Modeling 1: Polygons - An Introduction to Computer Graphics for Artists - page 70

6.2.1.4 Conversion

Conversion is when you already have geometry with the right shape or a similar
shape to what you need, but it is not polygonal, such as sub-division surfaces or
NURBS. When this is the case, the geometry must be converted to polygons before
it can be modifi ed with polygon editing tools. Conversion allows large numbers of
polygons to be added to a scene at once, but artists must be careful how this is done
because conversion creates weak geometry if the process isn't well-planned.
Because conversion is only applicable to non-polygon geometry, it won't be dis-
cussed in this chapter, but in Sect. 14.4.14 .

6.2.1.5 Import

To import geometry, you must have geometry that was made in another application
or an earlier session of the application you are using. Geometry that is brought into
a scene this way can be complete at the time of import, with no editing required. At
the time the geometry was originally made, one of the other methods of polygon
creation would have been used. One reason to import geometry is if your primary
application cannot perform a specifi c operation on the model but another applica-
tion can. In that situation, the model is exported to the other application, edited, and
then returned to the primary application for completion.

6.2.2 Conclusion

The purpose of these geometry creation methods is to quickly arrive at specifi c shapes
that would be diffi cult to describe otherwise. You could literally start with a cube or a
plane and, by adding vertices as they are needed, transform each point until the fi nal
model looks like anything you want it to be. In this way a cube can become a computer,
a tiger, a house, a rock, or anything at all. The problem is that for other shapes it is dif-
fi cult to position each point exactly correctly. If you want to make a sphere out of a cube
by adding and repositioning points, it would be very diffi cult to build the model without
making dents or small bumps all over its surface. This is why you would use a sphere
geometric primitive to make a sphere instead of building it out of something else.
Most objects are a product of a number of different types of shapes mixed
together. A pen is not really a simple cylinder, but it is similar to a cylinder. Some
CG artists might build a pen out of a cylinder primitive for this reason. They would
start with the cylinder, then add points and modify them until it looked right. Even
more likely, an artist would use a revolve tool. This type of tool will take a profi le
shape, turn it a specifi ed number of times around a specifi ed axis, and create faces
to connect each of the segments. With this tool, a model of most pens could be made
very quickly and accurately. For those that have some asymmetrical detail, or a
detail that cannot be built as part of the revolve operation, another object may be
attached, like a pen clip, if the pen has one.