<html>
<head>
<meta content="text/html; charset=windows-1252"
http-equiv="Content-Type">
</head>
<body bgcolor="#FFFFFF" text="#000000">
<p>Hello, <br>
</p>
Le 30/03/2017 à 13:31, Pierre Haessig a écrit :<br>
<blockquote
cite="mid:0db9ba2d-cb97-821e-62e5-b1b922c785a8@crans.org"
type="cite">
<meta http-equiv="content-type" content="text/html;
charset=windows-1252">
[....]
<p>But how come Julia is 4-5x faster since Numpy uses C
implementation for the entire process ? (Mersenne Twister ->
uniform double -> Box-Muller transform to get a Gaussian
<a moz-do-not-send="true" class="moz-txt-link-freetext"
href="https://github.com/numpy/numpy/blob/master/numpy/random/mtrand/randomkit.c">https://github.com/numpy/numpy/blob/master/numpy/random/mtrand/randomkit.c</a>).
Also I noticed that Julia uses a different algorithm (<span
class="pl-c">Ziggurat Method from Marsaglia and Tsang</span> ,
<a moz-do-not-send="true" class="moz-txt-link-freetext"
href="https://github.com/JuliaLang/julia/blob/master/base/random.jl#L700">https://github.com/JuliaLang/julia/blob/master/base/random.jl#L700</a>)
but this doesn't explain the difference for uniform rng.<br>
</p>
</blockquote>
Any ideas?<br>
<p>Do you think Stackoverflow would be a better place for my
question?</p>
<p>best,</p>
<p>Pierre</p>
</body>
</html>