plonski@aeroaero.org (Mike Plonski) (08/11/90)
Peter Hallett, of Precision Visuals, states that: It is incorrent to state that PV~WAVE is a shell that sits on top of IDL. Essentially, PV~WAVE is a superset of IDL. Currently, IDL procedures and applications will run using PV~WAVE. My question is: Does PV-Wave extract some sort of performance penalty (in terms of speed) by riding on top of IDL. One of my friends compared IDL and PV-Wave several years back and found that IDL seemed to run much faster than PV-Wave so they went with IDL. Is this still true? Does PV-Wave cost me anything relative to IDL (except for more $)? Are there things I can do in IDL, but not in PV-Wave? My target platform is a Sun 4 workstation. ------------------------------------------------------------------------------- . . .__. The opinions expressed herein are soley |\./| !__! Michael Plonski those of the author and do not represent | | | "plonski@aero.org" those of The Aerospace Corporation. _______________________________________________________________________________
sterner@warper.jhuapl.edu (Ray Sterner) (08/14/90)
plonski@aeroaero.org (Mike Plonski) writes: >My question is: Does PV-Wave extract some sort of performance penalty >(in terms of speed) by riding on top of IDL. One of my friends >compared IDL and PV-Wave several years back and found that IDL >seemed to run much faster than PV-Wave so they went with IDL. >Is this still true? Does PV-Wave cost me anything relative to >IDL (except for more $)? Are there things I can do in IDL, >but not in PV-Wave? My target platform is a Sun 4 workstation. Can somebody post a short IDL benchmark routine that we can time? It would be very interesting to know the speeds of various workstations, operating systems, and languages. The benchmark should be run on workstations with no other jobs (except what is normally running) on the system. The IDL function systime() might be useful for timing various stages of the routine.