MathGroup Archive 1998

[Date Index] [Thread Index] [Author Index]

Search the Archive

RE: Re: Introducing: Conix 3D Explorer!


  • To: mathgroup@smc.vnet.net
  • Subject: [mg11577] RE: [mg11540] Re: Introducing: Conix 3D Explorer!
  • From: "Barthelet, Luc" <lucb@ea.com>
  • Date: Tue, 17 Mar 1998 10:42:58 -0500

I am fine with requiring supplemental $ for extra features. So I would 
pay more to get Conix Explorer integrated in the front end.

But here is a performance issue:
Currently formatting and  transferring information from the kernel to 
Conix Explorer or any other mathlink program for that matter is 
fundamentally slow.
If WRI integrates well Conix explorer in Mathematica in the future we
will get  a tremendous performance improvement in rendering 3D graphics
and  manipulating them.
Also, the Graphics features in Mathematica are fine for the way they are
being  rendered in the front end, but do not have the consistency of
the rest  of the product. They were never challenged before, now they
need an  overall.
FullOptions for example is really deficient for 3D objects compared to 
2D objects.
Graphics3D primitive do not support texture mapping, smoothing, 
anti-aliasing....
As Mathematica is used as more than a math tool, but as a "fully 
integrated technical computing system" it is time to catch up on 3D 
graphics and animations.
So congratulation to WRI for enabling tools like Conix Explorer. Now 
let's get to the next level and provide the power we can get from a 
well integrated tool.

Thanks,

Luc Barthelet
General Manager
Maxis
"Where SimCity is built."


-----Original Message-----
From:	weber@math.uni-bonn.de [SMTP:weber@math.uni-bonn.de] To:
mathgroup@smc.vnet.net
Sent:	Saturday, March 14, 1998 10:56 AM To:	mathgroup@smc.vnet.net
Subject:	[mg11540] Re: Introducing: Conix 3D Explorer!

In article <6eacvd$o2e$9@dragonfly.wolfram.com>, SChandler
<SChandler@uh.edu> wrote:

> I have been using this product for about a week and agree 
completely.
> This is
> a terrific add on for Mathematica. It has so far proved quite 
reliable
> on my NT platform and the documentation is excellent. The program 
has a
> Mathematica flavor -- it is not just some port from existing 
graphics
code. Conix
> has answered my tech support questions promptly and fully. I agree 
with
> Mr. Barthelet that Wolfram should investigate integrating the
> capabilities of this program into its front end, although there may
> be some virtue in having pluralist graphics models that suit 
differing needs.
>
>

I can only agree concerning the quality of the GLExplorer.  I have
$DisplayFunction=GLShow in my init file which allows me to run a much
smaller front end now, because the explorer renders with less memory
(and faster).  The graphics are more flexible, and interactive.  Enough
of praise.

However, I do not agree that Wolfram should integrate the explorer  into
Mathematica. This would make Mathematica (3.5? 4.0?) way more  expensive
again (at least if Conix charges them the right price). There are
certainly many people who would say 'who cares?',  but there are also
other people.
Our department had not enough money to buy all the nice math software
available, so we decided to buy only a small number of Mathematica
licenses -- not enough to use it in classes. Selling a product and
counting on that there are enough people who just have to buy it is
very similar to the pricing philosophy of scientific journals. But 
(not
only) Wolfram should be aware of the fact that a lot of input for
mathmatical improvements comes from academia (which has a relatively
low budget).

It would be very helpful for us
if Mathematica came as a robust kernel & front end with many
'enhancements' which one could buy when needed. It is certainly true
that one of Mathematica's advantages lies in the fact that it  combines
all it abilities into one huge environment. But, for instance, I am  not
using the statistical functions or the sound capabilities. So there
could be a very small (cheap) version of which one could buy lots of
licenses, and smaller number of licenses for the add-ons. It would 
also
help to develop Mathematica more safely. The current version has so many
bugs that I sometimes have to switch back to Mathematica 2.2 to get the
correct result. I would really like to have the old Integrate funtion
somewhere in Mathematica 3.0 as "Integrate2", just to be on  the
save side.

So please, keep Mathematica small and reliable and don't charge the 
user
for too fancy and buggy features.

Matthias Weber




  • Prev by Date: RE: Automatic generation of piecewise functions
  • Next by Date: Re: commutativity
  • Prev by thread: Re: Introducing: Conix 3D Explorer!
  • Next by thread: Implicit differentiation