Delay in Plotting Next Overlaid Plot When Animating 2D Plot

I have a 10 line plots that are overlaid, and I want to animate plotting each of these plots--one after the other. So I do need hold on, or some form of method to keep the previous plot as the new plot is being plotted.

The only problem I am facing is that after one plot has completed plot, before the next iteration, it takes a significant amount of time to start the next plot.

This time adds up when I have multiple overlaid plots (it compounds too).

Below I have shared my code, and a snippet of the data I am using to plot. Please point out how I can fix this

clc;clear;close all;load('data.mat');
fv=data{1};rv=data{2};
cmap = jet(5);  figure;hold on
        %         v = VideoWriter('example', 'MPEG-4');
        %         open(v);
for i = 1:3
    f=fv(:,i);r=rv(:,i);
    color = cmap(i,:);
    tic
    for ii=1:length(f)
        plot(f(1:ii),r(1:ii),'LineWidth',2,'Color',color);pause(0.005)

    end
    toc;disp('Next Plot')
end
% close(v);

Answer 1

[DGM]

You're creating a new high-level graphics object for each plot() call. That gets very expensive, and everything will just keep getting slower. The reason for the big salient delay is that there's a large block of NaN at the end of the data. Even though nothing visible shows up, you're still creating 800+ more graphics objects during that time.

You can just trim the NaNs out and update the one object for each series.

load('data.mat');
fv = data{1};
rv = data{2};
cmap = jet(5);  

nseries = size(fv,2);
hp = gobjects(nseries,1);
for i = 1:nseries
    f = fv(:,i);
    r = rv(:,i);
    color = cmap(i,:);
    
    % a large chunk (~74%) of the data is invalid/blank
    % consider omitting that chunk to save time
    mk = ~isnan(f);
    f = f(mk);
    r = r(mk);
    
    % instead of creating 1200 overlapping line objects of increasing length, 
    % just create one and update it
    for ii = 1:length(f)
        if ii == 1
            hp(i) = plot(f(1:ii),r(1:ii),'LineWidth',2,'Color',color);
            hold on
        else
            hp(i).XData = f(1:ii);
            hp(i).YData = r(1:ii);
        end
        pause(0.005)
    end
    
    disp('Next Plot')
end

There may be a drawback to just omitting all NaNs. If there are scattered blocks of NaN in the middle of the data which should properly show up as gaps, omitting them will cause a line to be connected across the gap. At a glance, it looks like all the NaNs are concentrated in a block at the end, but you might want to make sure that everything is still okay. More care may need to be taken to actually isolate the trailing block alone.

[Cesium Modern]

Thanks, this works as I wanted it to! Appreciate the detailed answer.