Juan Ramon Saenz-Diez (yuri++at++metacore.es)
Tue, 20 Feb 96 10:53:24 -0100
I'm comparing the performance of two very simple database programs, one in
Performer and one in OpenGL (High Impact, PF 2.0, IRIX 5.3). In both cases
I ask for a GL display list compilation, to make some 1800 textured
triangles.
The question is that I get the same frame rate, but gr_osview reports a much
higher load on the CPU in the case of Performer (+-50% vs +-10%). The pf
hierarchy is so simple that I believe I'm push-mult-popping the same number
of times per frame. I need the whole CPU for my next project in Performer,
as I would have if I chose OGL.
So, what could be a possible reason for this? -- OK, you don't have the code to
see, but trust me ;-), its a silly program -- How can I more deeply check
what is happening in the CPU and how much hw graphics cache am I occupying?
Many thanks for any suggestion.
Juan R."Yuri" Saenz-Diez, MetaCore Visual Computing, Madrid, Spain.
yuri++at++metacore.es
This archive was generated by hypermail 2.0b2 on Mon Aug 10 1998 - 17:52:26 PDT