3 views (last 30 days)

Show older comments

Mike Garrity
on 20 Apr 2016

This code is creating 6,001 line objects which each have a single vertex. A line object with a single vertex is a degenerate case which isn't guaranteed to draw anything. Different renderers in different versions of MATLAB have handled this case differently, so this has never been a very reliable way to plot.

One option is to tell plot that you're drawing dots instead of lines. That's as easy as changing your call to plot to look like this:

plot(t,h,'.')

But creating 6,001 objects which each draw a single dot isn't a very efficient way to go, for this reasons I explained in this post on the MATLAB Graphics blog . You'd really be better off with a single call to plot. In your case, it would look something like this.

dt=1/1000;

t=0:dt:6;

g=9.81;

prev_h=2;

u=0;

t = 0:dt:6;

h = zeros(size(t));

for i=1:numel(t)

v = u + (g*dt);

dh = 0.5*(u+v)*dt;

h1 = prev_h-dh;

h(i) = h1;

prev_h = h(i);

u = v;

end

plot(t,h,'.')

The other advantage of this approach is that you could draw lines connecting the points instead of individual dots. With your current approach you can't do that because the plot command can't "see" more than a single data value at a time. By passing all of the data values into plot together, it can connect them.

Mike Garrity
on 25 Apr 2016

In that case, you'd want to create an empty plot first, and then append to its XData & YData in the loop. Something like this:

dt=1/1000;

g=9.81;

h=2;

u=0;

hobj = plot(nan,nan);

xlim([0 6])

ylim([-180 20])

for t=0:dt:6

v = u + (g*dt);

dh = 0.5*(u+v)*dt;

h1 = h-dh;

h = h1;

u = v;

hobj.XData(end+1) = t;

hobj.YData(end+1) = h;

drawnow limitrate

end

But there's a graphics object called animatedline (introduced in R2014b) that is designed for exactly this job:

dt=1/1000;

g=9.81;

h=2;

u=0;

hobj = animatedline;

xlim([0 6])

ylim([-180 20])

for t=0:dt:6

v = u + (g*dt);

dh = 0.5*(u+v)*dt;

h1 = h-dh;

h = h1;

u = v;

addpoints(hobj,t,h)

drawnow limitrate

end

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!