Re: chosing a CAS
- From: clicliclic@xxxxxxxxxx
- Date: Thu, 25 Sep 2008 11:22:22 -0700 (PDT)
jadoul.mic...@xxxxxxxx schrieb:
I forgot to say that I work on Windows ...
I looked at some alternative CAS but I did not find convincing
solutions to work under Windows.
For example, I browsed and wandered to dowload the Axiom for windows
and could not find a working download.
For Sage, I even didn't dare an install considering the size of the
install (654 MB) and the lack of further information about what will
happen when I install it. (I should be careful: I have 1 computer for
my business and my hobby!)
Did I miss something?
What would be your best advice?
The enormous size of Sage must be a result of this being a bundle of a
number of free open-source CA systems with a common programming
languageand user interface. Plus, there being no native Windows
version so far, they have to bring their own operating environment. I
am not sure whether the subsystems follow the latest development level
of the originals and what reference systems are used in the case of
forked developments (e.g. is Axiom, OpenAxiom, or FriCAS served?). If
any true development is done, does it also appear in the reference
systems? In any case, I think it unlikely that you need more than one
component for your project.
I suggest you download and try either Maxima or the Axiom fork FriCAS.
Serious development is being done on those. I believe the download
size for compiled Windows versions is something between 20 and 50 MB.
This may be partly due to (inflated?) documention or the fact that the
Windows versions again have to bring their own operating environment,
as the development for either system is done on Linux, I believe.
Between the two systems, the handling of objects differs noticeably:
Maxima is "weakly typed" while Axiom relatives are "strongly typed",
which may be unfamiliar to you. The latter, however, are excellent at
finding elementary antiderivatives of elementary functions (Risch
integrator), and FriCAS can solve systems of polynomial equations via
Groebner basis calculation.
By comparison, the size of the Windows version for the (discontinued)
low-cost commercial system Derive is about 2 MB for version 4.07 and
about 8 MB for version 6.10. Derive too would suffice for your
purpose, I think.
In all the system mentioned here, the definition of Poisson (or
Lagrange) brackets, which seem to resemble a Jacobian determinant, are
merely one- or two-liners. What is the best system for your purpose
should be determined by what capabilities you need beyond this level:
Do you need to handle (simplify, differentiate, etc.) special
functions? Do you need symbolic integration? Do you need to solve
ordinary (or partial?) differential equations?
You may get a feeling of the relative strengths and weaknesses of the
systems mentioned by looking at Michael Westers review "A Critique of
the Mathematical Abilities of CA Systems" (1999) on the web.
Martin.
.
- Follow-Ups:
- Re: chosing a CAS
- From: rjf
- Re: chosing a CAS
- References:
- chosing a CAS
- From: lalbatros
- Re: chosing a CAS
- From: jadoul . michel
- chosing a CAS
- Prev by Date: Re: chosing a CAS
- Next by Date: Re: chosing a CAS
- Previous by thread: Re: chosing a CAS
- Next by thread: Re: chosing a CAS
- Index(es):
Relevant Pages
|
