isfast: routines to measure if a given
OpenGL feature is fast or not.
Performance measurement covers issues
like displaying a program over a network or locally, flushing the
graphics pipeline properly, etc. Measuring all features needed by
an app can be time-consuming. Saving performance measurements and
reusing them whenever possible is an advantage. And measuring
operations other than graphics--disk and network throughput,
processing time for a particular set of data, performance on
uniprocessor and MP systems--will also be important.
isfast contains two libraries (and one demo program) which can help
with all the above-mentioned issues:
- libpdb: "Performance DataBase" routines
for measuring execution rates and maintaining a simple database.
- libisfast: set of routines demonstrating
libpdb that answer common questions about performance of OpenGL
features (using reasonable but subjective criteria).
- demo: contains a trivial main program to call
the routines in libisfast.
These libraries can't substitute for comprehensive benchmarking and
performance analysis, and don't replace more sophisticated tools (like
IRIS Performer and IRIS Inventor) that optimize application performance
in a variety of ways. However, they can handle simple tasks easily.