Is a Basic Understanding of Matlab Enough for Applied Numerical Methods?

Click For Summary

Discussion Overview

The discussion revolves around the adequacy of a basic understanding of Matlab for an applied numerical methods course. Participants explore the implications of limited Matlab instruction on future applications in industry or research, as well as the relevance of programming skills in general.

Discussion Character

  • Debate/contested
  • Conceptual clarification
  • Technical explanation

Main Points Raised

  • One participant expresses concern about the lack of extensive Matlab instruction in their numerical methods course and questions its potential impact on future work in industry or research.
  • Another participant argues against the necessity of Matlab, suggesting that the focus should be on understanding numerical techniques rather than specific programming syntax.
  • A different viewpoint emphasizes that basic programming concepts (like loops and conditionals) are sufficient for learning numerical methods, and that deeper programming skills can be acquired independently.
  • Some participants note that many numerical methods can be implemented in various programming environments, such as Python with libraries like scipy and numpy, which may be more beneficial in the long run.
  • There is a suggestion that practical programming skills are more important than familiarity with a specific language like Matlab, as understanding fundamental programming concepts can facilitate learning any language quickly.

Areas of Agreement / Disagreement

Participants express differing opinions on the importance of Matlab in learning numerical methods. Some believe that a basic understanding is sufficient, while others argue for the necessity of more extensive training in Matlab or alternative programming tools. The discussion remains unresolved regarding the best approach to teaching and learning numerical methods in relation to programming skills.

Contextual Notes

Participants highlight that the course may prioritize conceptual understanding over programming proficiency, and there is an acknowledgment of the limitations of Matlab compared to other programming options. The discussion reflects varying perspectives on the role of programming in numerical methods education.

eurekameh
Messages
209
Reaction score
0
So I'm taking an applied numerical methods course this year, and we are not putting any emphasis on Matlab. Well, we do use it on homeworks and assignments, but he does not teach us more than the very basics (for, while loops, if iterations). The weird thing is that all the math we learn in the class can be done on Matlab using these very basic operations.

I know that there is a time limit in which all of this material can be taught, but what I'm wondering is: will this lack of competence in Matlab effect me in any way further down the road (whether it be in industry or research)? Why isn't Matlab taught extensively, since it is what makes numerical methods possible, for lack of a better word?
 
Physics news on Phys.org
Matlab is terrible you should be glad not to be exposed to it. I have heard of classes using Matlab without warning students. That should be illegal. Minor differences in syntax are unimportant to learning ideas. While snapping a rubber band repeatedly on your nose is more productive and less painful, no one is preventing you from using Matlab on the side.
 
I never used MATLAB (well I did once...)and I suspect that this numerical method class is more focus on teaching you the ideas behind numerical tehcniques than how to do it via a program. Besides programs like matlab, mathematica, maple, etc, can all be learned with practice and should be learned if you feel you'll need them, but on your own time.
 
eurekameh said:
he does not teach us more than the very basics (for, while loops, if iterations).

Good for him. Those are all the logical constructons that you actually need, to write programs in any language. The only other "must know" thing about practical computer programming is recursion IMO, but numerical methods and Matlab don't provide much motivation for learning about that.

It's much better to learn and practise the basics of "how to program" thoroughly, than to fill your head with half-remembered and half-understood facts about all the special purpose commands and functions in one particular language. Learn the basics properly, and you will be able to get productive in ANY programming language in a couple of days, as and when you need to learn it.
 
Also note that lots of numerics is done outside of Matlab, and knowing the basics might be more useful in the long run. For example, for a huge percentage of problems, python with scipy/numpy is a viable and easily available alternative. Also most of the hardcore number crunching in science is done with home-made or special purpose programs (because general programs do not offer the required techniques, or offer them at an insufficent level of sophistication).
 

Similar threads

  • · Replies 13 ·
Replies
13
Views
3K
  • · Replies 5 ·
Replies
5
Views
3K
  • · Replies 1 ·
Replies
1
Views
3K
Replies
2
Views
2K
  • · Replies 3 ·
Replies
3
Views
6K
  • · Replies 1 ·
Replies
1
Views
3K
  • · Replies 2 ·
Replies
2
Views
3K
  • · Replies 4 ·
Replies
4
Views
7K
  • · Replies 2 ·
Replies
2
Views
3K
  • · Replies 4 ·
Replies
4
Views
3K