Matlab toolboxes essential for every physicist

  • Context: MATLAB 
  • Thread starter Thread starter Wrichik Basu
  • Start date Start date
  • Tags Tags
    Matlab Physicist
Click For Summary

Discussion Overview

The discussion revolves around the essential MATLAB toolboxes for physicists, focusing on both theoretical and experimental applications. Participants explore alternatives to MATLAB, including free software options and their respective functionalities.

Discussion Character

  • Exploratory
  • Technical explanation
  • Debate/contested

Main Points Raised

  • Some participants express that the core MATLAB software is sufficient for many applications, questioning the necessity of additional toolboxes.
  • Concerns are raised about the proprietary nature of MATLAB toolboxes, which limits understanding of their underlying algorithms, potentially affecting efficiency and accuracy.
  • Alternatives to MATLAB, such as FreeMat, GNU Octave, and Julia, are discussed, with some participants highlighting their advantages and disadvantages.
  • Julia is noted for its speed and interoperability with other programming languages, although some participants mention its IDE interface is still developing.
  • Specific toolboxes mentioned as useful include image processing, image acquisition, control, signal processing, and data acquisition, with some participants noting that these are typically included in student licenses.
  • One participant mentions the curve fitting toolbox as a useful addition not included in the student version.
  • There is a request for confirmation on the accuracy of links to resources related to JupyterLab, Jupyter Notebook, and Julia Pro.
  • Participants share experiences with the performance of MATLAB clones, noting that some may be slower than MATLAB itself.

Areas of Agreement / Disagreement

Participants express a mix of opinions regarding the necessity and performance of MATLAB toolboxes, with no clear consensus on which toolboxes are essential or the effectiveness of alternatives. The discussion remains unresolved regarding the best software options for physicists.

Contextual Notes

Some participants mention limitations in the performance of alternative software and the need for better tools in the Julia community. There are also references to the proprietary nature of MATLAB toolboxes affecting their use.

Wrichik Basu
Science Advisor
Insights Author
Gold Member
Messages
2,188
Reaction score
2,694
I was looking forward to buy Matlab student license sometime this year. Anyone who has already bought a license knows that a number of toolboxes are available. Here is the site.

I am not buying Simulink at this moment. So, please leave out anything related to Simulink.

Other than the basic software, what are the toolboxes that an experimental/theoretical physicist requires in day to day work?
 
Physics news on Phys.org
At our lab, folks just don’t use the toolboxes as the core is sufficient for must applications. Also the physicists here don’t like the fact that toolbox algorithms are proprietary to MATLAB and as such you can’t see how they really work and that’s important for efficiency, limitations and for accuracy.

There’s a free alternative to Matlab called freemat that works quite well and of course it’s free. Matlab gives away cheap licenses to students in the hope that they will insist that their future employers must buy the full product for work and it’s much more than the student pays. Many engineers use Matlab all the time in their work.

Also, have you checked out Julia? It’s has a flavor of Matlab but the ide interface isn’t there yet. However, the Jupiter notebooks, and Jupiterlab gives you a similar experience. There’s also Juliapro which is the atom-Juno editor packaged configured and packaged with Julia. Julia has an impressive speed advantage over other languages except for C code and can interoperate with other popular data science languages such as python, java, r, fortran and c.

Many physics problems focus on solving differential equations and partial differential equations numerically and plotting the results and much of that capability is already present in the core.
 
  • Like
Likes   Reactions: m4r35n357
jedishrfu said:
Also, have you checked out Julia? It’s has a flavor of Matlab but the ide interface isn’t there yet. However, the Jupiter notebooks, and Jupiterlab gives you a similar experience. There’s also Juliapro which is the atom-Juno editor packaged configured and packaged with Julia. Julia has an impressive speed advantage over other languages except for C code and can interoperate with other popular data science languages such as python, java, r, fortran and c.
Actually, when I first read about MATLAB a couple of years back, I found that they have an app for Android mobile. While for the desktop app one needs a license to do anything, for the mobile app, one gets access to a number of functions. I have been using them until now. I felt that I should buy the license to explore the software further.

However, recently I had the feeling that MATLAB has some shortcomings, though I against complaining about that. Having used java for nearly four years now, I find a lack of proper modularity, something that java supports to a great extent.

I will happily try out the softwares that you have recommended. Just help me out a bit:

JupyterLab: https://jupyterlab.readthedocs.io/en/stable/

Jupyter Notebook: http://jupyter.org

Julia Pro: https://juliacomputing.com/products/juliapro.html

Are these websites correct?

Which of these would you prefer the most? Which one do you use mostly for computations?

Also, I found people writing about SciLab. Do you have any opinions regarding that?
 
Last edited:
I’ve heard the Matlab clones are slower than Matlab. The clones are scilab and octave. I don’t know if freemat is slower but I imagine it is. I’ve used freemat when I needed to do some quick plots or calculations.

For Julia the Jupyter notebook seems to be the most effective ide. Jupiter lab looks good too. It’s based on the Jupyter notebook which is in fact an interactive web page. I’m playing with juliapro right now but feel it’s a bit slow. Juliapro is based on Juno which is a custom version of the atom webbased editor utility and is also sluggish. Some folks use vim and Julia directly. I feel things will change as the Julia community demands better tools.

Julia too is slow initially as it needs to build its environment but once that’s done it runs much faster.
 
  • Like
Likes   Reactions: Wrichik Basu
jedishrfu said:
There’s a free alternative to Matlab called freemat that works quite well and of course it’s free.
And GNU Octave, of course.
 
  • Like
Likes   Reactions: jedishrfu
jedishrfu said:
Yes, I mentioned that in my prior post #4
I was answering #2 and saw #4 after I posted . . . ;)
 
  • Like
Likes   Reactions: jedishrfu
I use and have used MATLAB for work in universities and national labs, both in physics in engineering. The toolboxes I use most are image processing, image acquisition, control, signal processing, and data acquisition. I think almost all of those are included in the student license.
 
The only other one that I use that is not in the student version is the curve fitting toolbox.
 

Similar threads

  • · Replies 10 ·
Replies
10
Views
3K
  • · Replies 5 ·
Replies
5
Views
3K
Replies
22
Views
3K
  • · Replies 3 ·
Replies
3
Views
3K
  • · Replies 17 ·
Replies
17
Views
3K
  • · Replies 4 ·
Replies
4
Views
4K
  • · Replies 12 ·
Replies
12
Views
3K
Replies
1
Views
3K
  • · Replies 1 ·
Replies
1
Views
9K
  • · Replies 7 ·
Replies
7
Views
3K