[SciPy-user] exponential integral behaviour near -20

Robert Kern robert.kern@gmail....
Mon Dec 15 23:43:40 CST 2008


On Mon, Dec 15, 2008 at 23:34, David Joyner <wdj@usna.edu> wrote:
>
> ---- Original message ----
>>Date: Mon, 15 Dec 2008 23:02:30 -0600
>>From: "Robert Kern" <robert.kern@gmail.com>
>>Subject: Re: [SciPy-user] exponential integral behaviour near -20
>>To: "SciPy Users List" <scipy-user@scipy.org>
>>
>>On Mon, Dec 15, 2008 at 22:39, David Joyner <wdj@usna.edu> wrote:
>>> Hi:
>>>
>>> A user reported a bug on the sage-support list (Sage, www.sagemath.org, includes scipy) for the exponential
>>> integral. This seems to illustrate the idea:
>>>
>>>>>> import scipy.special, math
>>>>>> -scipy.special.exp1(-complex(19.9999990))
>>> (25615628.4058083-3.1415926535897931j)
>>>>>> -scipy.special.exp1(-complex(19.9999991))
>>> (25615624.582086116+3.1415926535897931j)
>>>
>>> Any ideas on what would cause this?
>>
>>I cannot replicate this on OS X with a recent(ish) SVN scipy compiled
>>with gfortran 4.2.1. This function is implemented in FORTRAN, so the
>>FORTRAN compiler used may be relevant. Can you give us more
>>information about the platform, scipy version, and FORTRAN compiler?
>
> I think all the version information is here (I ran scipy
> from within sage using sage -python):
> http://www.sagemath.org/packages/standard/
> It was run on a machine with a phenom chip
> running amd64 ubuntu 8.10.

Please show me "g95 --version".

-- 
Robert Kern

"I have come to believe that the whole world is an enigma, a harmless
enigma that is made terrible by our own mad attempt to interpret it as
though it had an underlying truth."
  -- Umberto Eco


More information about the SciPy-user mailing list