Unable to install wolfram

[Egonz]

Hi

I followed the instructions on how to install wolfram on my rpi, but I get an error stating is unable to locate the package. Is there a specific repository I need to configure?

Thanks!
Egonz

[MoosePi]

Is this the software you have downloaded ?

http://www.wolfram.com/raspberry-pi/?fp=middle
http://www.raspberrypi.org/archives/5282

see http://www.raspberrypi.org/phpBB3/viewt ... 95&t=61489

[Egonz]

Here is the output of each command.

pi@raspberrypi ~ $ sudo apt-get update
Hit http://archive.raspberrypi.org wheezy Release.gpg
Hit http://archive.raspberrypi.org wheezy Release
Hit http://http.debian.net wheezy Release.gpg
Hit http://archive.raspberrypi.org wheezy/main armel Packages
Hit http://http.debian.net wheezy Release
Hit http://http.debian.net wheezy/main armel Packages
Hit http://http.debian.net wheezy/contrib armel Packages
Hit http://http.debian.net wheezy/contrib Translation-en
Hit http://http.debian.net wheezy/non-free armel Packages
Hit http://http.debian.net wheezy/main Translation-en
Ign http://archive.raspberrypi.org wheezy/main Translation-en_GB
Hit http://http.debian.net wheezy/non-free Translation-en
Ign http://archive.raspberrypi.org wheezy/main Translation-en
Reading package lists... Done
pi@raspberrypi ~ $ sudo apt-get install wolfram-engine
Reading package lists... Done
Building dependency tree
Reading state information... Done
E: Unable to locate package wolfram-engine


One thing I noticed from the other post is that I am using the soft float version. Perhaps the package is not available for the soft float version?

pi@raspberrypi ~ $ uname -a
Linux raspberrypi 3.6.11+ #538 PREEMPT Fri Aug 30 20:42:08 BST 2013 armv6l GNU/Linux

[sharix]

This is most probably the issue. Last time I checked there was no soft-float build available.

[dennisw]

I don't know which other post you're referring to.

Try this command:

Code: Select all

ls -ld /lib/arm-linux-gnueabi*

If the result shows a directory name that ends in "hf" then you're running the hard float distribution.

On my Pi, this command:

Code: Select all

cat /etc/apt/sources.list

outputs:

deb http://mirrordirector.raspbian.org/raspbian/ wheezy main contrib non-free rpi

If yours is different, try changing it to that source after backing up the file.

[dom]

One thing I noticed from the other post is that I am using the soft float version. Perhaps the package is not available for the soft float version?

Is there a good reason for using the soft-float raspbian?
Packages that use floating point (quite likely Mathmatica) run significantly faster on hard-float builds.

[sharix]

@dom: I guess some of us have worked with our Pis before the hardFP debian came along and can't be bothered to reinstall and set up everything again. That's why I still have one Pi with the softFP debian on it.

[Egonz]

Thanks for all the help. I appreciated.

I just wanted to play with Mathematica (I have heard great things about it). Although it will be slower in softfp systems, will it still be useable? Any one knows if there are plans for such a release?

[MoosePi]

Thanks for all the help. I appreciated.

I just wanted to play with Mathematica (I have heard great things about it). Although it will be slower in softfp systems, will it still be useable? Any one knows if there are plans for such a release?

Soft Float armel support was dropped quite a while ago, it is not needed now that the latest version of Java run on HF.

[ianstoba]

Is there an easy way to update from armel to armhf? I guess I could just start over with a new raspbian image, but I would rather not lose the work I have done on my Pi so far.

[mahjongg]

just create a second SD-card for this single purpose.

[alexatwri]

Thanks for all the help. I appreciated.

I just wanted to play with Mathematica (I have heard great things about it). Although it will be slower in softfp systems, will it still be useable? Any one knows if there are plans for such a release?

We don't currently have any plans to release a softfp build for the Raspberry Pi. We chose hardfp for this port because of how critical the floating point performance would be for tasks in Mathematica.