Wear, Mark (mark.wear++at++lmco.com)
Wed, 25 Aug 1999 07:25:37 -0600
BACKGROUND:
I have an application that uses a mix of Vega and Performer to produce a
large-scale, texture-intense visual simulation. Four channels are being
drawn. The application is currently running on an 8 processor, 1 pipe Onyx2
IR with 2 RMs and 1GB of memory. It is desirable for the simulation to run
at 30 Hz. I am using Vega API to configure the multiprocess mode and to
manage the frame rate.
PROBLEM:
When I limit the frame rate to 30Hz, the simulation runs 30 Hz with frequent
drops to 20 Hz. However, when I limit the frame rate to 60 Hz, the
simulation runs 30 Hz with occasional jumps to 60 Hz (during roll). If the
simulation can run 30 Hz when asking for 60 Hz, why can it not run 30 Hz
when limiting to 30 Hz? What is going on here?
When I ask for 60 Hz, the simulation is much smoother, but 'glitches' in the
scene due to oscillating between 60Hz and 30 Hz are equally as distracting
as 'glitches' due to oscillation between 30Hz and 20Hz.
If I can not find a solution that enables me to maintain and limit my frame
rate to 30 Hz using the API, does anyone have a suggestion as to how to
detect/predict and prevent a frame under-run?
Thanks,
Mark
Mark E. Wear
Senior Electronic Systems Engineer:
Real-Time Simulation
Lockheed Martin Missiles and Fire Control
(972) 603-2758
mark.wear++at++lmco.com
This archive was generated by hypermail 2.0b2 on Wed Aug 25 1999 - 06:25:59 PDT