W3C home > Mailing lists > Public > www-svg@w3.org > March 2000

bezier curves and transforms

From: Max Froumentin <masmef@maths.bath.ac.uk>
Date: 26 Mar 2000 16:37:37 +0100
To: www-svg@w3.org
Message-ID: <ptvpushg2ta.fsf@scratchy.maths.bath.ac.uk>
From the svg requirements:

Paths can be made up of any combination of the following:
- [bezier segments]
- No other curve types (Other curve types such as splines or NURBS are either 
  technically very difficult, industry-specific and/or have not established 
  themselves as industry standards as much as beziers)

I know it's probably futile to try and reopen the debate on basic
shapes or paths, but it seems to me that having rational beziers
wouldn't have been so "technically difficult", as they require only
one extra parameter for each control point, and the computations for
displaying them differ very little from non-rational beziers.  The
main advantage is that almost all standard types of parametric curves
can be represented (not approximated) by rational bezier: B-splines,
NURBS, beta splines, you name it. The only problem I can see for paths
is that rational bezier don't necessarily interpolate their end
control points. But I think that would have been a minor issue since
an SVG renderer could always connect loose ends with line segments.

My second point is about transformations: transformations can be
represented by a 3x3 matrix (section 7.4):

[a c e]
[b d f]
[0 0 1] <- why isn't it possible to modify this row? Being able to change
           the leftmost 2 values would allow nice perspective effects on 
           objects, e.g. scrolling text  la Star Wars.


Max Froumentin - http://www.bath.ac.uk/~masmef
Department of Mathematical Sciences
University of Bath
Bath BA2 7AY
United Kingdom
Received on Sunday, 26 March 2000 10:37:45 UTC

This archive was generated by hypermail 2.3.1 : Wednesday, 7 January 2015 15:29:12 UTC