Re: pfGeoSet Size

New Message Reply Date view Thread view Subject view Author view

Jenny Zhao (zhz++at++dandan.engr.sgi.com)
Sat, 17 Apr 1999 01:19:11 -0700


Brian wrote:
> These excerpts are simply making reccomendations as to the best lengths for
> database traversal and culling algorithm efficiency. I don't know of any limits
> to the number of primitives you can send in a pfGeoSet. The pfGeoSet class
> defines the number of primitives as an `int', so send lots if you are not
> worried about performance.

That is right. For example, if a geoset contains many triangles that
cover a large area, not all triangles are visible at the same time. Since
Performer culling is only down to getset level at best, all triangle
will be rendered if any is visible, which is not very efficient.
But there is no reason why you can not draw more than 64 triangles in a tstrip.
A good rule of thumb is "The longer the tstrip is, the faster it gets rendered."

>I think the best length for Onyx2 IR and IR2 is multiples of 8 and for
>Octane/Impact graphics it is 9.

It is multiples of 6 or 12 for IR/IR2. Any tstrip with length longer than
12 is good. The LONGER the BETTER.
Of course, this number is going to change in the new hardware as we try to
optimize the sweet spot for vis sim applications.
The tips from the Programmer's guide, like Brian pointed out, are
mostly referring to the balance between CULL/ISECT efficiency and
rendering speed. There is no hardware or software limitations on tstrip length.

Jenny Zhao zhz++at++engr.sgi.com 650 933-5091 IRIS Performer Manager


New Message Reply Date view Thread view Subject view Author view

This archive was generated by hypermail 2.0b2 on Sat Apr 17 1999 - 01:19:13 PDT

This message has been cleansed for anti-spam protection. Replace '++at++' in any mail addresses with the '@' symbol.