Raphael, I'm using a fresh install of squeeze on an old, plain vanilla 32 bit computer with an amd xp2000 processor and an agp nvidia graphics card, which I'm using to check out debian to see if I might like to use it on a new system ( I mostly used matlab on windows machines when I didn't have to pay for it myself. Now I do and so I've been looking for cheaper alternatives). There were no problems using the package manager to install octave after adding one of the squeeze-backports locations to the apt sources list. Then I launched octave and then just typed a=[1 2; 3 4]; b=[5 6;7 8]; a*b. Poof, octave disappeared in a pile of digital dust ( but a+b is ok). Next I fired up octave again and typed a=[1 2; 3 4]; eig(a). Poof again. eig was one of the test scripts which passed your tests so I guess this suggests that the problem is local to my computer but I don't see how. Unless...Before trying octave, I first had problems with scilab (I'm shopping around). When I used a scilab install script which deposited the code in my home directory, I was able to launch the software and perform basic numerical manipulations but the program crashed whenever I tried to plot anything. Their website said something to the effect that the standard nvidia driver used by most linux distributions didn't properly support their graphics implementation (Java binding for OpenGL). However, eig(a) doesn't seem to require any graphics. One final thing, I also used the synaptic package manager to load scilab. It installs fine but scilab promptly vaporizes when you try to launch it from the menu. Do you think that this nvidia graphics issue could also be the cause of my problems with octave? I hope not since that was the main reason that I tried to switch from scilab to octave.
Thanks again in advance for putting up with my grief, Bill -----Original Message----- From: Rafael Laboissiere [mailto:[email protected]] Sent: Thursday, January 31, 2013 9:42 AM To: whludescher Cc: [email protected] Subject: Re: [Pkg-octave-devel] Bug#699300: Octave crashes with Matrix multiply Bill, First of all, keep [email protected] in the Cc list, such that this discussion is recorded in the BTS. I am doing this and I hope you do not mind that I copied your private message here. As regards your question: * whludescher <[email protected]> [2013-01-31 07:46]: > > Yes I am running squeeze so I loaded up Octave 3.6.2 as per your > suggestion, again using the synaptic package manager . Unfortunately, > I have the same problem with matrix multiply. Is this a known bug or > must it be something unique to my system. I can't believe that the > octave community would tolerate such a problem. This is weird. I have never tested the backported 3.6.2 version of the octave package in a squeeze system. However, the build daemon succeeded to build the package and passed all the unit tests: https://buildd.debian.org/status/fetch.php?pkg=octave&arch=i386&ver=3.6.2-5~ bpo60%2B1&stamp=1350987957 (look for "test scripts" in the page above). I would say that there is indeed something wrong with your system, but I am not sure. At any rate, without more precise information from you, we are not able to even imagine the origin of this bug. Rafael -- To UNSUBSCRIBE, email to [email protected] with a subject of "unsubscribe". Trouble? Contact [email protected]
