16_Bessel_J_Random_Data">And
682 other failures.</a>)<br> (<span class="emphasis"><em>Rmath 3.0.2:</em></span>
683 Max = 6.74&#949; (Mean = 1.3&#949;))<br> (<span class="emphasis"><em>Cephes:</em></span> Max
684 = 16.7&#949; (Mean = 2.5&#949;))
685 </p>
686 </td>
687 <td>
688 <p>
689 <span class="blue">Max = 260&#949; (Mean = 34&#949;)</span>
690 </p>
691 </td>
692 </tr>
693 <tr>
694 <td>
695 <p>
696 Bessel J: Random Data (Tricky large values)
697 </p>
698 </td>
699 <td>
700 <p>
701 <span class="blue">Max = 59.2&#949; (Mean = 8.67&#949;)</span>
702 </p>
703 </td>
704 <td>
705 <p>
706 <span class="blue">Max = 785&#949; (Mean = 94.2&#949;)</span><br> <br>
707 (<span class="emphasis"><em>&lt;tr1/cmath&gt;:</em></span> <span class="red">Max
708 = 5.01e+17&#949; (Mean = 6.23e+16&#949;))</span>
709 </p>
710 </td>
711 <td>
712 <p>
713 <span class="blue">Max = 0&#949; (Mean = 0&#949;)</span><br> <br> (<span class="emphasis"><em>GSL
714 1.16:</em></span> Max = 2.48e+05&#949; (Mean = 5.11e+04&#949;))<br> (<span class="emphasis"><em>Rmath
715 3.0.2:</em></span> Max = 71.6&#949; (Mean = 11.7&#949;))<br> (<span class="emphasis"><em>Cephes:</em></span>
716 Max = 2.48e+05&#949; (Mean = 3.02e+04&#949;))
717 </p>
718 </td>
719 <td>
720 <p>
721 <span class="blue">Max = 785&#949; (Mean = 97.4&#949;)</span>
722 </p>
723 </td>
724 </tr>
725 </tbody>
726 </table></div>
727 </div>
728 <br class="table-break"><div class="table">
729 <a name="math_toolkit.bessel.bessel_first.table_cyl_neumann_integer_orders_"></a><p class="title"><b>Table&#160;6.42.&#160;Error rates for cyl_neumann (integer orders)</b></p>
730 <div class="table-contents"><table class="table" summary="Error rates for cyl_neumann (integer orders)">
731 <colgroup>
732 <col>
733 <col>
734 <col>
735 <col>
736 <col>
737 </colgroup>
738 <thead><tr>
739 <th>
740 </th>
741 <th>
742 <p>
743 Microsoft Visual C++ version 12.0<br> Win32<br> double
744 </p>
745 </th>
746 <th>
747 <p>
748 GNU C++ version 5.1.0<br> linux<br> double
749 </p>
750 </th>
751 <th>
752 <p>
753 GNU C++ version 5.1.0<br> linux<br> long double
754 </p>
755 </th>
756 <th>
757 <p>
758 Sun compiler version 0x5130<br> Sun Solaris<br> long double
759 </p>
760 </th>
761 </tr></thead>
762 <tbody>
763 <tr>
764 <td>
765 <p>
766 Y0: Mathworld Data (Integer Version)
767 </p>
768 </td>
769 <td>
770 <p>
771 <span class="blue">Max = 4.61&#949; (Mean = 2.29&#949;)</span><br> <br>
772 (<span class="emphasis"><em>&lt;math.h&gt;:</em></span> Max = 5.37e+003&#949; (Mean = 1.81e+003&#949;))
773 </p>
774 </td>
775 <td>
776 <p>
777 <span class="blue">Max = 0&#949; (Mean = 0&#949;)</span><br> <br> (<span class="emphasis"><em>GSL
778 1.16:</em></span> Max = 6.46&#949; (Mean = 2.38&#949;))<br> (<span class="emphasis"><em>Rmath
779 3.0.2:</em></span> Max = 167&#949; (Mean = 56.5&#949;))<br> (<span class="emphasis"><em>Cephes:</em></span>
780 Max = 5.37e+03&#949; (Mean = 1.81e+03&#949;))
781 </p>
782 </td>
783 <td>
784 <p>
785 <span class="blue">Max = 5.59&#949; (Mean = 2.54&#949;)</span><br> <br>
786 (<span class="emphasis"><em>&lt;tr1/cmath&gt;:</em></span> Max = 2.05e+05&#949; (Mean =
787 6.87e+04&#949;))
788 </p>
789 </td>
790 <td>
791 <p>
792 <span class="blue">Max = 5.53&#949; (Mean = 2.4&#949;)</span>
793 </p>
794 </td>
795 </tr>
796 <tr>
797 <td>
798 <p>
799 Y1: Mathworld Data (Integer Version)
800 </p>
801 </td>
802 <td>
803 <p>
804 <span class="blue">Max = 4.75&#949; (Mean = 1.72&#949;)</span><br> <br>
805 (<span class="emphasis"><em>&lt;math.h&gt;:</em></span> Max = 1.86e+004&#949; (Mean = 6.2e+003&#949;))
806 </p>
807 </td>
808 <td>
809 <p>
810 <span class="blue">Max = 0&#949; (Mean = 0&#949;)</span><br> <br> (<span class="emphasis"><em>GSL
811 1.16:</em></span> Max = 1.51&#949; (Mean = 0.839&#949;))<br> (<span class="emphasis"><em>Rmath
812 3.0.2:</em></span> Max = 193&#949; (Mean = 64.4&#949;))<br> (<span class="emphasis"><em>Cephes:</em></span>
813 Max = 1.86e+04&#949; (Mean = 6.2e+03&#949;))
814 </p>
815 </td>
816 <td>
817 <p>
818 <span class="blue">Max = 12.7&#949; (Mean = 4.34&#949;)</span><br> <br>
819 (<span class="emphasis"><em>&lt;tr1/cmath&gt;:</em></span> Max = 9.71e+03&#949; (Mean =
820 4.08e+03&#949;))
821 </p>
822 </td>
823 <td>
824 <p>
825 <span class="blue">Max = 6.33&#949; (Mean = 2.29&#949;)</span>
826 </p>
827 </td>
828 </tr>
829 <tr>
830 <td>
831 <p>
832 Yn: Mathworld Data (Integer Version)
833 </p>
834 </td>
835 <td>
836 <p>
837 <span class="blue">Max = 35&#949; (Mean = 11.8&#949;)</span><br> <br>
838 (<span class="emphasis"><em>&lt;math.h&gt;:</em></span> Max = 2.49e+005&#949; (Mean = 8.14e+004&#949;))
839 </p>
840 </td>
841 <td>
842 <p>
843 <span class="blue">Max = 0.993&#949; (Mean = 0.314&#949;)</span><br> <br>
844 (<span class="emphasis"><em>GSL 1.16:</em></span> Max = 2.41e+05&#949; (Mean = 7.62e+04&#949;))<br>
845 (<span class="emphasis"><em>Rmath 3.0.2:</em></span> Max = 1.24e+04&#949; (Mean = 4e+03&#949;))<br>
846 (<span class="emphasis"><em>Cephes:</em></span> Max = 2.49e+05&#949; (Mean = 8.14e+04&#949;))
847 </p>
848 </td>
849 <td>
850 <p>
851 <span class="blue">Max = 55.2&#949; (Mean = 17.7&#949;)</span><br> <br>
852 (<span class="emphasis"><em>&lt;tr1/cmath&gt;:</em></span> <span class="red">Max
853 = 2.2e+20&#949; (Mean = 6.97e+19&#949;) <a class="link" href="../logs_and_tables/logs.html#errors_GNU_C_version_5_1_0_linux_long_double_cyl_neumann_integer_orders___tr1_cmath__Yn_Mathworld_Data_Integer_Version_">And
854 other failures.</a>)</span>
855 </p>
856 </td>
857 <td>
858 <p>
859 <span class="blue">Max = 55.2&#949; (Mean = 17.8&#949;)</span>
860 </p>
861 </td>
862 </tr>
863 </tbody>
864 </table></div>
865 </div>
866 <br class="table-break"><div class="table">
867 <a name="math_toolkit.bessel.bessel_first.table_cyl_neumann"></a><p class="title"><b>Table&#160;6.43.&#160;Error rates for cyl_neumann</b></p>
868 <div class="table-contents"><table class="table" summary="Error rates for cyl_neumann">
869 <colgroup>
870 <col>
871 <col>
872 <col>
873 <col>
874 <col>
875 </colgroup>
876 <thead><tr>
877 <th>
878 </th>
879 <th>
880 <p>
881 Microsoft Visual C++ version 12.0<br> Win32<br> double
882 </p>
883 </th>
884 <th>
885 <p>
886 GNU C++ version 5.1.0<br> linux<br> double
887 </p>
888 </th>
889 <th>
890 <p>
891 GNU C++ version 5.1.0<br> linux<br> long double
892 </p>
893 </th>
894 <th>
895 <p>
896 Sun compiler version 0x5130<br> Sun Solaris<br> long double
897 </p>
898 </th>
899 </tr></thead>
900 <tbody>
901 <tr>
902 <td>
903 <p>
904 Y0: Mathworld Data
905 </p>
906 </td>
907 <td>
908 <p>
909 <span class="blue">Max = 4.61&#949; (Mean = 2.29&#949;)</span>
910 </p>
911 </td>
912 <td>
913 <p>
914 <span class="blue">Max = 0&#949; (Mean = 0&#949;)</span><br> <br> (<span class="emphasis"><em>GSL
915 1.16:</em></span> Max = 60.9&#949; (Mean = 20.4&#949;))<br> (<span class="emphasis"><em>Rmath
916 3.0.2:</em></span> Max = 167&#949; (Mean = 56.5&#949;))
917 </p>
918 </td>
919 <td>
920 <p>
921 <span class="blue">Max = 5.59&#949; (Mean = 2.54&#949;)</span><br> <br>
922 (<span class="emphasis"><em>&lt;tr1/cmath&gt;:</em></span> Max = 2.05e+05&#949; (Mean =
923 6.87e+04&#949;))
924 </p>
925 </td>
926 <td>
927 <p>
928 <span class="blue">Max = 5.53&#949; (Mean = 2.4&#949;)</span>
929 </p>
930 </td>
931 </tr>
932 <tr>
933 <td>
934 <p>
935 Y1: Mathworld Data
936 </p>
937 </td>
938 <td>
939 <p>
940 <span class="blue">Max = 4.75&#949; (Mean = 1.72&#949;)</span>
941 </p>
942 </td>
943 <td>
944 <p>
945 <span class="blue">Max = 0&#949; (Mean = 0&#949;)</span><br> <br> (<span class="emphasis"><em>GSL
946 1.16:</em></span> Max = 23.4&#949; (Mean = 8.1&#949;))<br> (<span class="emphasis"><em>Rmath
947 3.0.2:</em></span> Max = 193&#949; (Mean = 64.4&#949;))
948 </p>
949 </td>
950 <td>
951 <p>
952 <span class="blue">Max = 12.7&#949; (Mean = 4.34&#949;)</span><br> <br>
953 (<span class="emphasis"><em>&lt;tr1/cmath&gt;:</em></span> Max = 9.71e+03&#949; (Mean =
954 4.08e+03&#949;))
955 </p>
956 </td>
957 <td>
958 <p>
959 <span class="blue">Max = 6.33&#949; (Mean = 2.29&#949;)</span>
960 </p>
961 </td>
962 </tr>
963 <tr>
964 <td>
965 <p>
966 Yn: Mathworld Data
967 </p>
968 </td>
969 <td>
970 <p>
971 <span class="blue">Max = 35&#949; (Mean = 11.8&#949;)</span>
972 </p>
973 </td>
974 <td>
975 <p>
976 <span class="blue">Max = 0.993&#949; (Mean = 0.314&#949;)</span><br> <br>
977 (<span class="emphasis"><em>GSL 1.16:</em></span> Max = 2.41e+05&#949; (Mean = 7.62e+04&#949;)
978 <a class="link" href="../logs_and_tables/logs.html#errors_GNU_C_version_5_1_0_linux_double_cyl_neumann_GSL_1_16_Yn_Mathworld_Data">And
979 other failures.</a>)<br> (<span class="emphasis"><em>Rmath 3.0.2:</em></span>
980 Max = 1.24e+04&#949; (Mean = 4e+03&#949;))
981 </p>
982 </td>
983 <td>
984 <p>
985 <span class="blue">Max = 55.2&#949; (Mean = 17.7&#949;)</span><br> <br>
986 (<span class="emphasis"><em>&lt;tr1/cmath&gt;:</em></span> <span class="red">Max
987 = 2.2e+20&#949; (Mean = 6.97e+19&#949;) <a class="link" href="../logs_and_tables/logs.html#errors_GNU_C_version_5_1_0_linux_long_double_cyl_neumann__tr1_cmath__Yn_Mathworld_Data">And
988 other failures.</a>)</span>
989 </p>
990 </td>
991 <td>
992 <p>
993 <span class="blue">Max = 55.2&#949; (Mean = 17.8&#949;)</span>
994 </p>
995 </td>
996 </tr>
997 <tr>
998 <td>
999 <p>
1000 Yv: Mathworld Data
1001 </p>
1002 </td>
1003 <td>
1004 <p>
1005 <span class="blue">Max = 7.89&#949; (Mean = 3.27&#949;)</span>
1006 </p>
1007 </td>
1008 <td>
1009 <p>
1010 <span class="blue">Max = 10&#949; (Mean = 3.02&#949;)</span><br> <br>
1011 (<span class="emphasis"><em>GSL 1.16:</em></span> Max = 1.07e+05&#949; (Mean = 3.22e+04&#949;)
1012 <a class="link" href="../logs_and_tables/logs.html#errors_GNU_C_version_5_1_0_linux_double_cyl_neumann_GSL_1_16_Yv_Mathworld_Data">And
1013 other failures.</a>)<br> (<span class="emphasis"><em>Rmath 3.0.2:</em></span>
1014 Max = 1.05e+03&#949; (Mean = 326&#949;))
1015 </p>
1016 </td>
1017 <td>
1018 <p>
1019 <span class="blue">Max = 10.7&#949; (Mean = 4.92&#949;)</span><br> <br>
1020 (<span class="emphasis"><em>&lt;tr1/cmath&gt;:</em></span> <span class="red">Max
1021 = 3.49e+15&#949; (Mean = 1.05e+15&#949;) <a class="link" href="../logs_and_tables/logs.html#errors_GNU_C_version_5_1_0_linux_long_double_cyl_neumann__tr1_cmath__Yv_Mathworld_Data">And
1022 other failures.</a>)</span>
1023 </p>
1024 </td>
1025 <td>
1026 <p>
1027 <span class="blue">Max = 10.7&#949; (Mean = 5.1&#949;)</span>
1028 </p>
1029 </td>
1030 </tr>
1031 <tr>
1032 <td>
1033 <p>
1034 Yv: Mathworld Data (large values)
1035 </p>
1036 </td>
1037 <td>
1038 <p>
1039 <span class="blue">Max = 0.682&#949; (Mean = 0.35&#949;)</span>
1040 </p>
1041 </td>
1042 <td>
1043 <p>
1044 <span class="blue">Max = 0&#949; (Mean = 0&#949;)</span><br> <br> (<span class="emphasis"><em>GSL
1045 1.16:</em></span> Max = 60.8&#949; (Mean = 23&#949;) <a class="link" href="../logs_and_tables/logs.html#errors_GNU_C_version_5_1_0_linux_double_cyl_neumann_GSL_1_16_Yv_Mathworld_Data_large_values_">And
1046 other failures.</a>)<br> (<span class="emphasis"><em>Rmath 3.0.2:</em></span>
1047 <span class="red">Max = +INF&#949; (Mean = +INF&#949;) <a class="link" href="../logs_and_tables/logs.html#errors_GNU_C_version_5_1_0_linux_double_cyl_neumann_Rmath_3_0_2_Yv_Mathworld_Data_large_values_">And
1048 other failures.</a>)</span>
1049 </p>
1050 </td>
1051 <td>
1052 <p>
1053 <span class="blue">Max = 1.57&#949; (Mean = 1.17&#949;)</span><br> <br>
1054 (<span class="emphasis"><em>&lt;tr1/cmath&gt;:</em></span> Max = 43.2&#949; (Mean = 16.3&#949;)
1055 <a class="link" href="../logs_and_tables/logs.html#errors_GNU_C_version_5_1_0_linux_long_double_cyl_neumann__tr1_cmath__Yv_Mathworld_Data_large_values_">And
1056 other failures.</a>)
1057 </p>
1058 </td>
1059 <td>
1060 <p>
1061 <span class="blue">Max = 1.57&#949; (Mean = 1.24&#949;)</span>
1062 </p>
1063 </td>
1064 </tr>
1065 <tr>
1066 <td>
1067 <p>
1068 Y0 and Y1: Random Data
1069 </p>
1070 </td>
1071 <td>
1072 <p>
1073 <span class="blue">Max = 4.17&#949; (Mean = 1.24&#949;)</span>
1074 </p>
1075 </td>
1076 <td>
1077 <p>
1078 <span class="blue">Max = 0&#949; (Mean = 0&#949;)</span><br> <br> (<span class="emphasis"><em>GSL
1079 1.16:</em></span> Max = 34.4&#949; (Mean = 8.9&#949;))<br> (<span class="emphasis"><em>Rmath
1080 3.0.2:</em></span> Max = 83&#949; (Mean = 14.2&#949;))
1081 </p>
1082 </td>
1083 <td>
1084 <p>
1085 <span class="blue">Max = 11.8&#949; (Mean = 3.28&#949;)</span><br> <br>
1086 (<span class="emphasis"><em>&lt;tr1/cmath&gt;:</em></span> Max = 2.59e+03&#949; (Mean =
1087 500&#949;))
1088 </p>
1089 </td>
1090 <td>
1091 <p>
1092 <span class="blue">Max = 10.8&#949; (Mean = 3.04&#949;)</span>
1093 </p>
1094 </td>
1095 </tr>
1096 <tr>
1097 <td>
1098 <p>
1099 Yn: Random Data
1100 </p>
1101 </td>
1102 <td>
1103 <p>
1104 <span class="blue">Max = 117&#949; (Mean = 10.2&#949;)</span>
1105 </p>
1106 </td>
1107 <td>
1108 <p>
1109 <span class="blue">Max = 0&#949; (Mean = 0&#949;)</span><br> <br> (<span class="emphasis"><em>GSL
1110 1.16:</em></span> Max = 500&#949; (Mean = 47.8&#949;))<br> (<span class="emphasis"><em>Rmath
1111 3.0.2:</em></span> Max = 691&#949; (Mean = 67.9&#949;))
1112 </p>
1113 </td>
1114 <td>
1115 <p>
1116 <span class="blue">Max = 338&#949; (Mean = 28.2&#949;)</span><br> <br>
1117 (<span class="emphasis"><em>&lt;tr1/cmath&gt;:</em></span> Max = 4.01e+03&#949; (Mean =
1118 348&#949;))
1119 </p>
1120 </td>
1121 <td>
1122 <p>
1123 <span class="blue">Max = 338&#949; (Mean = 27.5&#949;)</span>
1124 </p>
1125 </td>
1126 </tr>
1127 <tr>
1128 <td>
1129 <p>
1130 Yv: Random Data
1131 </p>
1132 </td>
1133 <td>
1134 <p>
1135 <span class="blue">Max = 1.23e+003&#949; (Mean = 69.9&#949;)</span>
1136 </p>
1137 </td>
1138 <td>
1139 <p>
1140 <span class="blue">Max = 1.53&#949; (Mean = 0.102&#949;)</span><br> <br>
1141 (<span class="emphasis"><em>GSL 1.16:</em></span> Max = 1.41e+06&#949; (Mean = 7.67e+04&#949;))<br>
1142 (<span class="emphasis"><em>Rmath 3.0.2:</em></span> Max = 1.79e+05&#949; (Mean = 9.64e+03&#949;))
1143 </p>
1144 </td>
1145 <td>
1146 <p>
1147 <span class="blue">Max = 2.08e+03&#949; (Mean = 149&#949;)</span><br>
1148 <br> (<span class="emphasis"><em>&lt;tr1/cmath&gt;:</em></span> <span class="red">Max
1149 = +INF&#949; (Mean = +INF&#949;) <a class="link" href="../logs_and_tables/logs.html#errors_GNU_C_version_5_1_0_linux_long_double_cyl_neumann__tr1_cmath__Yv_Random_Data">And
1150 other failures.</a>)</span>
1151 </p>
1152 </td>
1153 <td>
1154 <p>
1155 <span class="blue">Max = 2.08e+03&#949; (Mean = 149&#949;)</span>
1156 </p>
1157 </td>
1158 </tr>
1159 </tbody>
1160 </table></div>
1161 </div>
1162 <br class="table-break"><p>
1163 Note that for large <span class="emphasis"><em>x</em></span> these functions are largely dependent
1164 on the accuracy of the <code class="computeroutput"><span class="identifier">std</span><span class="special">::</span><span class="identifier">sin</span></code> and
1165 <code class="computeroutput"><span class="identifier">std</span><span class="special">::</span><span class="identifier">cos</span></code> functions.
1166 </p>
1167 <p>
1168 Comparison to GSL and <a href="http://www.netlib.org/cephes/" target="_top">Cephes</a>
1169 is interesting: both <a href="http://www.netlib.org/cephes/" target="_top">Cephes</a>
1170 and this library optimise the integer order case - leading to identical results
1171 - simply using the general case is for the most part slightly more accurate
1172 though, as noted by the better accuracy of GSL in the integer argument cases.
1173 This implementation tends to perform much better when the arguments become
1174 large, <a href="http://www.netlib.org/cephes/" target="_top">Cephes</a> in particular
1175 produces some remarkably inaccurate results with some of the test data (no
1176 significant figures correct), and even GSL performs badly with some inputs
1177 to J<sub>v</sub>. Note that by way of double-checking these results, the worst performing
1178 <a href="http://www.netlib.org/cephes/" target="_top">Cephes</a> and GSL cases were
1179 recomputed using <a href="http://functions.wolfram.com" target="_top">functions.wolfram.com</a>,
1180 and the result checked against our test data: no errors in the test data
1181 were found.
1182 </p>
1183 <h5>
1184 <a name="math_toolkit.bessel.bessel_first.h4"></a>
1185 <span class="phrase"><a name="math_toolkit.bessel.bessel_first.implementation"></a></span><a class="link" href="bessel_first.html#math_toolkit.bessel.bessel_first.implementation">Implementation</a>
1186 </h5>
1187 <p>
1188 The implementation is mostly about filtering off various special cases:
1189 </p>
1190 <p>
1191 When <span class="emphasis"><em>x</em></span> is negative, then the order <span class="emphasis"><em>v</em></span>
1192 must be an integer or the result is a domain error. If the order is an integer
1193 then the function is odd for odd orders and even for even orders, so we reflect
1194 to <span class="emphasis"><em>x &gt; 0</em></span>.
1195 </p>
1196 <p>
1197 When the order <span class="emphasis"><em>v</em></span> is negative then the reflection formulae
1198 can be used to move to <span class="emphasis"><em>v &gt; 0</em></span>:
1199 </p>
1200 <p>
1201 <span class="inlinemediaobject"><img src="../../../equations/bessel9.svg"></span>
1202 </p>
1203 <p>
1204 <span class="inlinemediaobject"><img src="../../../equations/bessel10.svg"></span>
1205 </p>
1206 <p>
1207 Note that if the order is an integer, then these formulae reduce to:
1208 </p>
1209 <p>
1210 J<sub>-n</sub> = (-1)<sup>n</sup>J<sub>n</sub>
1211 </p>
1212 <p>
1213 Y<sub>-n</sub> = (-1)<sup>n</sup>Y<sub>n</sub>
1214 </p>
1215 <p>
1216 However, in general, a negative order implies that we will need to compute
1217 both J and Y.
1218 </p>
1219 <p>
1220 When <span class="emphasis"><em>x</em></span> is large compared to the order <span class="emphasis"><em>v</em></span>
1221 then the asymptotic expansions for large <span class="emphasis"><em>x</em></span> in M. Abramowitz
1222 and I.A. Stegun, <span class="emphasis"><em>Handbook of Mathematical Functions</em></span>
1223 9.2.19 are used (these were found to be more reliable than those in A&amp;S
1224 9.2.5).
1225 </p>
1226 <p>
1227 When the order <span class="emphasis"><em>v</em></span> is an integer the method first relates
1228 the result to J<sub>0</sub>, J<sub>1</sub>, Y<sub>0</sub> &#160; and Y<sub>1</sub> &#160; using either forwards or backwards recurrence
1229 (Miller's algorithm) depending upon which is stable. The values for J<sub>0</sub>, J<sub>1</sub>,
1230 Y<sub>0</sub> &#160; and Y<sub>1</sub> &#160; are calculated using the rational minimax approximations on root-bracketing
1231 intervals for small <span class="emphasis"><em>|x|</em></span> and Hankel asymptotic expansion
1232 for large <span class="emphasis"><em>|x|</em></span>. The coefficients are from:
1233 </p>
1234 <p>
1235 W.J. Cody, <span class="emphasis"><em>ALGORITHM 715: SPECFUN - A Portable FORTRAN Package
1236 of Special Function Routines and Test Drivers</em></span>, ACM Transactions
1237 on Mathematical Software, vol 19, 22 (1993).
1238 </p>
1239 <p>
1240 and
1241 </p>
1242 <p>
1243 J.F. Hart et al, <span class="emphasis"><em>Computer Approximations</em></span>, John Wiley
1244 &amp; Sons, New York, 1968.
1245 </p>
1246 <p>
1247 These approximations are accurate to around 19 decimal digits: therefore
1248 these methods are not used when type T has more than 64 binary digits.
1249 </p>
1250 <p>
1251 When <span class="emphasis"><em>x</em></span> is smaller than machine epsilon then the following
1252 approximations for Y<sub>0</sub>(x), Y<sub>1</sub>(x), Y<sub>2</sub>(x) and Y<sub>n</sub>(x) can be used (see: <a href="http://functions.wolfram.com/03.03.06.0037.01" target="_top">http://functions.wolfram.com/03.03.06.0037.01</a>,
1253 <a href="http://functions.wolfram.com/03.03.06.0038.01" target="_top">http://functions.wolfram.com/03.03.06.0038.01</a>,
1254 <a href="http://functions.wolfram.com/03.03.06.0039.01" target="_top">http://functions.wolfram.com/03.03.06.0039.01</a>
1255 and <a href="http://functions.wolfram.com/03.03.06.0040.01" target="_top">http://functions.wolfram.com/03.03.06.0040.01</a>):
1256 </p>
1257 <p>
1258 <span class="inlinemediaobject"><img src="../../../equations/bessel_y0_small_z.svg"></span>
1259 </p>
1260 <p>
1261 <span class="inlinemediaobject"><img src="../../../equations/bessel_y1_small_z.svg"></span>
1262 </p>
1263 <p>
1264 <span class="inlinemediaobject"><img src="../../../equations/bessel_y2_small_z.svg"></span>
1265 </p>
1266 <p>
1267 <span class="inlinemediaobject"><img src="../../../equations/bessel_yn_small_z.svg"></span>
1268 </p>
1269 <p>
1270 When <span class="emphasis"><em>x</em></span> is small compared to <span class="emphasis"><em>v</em></span> and
1271 <span class="emphasis"><em>v</em></span> is not an integer, then the following series approximation
1272 can be used for Y<sub>v</sub>(x), this is also an area where other approximations are
1273 often too slow to converge to be used (see <a href="http://functions.wolfram.com/03.03.06.0034.01" target="_top">http://functions.wolfram.com/03.03.06.0034.01</a>):
1274 </p>
1275 <p>
1276 <span class="inlinemediaobject"><img src="../../../equations/bessel_yv_small_z.svg"></span>
1277 </p>
1278 <p>
1279 When <span class="emphasis"><em>x</em></span> is small compared to <span class="emphasis"><em>v</em></span>,
1280 J<sub>v</sub>x &#160; is best computed directly from the series:
1281 </p>
1282 <p>
1283 <span class="inlinemediaobject"><img src="../../../equations/bessel2.svg"></span>
1284 </p>
1285 <p>
1286 In the general case we compute J<sub>v</sub> &#160; and Y<sub>v</sub> &#160; simultaneously.
1287 </p>
1288 <p>
1289 To get the initial values, let &#956; &#160; = &#957; - floor(&#957; + 1/2), then &#956; &#160; is the fractional part
1290 of &#957; &#160; such that |&#956;| &lt;= 1/2 (we need this for convergence later). The idea
1291 is to calculate J<sub>&#956;</sub>(x), J<sub>&#956;+1</sub>(x), Y<sub>&#956;</sub>(x), Y<sub>&#956;+1</sub>(x) and use them to obtain J<sub>&#957;</sub>(x), Y<sub>&#957;</sub>(x).
1292 </p>
1293 <p>
1294 The algorithm is called Steed's method, which needs two continued fractions
1295 as well as the Wronskian:
1296 </p>
1297 <p>
1298 <span class="inlinemediaobject"><img src="../../../equations/bessel8.svg"></span>
1299 </p>
1300 <p>
1301 <span class="inlinemediaobject"><img src="../../../equations/bessel11.svg"></span>
1302 </p>
1303 <p>
1304 <span class="inlinemediaobject"><img src="../../../equations/bessel12.svg"></span>
1305 </p>
1306 <p>
1307 See: F.S. Acton, <span class="emphasis"><em>Numerical Methods that Work</em></span>, The Mathematical
1308 Association of America, Washington, 1997.
1309 </p>
1310 <p>
1311 The continued fractions are computed using the modified Lentz's method (W.J.
1312 Lentz, <span class="emphasis"><em>Generating Bessel functions in Mie scattering calculations
1313 using continued fractions</em></span>, Applied Optics, vol 15, 668 (1976)).
1314 Their convergence rates depend on <span class="emphasis"><em>x</em></span>, therefore we need
1315 different strategies for large <span class="emphasis"><em>x</em></span> and small <span class="emphasis"><em>x</em></span>.
1316 </p>
1317 <p>
1318 <span class="emphasis"><em>x &gt; v</em></span>, CF1 needs O(<span class="emphasis"><em>x</em></span>) iterations
1319 to converge, CF2 converges rapidly
1320 </p>
1321 <p>
1322 <span class="emphasis"><em>x &lt;= v</em></span>, CF1 converges rapidly, CF2 fails to converge
1323 when <span class="emphasis"><em>x</em></span> <code class="literal">-&gt;</code> 0
1324 </p>
1325 <p>
1326 When <span class="emphasis"><em>x</em></span> is large (<span class="emphasis"><em>x</em></span> &gt; 2), both
1327 continued fractions converge (CF1 may be slow for really large <span class="emphasis"><em>x</em></span>).
1328 J<sub>&#956;</sub>, J<sub>&#956;+1</sub>, Y<sub>&#956;</sub>, Y<sub>&#956;+1</sub> can be calculated by
1329 </p>
1330 <p>
1331 <span class="inlinemediaobject"><img src="../../../equations/bessel13.svg"></span>
1332 </p>
1333 <p>
1334 where
1335 </p>
1336 <p>
1337 <span class="inlinemediaobject"><img src="../../../equations/bessel14.svg"></span>
1338 </p>
1339 <p>
1340 J<sub>&#957;</sub> and Y<sub>&#956;</sub> are then calculated using backward (Miller's algorithm) and forward
1341 recurrence respectively.
1342 </p>
1343 <p>
1344 When <span class="emphasis"><em>x</em></span> is small (<span class="emphasis"><em>x</em></span> &lt;= 2), CF2
1345 convergence may fail (but CF1 works very well). The solution here is Temme's
1346 series:
1347 </p>
1348 <p>
1349 <span class="inlinemediaobject"><img src="../../../equations/bessel15.svg"></span>
1350 </p>
1351 <p>
1352 where
1353 </p>
1354 <p>
1355 <span class="inlinemediaobject"><img src="../../../equations/bessel16.svg"></span>
1356 </p>
1357 <p>
1358 g<sub>k</sub> &#160; and h<sub>k</sub> &#160;
1359 are also computed by recursions (involving gamma functions), but
1360 the formulas are a little complicated, readers are refered to N.M. Temme,
1361 <span class="emphasis"><em>On the numerical evaluation of the ordinary Bessel function of
1362 the second kind</em></span>, Journal of Computational Physics, vol 21, 343
1363 (1976). Note Temme's series converge only for |&#956;| &lt;= 1/2.
1364 </p>
1365 <p>
1366 As the previous case, Y<sub>&#957;</sub> &#160; is calculated from the forward recurrence, so is Y<sub>&#957;+1</sub>.
1367 With these two values and f<sub>&#957;</sub>, the Wronskian yields J<sub>&#957;</sub>(x) directly without backward
1368 recurrence.
1369 </p>
1370 </div>
1371 <table xmlns:rev="http://www.cs.rpi.edu/~gregod/boost/tools/doc/revision" width="100%"><tr>
1372 <td align="left"></td>
1373 <td align="right"><div class="copyright-footer">Copyright &#169; 2006-2010, 2012-2014 Nikhar Agrawal,
1374 Anton Bikineev, Paul A. Bristow, Marco Guazzone, Christopher Kormanyos, Hubert
1375 Holin, Bruno Lalande, John Maddock, Jeremy Murphy, Johan R&#229;de, Gautam Sewani,
1376 Benjamin Sobotta, Thijs van den Berg, Daryle Walker and Xiaogang Zhang<p>
1377 Distributed under the Boost Software License, Version 1.0. (See accompanying
1378 file LICENSE_1_0.txt or copy at <a href="http://www.boost.org/LICENSE_1_0.txt" target="_top">http://www.boost.org/LICENSE_1_0.txt</a>)
1379 </p>
1380 </div></td>
1381 </tr></table>
1382 <hr>
1383 <div class="spirit-nav">
1384 <a accesskey="p" href="bessel_over.html"><img src="../../../../../../doc/src/images/prev.png" alt="Prev"></a><a accesskey="u" href="../bessel.html"><img src="../../../../../../doc/src/images/up.png" alt="Up"></a><a accesskey="h" href="../../index.html"><img src="../../../../../../doc/src/images/home.png" alt="Home"></a><a accesskey="n" href="bessel_root.html"><img src="../../../../../../doc/src/images/next.png" alt="Next"></a>
1385 </div>
1386 </body>
1387 </html>