Message Boards Message Boards

0
|
3200 Views
|
11 Replies
|
7 Total Likes
View groups...
Share
Share this post:

Mathematica support for general ARM chips

Posted 8 months ago

Withe the apple silicon and the new snapdragon elite X it seems that the trend is to move towards arm based chips. I know that mathematica does support apple silicon, but are there any plans to support other arm chips? I mean, like a general version of mathematica for the arm chips in linux and windows machines.

POSTED BY: Felipe Barbosa
11 Replies
Posted 6 months ago

I am interested in this too

POSTED BY: Updating Name

Hi Arnoud,

I am interested in using Mathematica on a Samsung tablet device. My current device is the Tab S9 sporting a Qualcomm Snapdragon 8 gen 2 CPU, which should be plenty compared to the R-Pi, thus performance should not be an issue.

OS wise, I am using termux as the container, I can also can use proot to install other Linux Distro if necessary, although this will hurt the performance due to Proot. Hence, ideally I want to do everything inside termux.

I just wonder whether anyone attempt successfully to install Mathematica R-Pi version on such set up. I am willing to actually participate in debugging efforts, thus if there is already a group attempting it in this community, I would love to join. Consider Samsung tablet sales literally have been doubling gen over gen in the last 3 gens (of the order of 10mils unit per quarter now), I think this is an absolute worthwhile effort!

I will try to start something later today and see what the initial problems are.

Kind regards, Quan Le Thien

POSTED BY: Quan Le Thien

We don't have an official build for Linux-ARM for customers, but we do have a Linux-ARM build for Raspberry Pi: https://www.wolfram.com/raspberry-pi/

Do you have a specific Linux-ARM machine that you are asking this for?

We don't have any plans for a Windows-ARM version at the moment

POSTED BY: Arnoud Buzing
Posted 8 months ago

Hi On a slightly different note,

I have a AMD Ryzen Threadripper CPU. I am wondering whether it is possible for Mathematica to use AMD's Optimizing CPU Libraries (AOCL) instead of MKL. Is there an easy way to make this happen?

POSTED BY: Asim Ansari
Posted 8 months ago

I don't have any machine in particular, I am just looking to buy a new laptop and searching the possibilities.

POSTED BY: Felipe Barbosa

Hi Arnoud,

I am not sure whether you see my other direct reply to OP, really sorry that message was intended as a direct question to you. I never realized that I did not reply directly to you till now.

The good news is that I managed to successfully install the Pi's ARM version of both the Wolfram Engine and the front-end Mathematica to with full GUI working with Termux-X11. Because this is my first time getting involved with the Wolfram Community, I am not sure how to share it as a tutorial. Do you know whether it is appropriate for me to make a tutorial for that on WC? If yes, where should I put that?

Kind regards, Quan Le Thien

POSTED BY: Quan Le Thien

Should be fine to make a Wolfram Community post about it. Glad this worked for you.

POSTED BY: Arnoud Buzing
Posted 6 months ago

Any chance you could write up some notes on this endeavor for other folks trying to accomplish the same objective?

POSTED BY: Mma Usr
Posted 6 months ago

Please!

POSTED BY: Codrut Popescu
Posted 5 months ago

Hi again,

I have a M1 Mac and want to install Fedora Asahi on it, is this Raspberry Pi version expected to work?

POSTED BY: Felipe Barbosa
Posted 5 months ago

Unlikely ... or at least not without a lot of effort on your part. There is a post on the web somewhere about someone managing to accomplish something similar for a non-Raspbian distribution on their Pi.

It would be helpful if Wolfram would start packaging the ARM version of Mathematica as a flatpak or a snap so that users could install it across different distributions.

POSTED BY: Mma Usr
Reply to this discussion
Community posts can be styled and formatted using the Markdown syntax.
Reply Preview
Attachments
Remove
or Discard

Group Abstract Group Abstract