From bug-request at octave dot org Mon Feb 7 09:48:57 2005 Subject: Re: Inconsistency in complex division by zero From: David Bateman To: "Dmitri A. Sergatskov" CC: bug at octave dot org Date: Mon, 07 Feb 2005 16:50:55 +0100 Dmitri A. Sergatskov wrote: > David Bateman wrote: > >> There is an inconsistency in the division by zero between octave and >> matlab. Consider 1i/0 that returns "NaN - NaNi" for octave and "NaN >> -Infi" for matlab. The Matlab return value makes much more sense that >> the octave one as it treats the real and imaginary parts separately.. >> Tested against octave 2.1.64 and matlab 7.0sp1. >> > > I think the result depends on gcc version used. It may also be related > to the gcc bug: > > http://gcc.gnu.org/bugzilla/show_bug.cgi?id=18902 Perhaps, but in that case the problem should not appear with a gcc 3.3 compiled version of libc... I have version linked against a Mandrake gcc 3.3.1 version, that demonstrates the same problem. However mandrake includes a plethora of patches in their gcc, and so there it is not easy to tell whether Mandrake has included the problematic patch in their gcc 3.3.1 version... Regards David -- David Bateman David dot Bateman at motorola dot com Motorola Labs - Paris +33 1 69 35 48 04 (Ph) Parc Les Algorithmes, Commune de St Aubin +33 1 69 35 77 01 (Fax) 91193 Gif-Sur-Yvette FRANCE The information contained in this communication has been classified as: [x] General Business Information [ ] Motorola Internal Use Only [ ] Motorola Confidential Proprietary ------------------------------------------------------------- Octave is freely available under the terms of the GNU GPL. Octave's home on the web: http://www.octave.org How to fund new projects: http://www.octave.org/funding.html Subscription information: http://www.octave.org/archive.html -------------------------------------------------------------