Information Technology Reference
In-Depth Information
3.2.2 A. Michael Noll
A. Michael Noll's “Gaussian-Quadratic” graphic makes use, in one direction (the
horizontal, viz. Fig.
3.2
), of the Gaussian distribution. The coordinates of vertices in
the horizontal
x
-direction are chosen according to a Gaussian distribution, the most
important alternative to the uniform distribution. The co-ordinates of vertices in ver-
tical direction are calculated in a deterministic way (their values increase quadrati-
cally).
Whereas Nees' design follows a definite, if simple, compositional rule, Noll's is
really basic: one polygon whose points are determined according to two distribu-
tions. It is not unfair to say that this is a simple visualisation of a simple mathemat-
ical process.
3.2.3 Frieder Nake
The same is true of Nake's polygon (Fig.
3.3
). The algorithmic principle behind the
visual rendition is exactly the same as that of Fig.
3.2
: repeatedly choose an
x
- and
a
y
-coordinate, applying distribution functions
F
x
and
F
y
, and draw a straight line
from the previous point to the new point
(x, y)
; let then
(x, y)
take on the role of
the previous point for the next iteration.
In this formulation,
F
x
and
F
y
stand for functional parameters that must be pro-
vided by the artist when his intention is to realise an image by executing the algo-
rithm.
7
Some experience, intuition, or creativity—whatever you prefer—flows into
this choice.
The visual appearance of Nake's polygon may look more complex, a bit more
like a composition. The fact that it owes its look to the simple structure of one poly-
gon, does not show explicitly. At least, it seems to be difficult to visually follow
the one continuous line that constitutes the entire drawing. However, we can clearly
discover the solitary line, when we read the algorithm. The description of the sim-
ple drawing contains more (or other) facts than we see. So the algorithmic structure
may disappear behind the visual appearance even in such a trivial case. Algorithmic
simplicity (happening at the
subface
of the image, its invisible side) may gener-
ate visual complexity (visible
surface
of the image). If this is already happening
in such trivial situations, how much more should we expect a non-transparent re-
lation between simplicity (algorithmic) and complexity (visual) in cases of greater
algorithmic effort?
8
7
Only a few steps must be added to complete the algorithm: a first point must be chosen, the total
number of points for the polygon must be selected, the size of the drawing area is required, and the
drawing instrument must be defined (colour, stroke weight).
8
The digital image, in my view, exists as a double. I call them the
subface
and the
surface
.They
always come together, you cannot have one without the other. The subface is the computer's view,
and since the computer cannot see, it is invisible, but computable. The surface is the observer's
view. It is visible to us.
