Ask Your Question

Revision history [back]

click to hide/show revision 1
initial version

Improve time to plot a set of points

I try to draw clouds of coloured dots. The coordinates of the points are given by a formula. Let's say they're randomly distributed in [0,1]: that doesn't really matter here and isn't the problem.

For the plot, I use the following code. It's remarkably slow (in the sense that I didn't expect it to take so long).

The options don't seem to be the cause of the slowdown (*), but they do limit my use of list_plot, which seems to be much more efficient.

For example, with the following code:

  • the total CPU time for Trace() is 1min
  • the total CPU time for Trace2() is 121ms

I'm looking for ideas on how to improve the time it takes to draw my colored dots (I'd probably have to find a suitable python library, but I'm far from an expert...).

N = 10^4
R = range(N)
P=[(randrange(N)/N*1.0,randrange(N)/N*1.0) for k in R]
#..................
def mycolorscale(k):
    return Color(1 - k/N, 0, 0)
#..................
def Trace():
    g=Graphics()
    k=0
    for p in P:
        g+=point(p,rgbcolor=mycolorscale(k),size=1)
        k+=1
    g.show()
#..................
def Trace2():
    list_plot(P,plotjoined=False,size=1).show()

(*) Replacing

g+=point(p,rgbcolor=mycolorscale(k),size=1)

by

g+=point(p)

in the Trace code changes essentially nothing to the duration.