Hi,
I'm using vpython in computational simulation. A little concerned about
the efficiency, I ran the following tests:
from visual import *
import profile
import datetime
def dostuff(n):
for i in range(n):
for i in range(n):
pass
timea = datetime.datetime.today()
profile.run("dostuff(1500)")
timeb = datetime.datetime.today()
t = timeb - timea
print "Actual running time: %i.%i seconds" % (t.seconds, t.microseconds)
ball = sphere(pos=(0,0,0), radius=1)
timea = datetime.datetime.today()
profile.run("dostuff(1500)")
timeb = datetime.datetime.today()
t = timeb - timea
print "Actual running time: %i.%i seconds" % (t.seconds, t.microseconds)
And I got this output:
Visual-2003-10-05
3 function calls in 1.250 CPU seconds
Ordered by: standard name
ncalls tottime percall cumtime percall filename:lineno(function)
1 0.000 0.000 1.240 1.240 <string>:1(?)
1 0.010 0.010 1.250 1.250 profile:0(dostuff(1500))
0 0.000 0.000 profile:0(profiler)
1 1.240 1.240 1.240 1.240 visualanalysis.py:5(dostuff)
Actual running time: 1.338290 seconds
Xlib: extension "XFree86-DRI" missing on display ":0.0".
3 function calls in 2.040 CPU seconds
Ordered by: standard name
ncalls tottime percall cumtime percall filename:lineno(function)
1 0.000 0.000 2.040 2.040 <string>:1(?)
1 0.000 0.000 2.040 2.040 profile:0(dostuff(1500))
0 0.000 0.000 profile:0(profiler)
1 2.040 2.040 2.040 2.040 visualanalysis.py:5(dostuff)
Actual running time: 10.847798 seconds
This in my view is pretty poor efficiency, considering that visual
python is doing nothing but displaying a static ball on the second call
to dostuff(). Is there anyway to tell visual python to lay off and do
nothing for a while? So... in the context of my work, when I call the
functions that do all the hard work, i would tell visual python to do
nothing, and then when the functions return, i'd tell visual python to
start doing stuff again. BTW, what is it doing that slows everything
down so much?
--
James Roper
Summer Research Scholar Department of Computer Science
Fourth Year Software Engineer Department of Computer Science
Australian National University
"A deceptively simple architectural challenge faced by many
multi-tasking applications is gracefully doing nothing."
Glyph Lefkowitz
|
