Hello I have a problem with my coding, with drawnow in my opinion? But it could be something else

Hello, so basically I want to code beads which move left and write independently to each other unless if they collide. I haven't written the collision algorithm so far, but right now i've gotten a problem with my plotting: When running my code, instead of having 2 balls simply move along the screen (left and right), i get the balls leaving a trail behind them while moving (as seen in attached pic.)

Part of my code is this (which i believe causes the problem since the other is just sorting out):

for t = 0:500

      for n=1:(nMolecules-1) %MMCollision check
           m=n+1;
      end

for n = 1:nMolecules
    % Wall collision check (Only 1st and last molecules)
    if (Molecule(1,n)<= 0.5*Molecule(3,n))
     Molecule(4,n) = -Molecule(4,n); %Elastic Collision
    elseif (Molecule(1,n)>= Right-0.5*Molecule(3,n)) 
     Molecule(4,n) = -Molecule(4,n); %Elastic Collision   
    end
    Molecule(1,n) = Molecule(1,n) + Molecule(4,n);
end

hold on

for n=1:nMolecules

xlim([Left Right]);
ylim([Down Up]);

if (Molecule(2,n)==4) %Drawing different balls based on rng
h = plot(Molecule(1,n),Molecule(5,n),'o','Color','black','MarkerFaceColor','red','MarkerSize',100*Molecule(3,n));      
else
h = plot(Molecule(1,n),Molecule(5,n),'o','Color','black','MarkerFaceColor','blue','MarkerSize',100*Molecule(3,n));      
end

end
hold off
drawnow

end

Basically my code randomly attributes a size to a ball and its "properties" like mass and radius data (not in provided code). The provided image shows both being of the same size and yes, there's supposed to be only 2 balls on the screen, instead of 10+

Anybody can help me?

[Rena Berman]

(Answers Dev) Restored Question.

Answer 1

[Star Strider]

We’re missing some code. Even if I define ‘nMolecules’, ‘Molecule’ remains undefined.

I can’t run your code, but from my experience, if you don’t want the ‘trail’ effect, eliminate the hold calls.

I wrote this for a — not yet Accepted but hope but never fades — Answer a week ago.

It’s not your intended application, but perhaps you can glean some ideas from it, since I did not use hold, and put the axis limits just after the plot (in my code, surf) call:

[X,Y] = meshgrid(linspace(-5, 5, 50));
fcn = @(x,y,k) k*x.^2 + y.^2;

v = [1:-0.05:-1;  -1:0.05:1];
for k1 = 1:2
    for k2 = v(k1,:)
        surfc(X, Y, fcn(X,Y,k2))
        axis([-5  5    -5  5    -30  50])
        drawnow
        pause(0.1)
    end
end

Run it! It’s fun to watch! I goes from an attractor basin to a saddle-point and back. I put the pause call in to make it appear relatively smooth. Tweak that as necessary.

[Star Strider]

All relevant complete code and a clear description of what you want to do is always appropriate. Otherwise we are left to guess, and our guesses are rarely consistent with what you want. For all our other areas of expertise, our individual and collective mind reading skills fail almost completely.

The hold function is likely unnecessary if you write your code correctly.

[Star Strider]

Well, I eliminated the ‘tails’, but I eliminated all but one of the molecules as well, which was not my objective. It’s difficult for me to follow what you’re doing, and it’s the end of my day anyway, so I’ll return this to you and let you add the other molecules without their tails.

My (slightly) revised code:

nMolecules =5; % Number of molecules
BSize = nMolecules; % Box Size
Molecule=zeros(5,nMolecules);
% Data 1: Position x
% Data 2: Mass
% Data 3: Size
% Data 4: Velocity
% Data 5: Position y - Fixed
Molecule(5,:) = 0.5; %Midway of box
% %%
% Ball Type A
A = zeros(4,1);
A(2,1)= 1; % Mass = 1 unit
A(3,1)= 0.1; % Diameter = 0.1 unit
% Ball Type B
B = zeros(4,1);
B(2,1)= 4; % Mass = 4 units
B(3,1)= 0.2; % Diameter = 0.2 unit
% Wall
Left = 0;
Right = BSize+1;
Down = 0;
Up = 1;
Wall = [Left Right Down Up];

figure(1)
xlim([Left Right]);
ylim([Down Up]);
% %%Molecule Set up
T = 0; %Type randomiser
for n = 1: nMolecules
    Molecule(1,n) = n; %Position
    T = rand(1);
    Molecule(4,n) = cos(2*pi*rand(1))/50; %Random Velocity

        if (T <= 0.5) %Choose 0.5 chance of being Type A
            Molecule(2,n) = A(2,1);
            Molecule(3,n) = A(3,1);
        else %else Type B
            Molecule(2,n) = B(2,1);
            Molecule(3,n) = B(3,1);
        end
    end

% %%
figure (1)
n=1;
for t = 0:500

        for n=1:(nMolecules-1) %MMCollision check
             m=n+1;
        end

for n = 1:nMolecules
    % Wall collision check (Only 1st and last molecules)
    if (Molecule(1,n)<= 0.5*Molecule(3,n))
     Molecule(4,n) = -Molecule(4,n); %Elastic Collision
    elseif (Molecule(1,n)>= Right-0.5*Molecule(3,n)) 
     Molecule(4,n) = -Molecule(4,n); %Elastic Collision   
    end
    Molecule(1,n) = Molecule(1,n) + Molecule(4,n);
end
% xlim([Left Right]);
% ylim([Down Up]);

for n=1:nMolecules

if (Molecule(2,n)==4)
h = plot(Molecule(1,n),Molecule(5,n),'o','Color','black','MarkerFaceColor','red','MarkerSize',100*Molecule(3,n));      
axis([-1  5    0  1])
hold on
else
h = plot(Molecule(1,n),Molecule(5,n),'o','Color','black','MarkerFaceColor','blue','MarkerSize',100*Molecule(3,n));      
axis([-1  5    0  1])
end
hold off

end

drawnow
% delete(h)

end

[Klyde]

nvm. I may change my code completely. Thanks though

[Star Strider]

My pleasure.