Matlab vs mathematica

  • Mathematica
  • Thread starter Ai52487963
  • Start date
  • #1
115
0
I'm trying to have a math program on my computer for math-related things. I've been attempting to use matlab for a while now, but its far too confusing for my tastes. Is mathematica any easier to use?
 

Answers and Replies

  • #2
39
0
like any program you will need to take time to get use to the notations but i found it to be easier than matlab. also mathematica has an extensive help library that you can use if you forget
 
  • #3
329
1
Mathematica is easier to use. Like the Midy said, it'll take time to get use to the notation, but it's pretty good at telling you what you did wrong.
 
  • #4
uart
Science Advisor
2,776
9
Asking the question of matlab vs mathematica is a little like asking the question of which one should I choose between Microsoft Word 10 and Motocross Madness 2. The simple answer is that they are different programs intended for different purposes and not directly comparable.

Mathematica is a algebra and symbolic maths package whereas matlab is predominately a numerical computation package (though you can get a symbolic toolkit for matlab which then gives it some of the features of a program like Maple or Mathematica). Generally speaking though you'd use Matlab if you want to be able to manipulate lots of numerical data easily and you'd use mathematica if you want an aid for symbolic mathematical manipulations. Engineers for example would mostly tend to find Matlab more useful than Mathematica, while it would normally be the other way around for a Mathematician.
 
Last edited:
  • #5
vanesch
Staff Emeritus
Science Advisor
Gold Member
5,028
16
I use both, and as pointed out previously, they have different scopes of application. Matlab is much more efficient when it comes to intensive numerical calculation. However, I find Mathematica much richer, and even when I have to do not-too-intensive numerical work, I prefer Mathematica. But for the big number crunching, Mathematica is simply not efficient enough with computer ressources.

Mathematica has however, a longer "learning curve" IMO. But once you master it, it is terribly powerful.

If it is a matter of budget, I'd go for mathematica, because there's no clone available, while there are good free clones of matlab (scilab for instance).
 
  • #6
D H
Staff Emeritus
Science Advisor
Insights Author
15,393
684
If it is a matter of budget, I'd go for mathematica, because there's no clone available, while there are good free clones of matlab (scilab for instance).
There is Maxima (SourceForge link), based on a 1982 version of Macsyma. Macysyma is the granddaddy of Mathematica and Maple. Sorry, I haven't tried it.

Matlab provides a symbolic toolkit powered by the Maple engine. Unfortunately, the integration is not clean (and that is saying it nicely). Do the Matlab clones provide a symbolic toolkit? To date I have not been impressed by the Matlab clones.
 
  • #7
486
0
There is Maxima (SourceForge link), based on a 1982 version of Macsyma. Macysyma is the granddaddy of Mathematica and Maple. Sorry, I haven't tried it.

Matlab provides a symbolic toolkit powered by the Maple engine. Unfortunately, the integration is not clean (and that is saying it nicely). Do the Matlab clones provide a symbolic toolkit? To date I have not been impressed by the Matlab clones.
I have tried maxima before. It does give you a feel of what symbolic package is. However it feature is nowhere near maple or mathematica. I personally prefer mathematica because it is TOO POWERFUL compare to anyhting else on earth
 
  • #8
Chris Hillman
Science Advisor
2,345
8
Sage

I've tried Maxima and it can perform various tasks quite well, but in general the user interface is much less well developed than what the OP might be used to, and the power lags far behind Mathematica and Maple. The syntax of Maxima is much closer to Maple than Mathematica (a good thing, since most observers agree that Mathematica syntax is not only harder to learn, it leads to some real problems in coding).

Matlab, Mathematica, and Maple are all highly developed, incredibly powerful and useful environments for various types of symbolic and numerical computations. As already noted, they have different and to some extent complementary strengths and weaknesses. For example, I generally prefer Maple to Mathematica, but there is no denying that Mathematica produces prettier print-quality graphics!

It might be worth mentioning for benefit of lurkers that in addition to Maxima, which is a general symbolic computation environment and which is available for free, there are quite a few specialized packages like GAP and Macaulay 2 which are free, easy to install, and extremely powerful. Noone would say that GAP syntax is particularly enchanting, but it is standard in computational algebra due to its power and the care with which it has been developed over so many years. Macaulay 2 has a syntax I like so much that I recommend it to CS students just to see an example of syntax which mathematicians like.

http://modular.math.washington.edu/sage/index.html [Broken] is an open source initiative led by http://modular.math.washington.edu/ [Broken] (Mathematics, University of Washington) and funded by various leading institutions in the mathematical sciences, but it is hugely ambitious so it always needs more developers! SAGE aims to ultimately provide a user interface comparable to Maple which enables users to run commands in Maple, Mathematica, MATLAB, MuPAD, Magma, Axiom, GAP, GP/PARI, Macaulay2, Maxima, Octave, and Singular, while freely using python scripts. See http://modular.math.washington.edu/sage/components.html [Broken] for a more complete list. As I understand it, there is some hope that development of SAGE might revive development of Maxima, which currently lags far behind Maple by virtue of lacking such powerful commands as casesplit and such useful packages as Groebner. I think this initiative is extremely promising; if it succeeds it will fundamentally alter how mathematicians work and play (for the better, I think). To mention just one point, working mathematicians are probably well aware, for example, of the desirability of facile checking of Mathematica and Maple results against each other.

Anyone interested in making financial contributions or contributing to code can http://modular.math.washington.edu/sage/ack.html [Broken]. I myself am unaffiliated with SAGE but I am seriously interested in trying to set up a "Sage Wiki" somewhat like Dispersive Wiki, with the goals of
  • providing a discussion forum for SAGE developers around the world,
  • providing tutorials for anyone interested in learning to use components of SAGE such as Maple, GAP, etc.,
  • promoting the project to the world at large.
Note that SAGE-2.8.12 is http://modular.math.washington.edu/sage/announce/sage-2.8.12.txt [Broken] for installation and beta testing.

(Confusingly, there is a http://www.npl.washington.edu/npl/ar96/ch2_9.html [Broken] also called SAGE and also run at UW, and there is a well known grant program also called SAGE and also admininstered at UW. All three programs are distinct!)

BTW, a relevant conference ("camp meeting"?) which starts tommorrow (!!!!) is here.
 
Last edited by a moderator:
  • #9
2,063
2
http://modular.math.washington.edu/sage/index.html [Broken] for a more complete list. As I understand it, there is some hope that development of SAGE might revive development of Maxima, which currently lags far behind Maple by virtue of lacking such powerful commands as casesplit and such useful packages as Groebner. I think this initiative is extremely promising; if it succeeds it will fundamentally alter how mathematicians work and play (for the better, I think). To mention just one point, working mathematicians are probably well aware, for example, of the desirability of facile checking of Mathematica and Maple results against each other.

Anyone interested in making financial contributions or contributing to code can http://modular.math.washington.edu/sage/ack.html [Broken], with the goals of
  • providing a discussion forum for SAGE developers around the world,
  • providing tutorials for anyone interested in learning to use components of SAGE such as Maple, GAP, etc.,
  • promoting the project to the world at large.
Note that SAGE-2.8.12 is http://modular.math.washington.edu/sage/announce/sage-2.8.12.txt [Broken] for installation and beta testing.

(Confusingly, there is a http://www.npl.washington.edu/npl/ar96/ch2_9.html [Broken] also called SAGE and also admininstered at UW. All three programs are distinct!)

BTW, a relevant conference ("camp meeting"?) which starts tommorrow (!!!!) is here.
It has made it to /., so everyone will know about it now. :biggrin: The site and its mirrors are quite slow, so I haven't read much about it. If anyone's interested, I managed to reach to the download page (32-bit Linux).
 
Last edited by a moderator:
  • #10
67
0
When I'm working on maths I have Matlab, Mathematica, and Maple all going. It's hard to be quite proficient at all of them so I just use what I know from each. I like mathematica a lot, if I were to choose only one, because the documentation seems the most extensive and accessible. Just my $0.02
 
  • #11
1,860
0
I think it comes down to solely what you want

Symbolic/Analytic - Mathematica
Numerical - Matlab

Don't get one or the other because one is easier to use, get it for the right purpose. People with programming experience would probably say Matlab is easier than Mathematica, while people without might have an easier time with Mathematica. Basically anything you would see in a course, such as Calculus, Linear Algebra, and others of the sort, you can use Mathematica for. But if you want to look at bifurcations of a nonlinear pendulum, for example, you don't (at least I don't) even want to look at Mathematica.
 
  • #13
chroot
Staff Emeritus
Science Advisor
Gold Member
10,226
34
Of course, we haven't even mentioned Mathcad, which is one of my personal favorites. It does almost anything that Mathematica does (at least in terms of what I need), it's pretty, and it's significantly easier to learn than Mathematica.

- Warren
 
  • #14
67
0
holy crap! I just realized how expensive these programs are.

@chroot
thanks I'll try that out
 
  • #15
chroot
Staff Emeritus
Science Advisor
Gold Member
10,226
34
Most of these programs have student editions that are reasonably priced (or about as "reasonable" as a textbook).

- Warren
 
  • #16
mathematica rules. maple is garbage, just as an aside.
 
  • #17
Gib Z
Homework Helper
3,346
5
I've only ever used the trial of Mathematicia because my parents don't reckon its worth $150 :( It was definitely powerful, although because of the Nature of its computations, some symbolic expressions were more complicated than they needed to be. That can be said of any symbolic math program though.
 
  • #18
189
0
Most of these programs have student editions that are reasonably priced (or about as "reasonable" as a textbook).

- Warren
Some of them are pretty asinine with their treatment of "Student Edition" licenses.

For instance, Wolfram donated a student edition of Mathematica 3.0 to a conference I presented at just after I finished my undergrad. My presentation took second or third place, and won the copy of Mathematica 3.0.

Which I then could not register, because (having graduated) I was no longer a student.

Which they balked at registering when I was a grad student, but had a T/A so I wasn't "full time" to them.

Which they assert I cannot use if I am not registered as a full time student.

So it sits, on a shelf, gathering dust.

ETA: I'd love to try Mathematica now, but even as a teacher and a grad student, I'm only eligible for the $1000 license, which is about 5% of my gross salary. I can't justify that when I have no exact pressing reason to need the software. $250 I could justify as a "toy".

Plus, I'm headed toward a primarily Linux home office...if I bought the Windows Mathematica, would they let me switch to Linux, or would I have to buy a new license?
 
Last edited:
  • #19
X=7
10
0
Which I then could not register, because (having graduated) I was no longer a student.
Oh, the irony!:eek:

Another numeric computation one that hasn't been mentioned is the über-cruncher program IDL!:surprised
 
  • #20
30,025
6,414
Die-hard Mathematica fan here. I wouldn't even consider any package without symbolic capabilities. You can just do so much more symbolically than you can if you are constrained to numerical routines.

I always found Maple very unstable, and even more unstable when called through Matlab. But that was years ago and it may be more stable now.
 

Related Threads on Matlab vs mathematica

  • Last Post
Replies
12
Views
1K
Replies
10
Views
20K
Replies
18
Views
138K
Replies
1
Views
3K
Replies
3
Views
2K
Replies
60
Views
163K
  • Last Post
Replies
7
Views
6K
  • Last Post
Replies
2
Views
1K
Replies
1
Views
3K
Top