From: Dr. Edmund Weitz
Subject: Re: Windows LISP Compiler
Date:
Message-ID: <3d228vzz.fsf@agharta.de>
··········@hotmail.com (Nonzero) writes:
> Hello all,
>
> What free windows LISP compiler would you guys recommend and where can I find it?
>
> Thanks in advance
> -Nonzero314
<http://groups.google.com/groups?selm=m3y9mzeh2u.fsf%40bird.agharta.de>
Nonzero <··········@hotmail.com> wrote:
> Hello all,
>
> What free windows LISP compiler would you guys recommend and where can I
find it?
-I would recommend Allegro Common Lisp from Franz. Inc.
The compiler and IDE (which is really intelligent, because Enter and Return
is the same) is not absolutely free (as a student you must update your 60
day trial license again every 2 months).
-The Lisp compiler from Xanalysis is also free for non commercial use. But
it is more tied to Emacs (whereas AL CL is not). And it is a tad behind the
AL CL concerning the performance for numerical calculations.
-Corman Lisp (which is Shareware) has got also a very good programming
environment. But it is not appropiate for number crunching.
But only Allegro Common Lisp is really a horse, when it comes to numerical
calculations.
S. Gonzi
In article <·················@hagakure.utanet.at>,
Siegfried Gonzi <···············@kfunigraz.ac.at> wrote:
>
>
>But only Allegro Common Lisp is really a horse, when it comes to numerical
>calculations.
What exactly are you saying? Is Allegro sucky when it
comes to numerical calculations? One doesn't usually
associate horses with great calculating prowess.
Neigh, let me take that back. There is record of some
equine calculation in certain vaudeville acts of yore,
but unless I'm mistaken, even those horses rarely
mastered floating point.
--d
From: "Dorai Sitaram" <····@goldshoe.gte.com>
> In article <·················@hagakure.utanet.at>,
> Siegfried Gonzi <···············@kfunigraz.ac.at> wrote:
> >
> >
> >But only Allegro Common Lisp is really a horse, when it comes to
numerical
> >calculations.
>
> What exactly are you saying? Is Allegro sucky when it
> comes to numerical calculations? One doesn't usually
> associate horses with great calculating prowess.
We often use terms in our everyday usage, which might not actually define an
appropiate meaning. Instead of horse you could use rocket but "horse power"
is not bad? ("Pferdekraft" in German)
> Neigh, let me take that back. There is record of some
> equine calculation in certain vaudeville acts of yore,
> but unless I'm mistaken, even those horses rarely
> mastered floating point.
Yes, yes a micro benchmark is hardly ever a good measure. In bigger programs
the complexity compensate for performace. I once red an interesting article
in "Computer in Science and Engineering", where they reported a project
where 3 postdocs transfered a 20.000 lines of Fortran code (as far as I can
remember in Chemistry) to C++.
Before the project has commenced everybody had assumed that C++ will have to
work hard in order to gain Fortran performace. But quite the contrary
emerged: the two codes are about the same, when it comes to execution
speeds.
S. Gonzi
Siegfried Gonzi wrote:
>
> We often use terms in our everyday usage, which might not actually define an
> appropiate meaning. Instead of horse you could use rocket but "horse power"
> is not bad? ("Pferdekraft" in German)
Here's a new word for Christmas Day, 2001: "Lispvergnuegen"
From: "Dorai Sitaram" <····@goldshoe.gte.com>
> In article <·················@hagakure.utanet.at>,
> Siegfried Gonzi <···············@kfunigraz.ac.at> wrote:
> >
> >
> >But only Allegro Common Lisp is really a horse, when it comes to
numerical
> >calculations.
>
> What exactly are you saying? Is Allegro sucky when it
> comes to numerical calculations? One doesn't usually
> associate horses with great calculating prowess.
We often use terms in our everyday usage, which might not actually define an
appropiate meaning. Instead of horse you could use rocket but "horse power"
is not bad? ("Pferdekraft" in German)
> Neigh, let me take that back. There is record of some
> equine calculation in certain vaudeville acts of yore,
> but unless I'm mistaken, even those horses rarely
> mastered floating point.
Yes, yes a micro benchmark is hardly ever a good measure. In bigger programs
the complexity compensate for performace. I once red an interesting article
in "Computer in Science and Engineering", where they reported a project
where 3 postdocs transfered a 20.000 lines of Fortran code (as far as I can
remember in Chemistry) to C++.
Before the project has commenced everybody had assumed that C++ will have to
work hard in order to gain Fortran performace. But quite the contrary
emerged: the two codes are about the same, when it comes to execution
speeds.
S. Gonzi