> From: "Andrew R. Thomas-Cramer" <[EMAIL PROTECTED]>
>
> I can confirm this approach works. I implemented it to display a surface graph
> with both colors and wire-frame.
> To be more specific about the Z value, you'd need to set the polygon offset
for
> the wire-frame _less_ than that for the filled polygons. For the wire-frame's
> appearance:
>
> PolygonAttributes attrPolygon = new PolygonAttributes();
> attrPolygon.setPolygonMode( PolygonAttributes.POLYGON_LINE );
> attrPolygon.setPolygonOffset( -1 ); // Nudge toward viewer.
>
> Also, I agree with Mark. We shouldn't have to have two separate Shape3Ds to
> produce a single visual object that's both filled and wire-framed. Memory
> requirements might be drastically reduced if a single Shape3D could be both
> filled and wire-framed.
If both Shape3Ds use the same Geometry, then the memory requirements are pretty
small-- just the different attributes. If you need indiviual control over the
edges, then you'll need to use a separate Geometry (i.e. a LineStripArray).
Earlier APIs have had individual polygon edge flags which the application could
use to control the edges. These saved memory, but my experience was that these
were much slower than storing and drawing the edges as lines. A classic
memory/speed tradeoff.
> However, wouldn't a little more be required than just changing the polygon
mode
> to a bitwise-OR? How would the color of the wire-frame be specified
> independently of the color of the fill? (If they had the same coloring,
there'd
> be no point -- or rather, no lines. :P )
Exactly. By using two shapes you have control over the appearance of both
primitives. If edges were built in to a single Shape then there would have to
be separate edge attributes.
Doug
===========================================================================
To unsubscribe, send email to [EMAIL PROTECTED] and include in the body
of the message "signoff JAVA3D-INTEREST". For general help, send email to
[EMAIL PROTECTED] and include in the body of the message "help".
