|
Need help answering general spline modelling questions.
spline modeling is like poly modeling, in that you're essentially making faces still, because you need areas with 3 or 4 points for it to become an actual surface... but where it is different is you control the smoothing through the bezier handles on the spline object rather than the position, proximity, and number of vertices on a mesh to control smoothing.
some advantages are that you never have to worry about geometry that doesn't have the right vertex arrangement to have a good-looking meshsmooth on it... some users might feel it's easier to create curved surfaces using spline cages as well
the disadvantages outweigh it IMO, because first off, it is very buggy...
you create a section of splines that look completely correct, only to find parts of the model are duplicated and inverted, leaving hours of cleanup to do, when just creating it with polymodeling wouldn't have that problem...
complex objects can get tedious at times, because of the sheer number of vertices that cannot be welded properly that are on top of eachother... so you can possibly have as many as 6 vertices on top of eachother you've got to keep track of...
if you're trying to line up objects, the curvature has to be exactly the same as the other object, or you'll notice seams
sometimes the surface is calculated in a peculiar manner, and is very hard to correct
personally I think poly modeling is the easiest and quickest form of modeling out there...
the 'order' of modeling methods from worst to greatest as I see them would probably fall something like this...
5.) nurbs
4.) box modeling
3.) extreme low poly modeling, then meshsmooth control levels to alter final shape
2.) spline modeling
1.) poly modeling
|