How do I get gif file function to work properly?

Hello I am trying to get the gif from my following code by I keep running into an error. This is my first time dealing with it so I am not sure where my issue is.

clear all; close all
% Initial Conditions
theta_p= 0.01*pi/180; % initial theta position (rad)
%theta_v=0;  % inital theta velocity (rad/s)

beta_p=90*pi/180-theta_p;
beta_v=0;

d_p=0.15; %inital position of collar A (m)
d_v=0; %inital velocity of collar A (m/s)
g=9.81; %gravity constant (m/s^2)
L=0.3; % length of bar (m)
mB=2; % mass of bar (kg)
mA=1; %mass of collar (kg)
dt=.0001;
I = (mB*L^2)/12;

Wa=mA*g;
Wb=mB*g;
x=(mB*L)/2;
z=(L/2);

for i = 1:50000
    A = [-1 0 -mA 0;
        1 0 -mB (x*sin(beta_p))
        0 1 0 -x*cos(beta_p)
        (sin(beta_p)*z) (-cos(beta_p)*z) 0 -I];
    B = [0;
        -x*(beta_v)^2*cos(beta_p); 
        Wb-x*(beta_v)^2*sin(beta_p); 
        0];
    X=A\B;
    beta_v = beta_v + X(4)*dt;
    beta_p=beta_p + beta_v*dt;
    d_v=d_v+X(3)*dt;
    d_p=d_p+d_v*dt;
    N = X(2)+Wa;
    
    time(i)= i*dt;
    dp(i)=d_p;
    dv(i)=d_v;
    beta(i)=beta_p;
    omega(i)=beta_v;
    Rx(i)=X(1);
    Ry(i)=X(2);
    d_a(i)=X(3);
    beta_a(i)= X(4);
    
end
%% Animation Setup
close all; clc;
scale = 10;
% Collar A 
xBl = scale*[-0.02 0.02 0.02 -0.02];
yBl = scale*[0.02 0.02 -0.02 -0.02];
% Pin
gam = linspace(0,2*pi,100);
xPin = scale*0.005.*cos(gam);
yPin = scale*0.005.*sin(gam);
% Swinging Bar:
xSBar = scale*[-0.02 0.25 0.25 -0.02];
ySBar = scale*[0.01 0.01 -0.01 -0.01];
% Frictionless Rod
xFBar = scale*[-0.2 0.2 0.2 -0.2];
yFBar = scale*[0.01 0.01 -0.01 -0.01];

%% Animation
% Initialize figure
h = figure; 
filename = 'Test2.gif';
% Initialize some vectors
xBlP = xBl;
yBlP = yBl;
xPinP = xPin; 
yPinP = yPin;
xSBarP = xSBar; 
ySBarP = ySBar;

for i=1:length(time)
    if mod(i,200) == 0 
        clf
        % Update coordinates of shapes
            % Rotation of bar
            xSBarP = xSBar*cos(beta(i)) - ySBar*sin(beta(i));
            ySBarP = xSBar*sin(beta(i)) + ySBar*cos(beta(i));
            % Translate of bar
            xSBarP = dp(i)+xSBarP;
            % Translation of block
            xBlP = dp(i)+xBl;
            yBlP = yBl;
            % Translation of pin
            xPinP = dp(i)+xPin;
            yPinP = yPin;
        % Plot shapes 
        patch(xFBar, yFBar,'b'); 
        patch(xBlP, yBlP, 'y');
        patch(xSBarP, ySBarP, 'r');
        patch(xPinP, yPinP, 'k'); 
        axis('square'); axis([-2.5 2.5 -2.5 2.5]); axis('off');
        % Process frame
        frame = getframe(h);
        im = frame2im(frame);
        [imind,cm] = rgb2ind(im,256);
        if i == 1
            imwrite(imind,cm,filename,'gif','Loopcount',inf);
        else
            imwrite(imind,cm,filename,'gif','WriteMode','append');
        end
    end
end

[Walter Roberson]

Error using wgifc Can only append to GIF89a format GIFs.

Error in writegif (line 306) wgifc(mat, map, filename,writemode,disposalmethod,delaytime,...

Error in imwrite (line 566) feval(fmt_s.write, data, map, filename, paramPairs{:});

[Merve Buyukbas]

When I run the program the error line is frame = getframe (h); found as.

There may be several reasons for this, it may be necessary to plot the x, y coordinates and under the heading plot sahapes with the plot command. Then, continue by adding the drawnow command. If we want to run a sample file without any problem, you can try the following codes. Here frame = getframe (h); command works without errors.

[Merve Buyukbas]

I think this link is useful: https://www.mathworks.com/matlabcentral/answers/526432-error-using-imwrite

[Walter Roberson]

I had no problem with the getframe()

The code already has

h = figure; 

and there should be no need to generate another figure.

Answer 1

[Walter Roberson]

imwrite cannot handle different colormaps for each frame. You have to write in rgb mode. It will automatically reprocess the color data of the appended frames to match the colormap it deduced from the first (rgb) frame.

Yes, this is more limited than one might prefer, and does tend to imply that you should write a colorful frame first so colors on the remaining frames are not accidentally mapped through a restricted colormap.

[Umut Ayyildiz]

I've doen another animation using the same format in my code above and it still worked. I'm still sort of confused as to what I need to do.

This is the code I am refering to where it worked:

clear all; close all
m = 0.5; %mass kg
dt = .04; % timestep
r(1) = 0.8; %r position m
r_vel(1)=0; %r velocity m/s 
theta(1) = .7854; %theta position rad  
theta_vel(1)=0; %theta velocity rad/s
theta_acc = 10; %theta constant acceleration rad/s^2 
g=9.81; %gravity constant m/s^2

for ii = 1:200
A = [sin(theta(ii))/m -(sin(theta(ii))+2*dt*theta_vel(ii)*cos(theta(ii))+2*dt^2*theta_acc*cos(theta(ii)))
     cos(theta(ii))/m -(-cos(theta(ii))+2*dt*theta_vel(ii)*sin(theta(ii))+2*dt^2*theta_acc*sin(theta(ii)))];
 
B=[2*r_vel(ii)*theta_vel(ii)*cos(theta(ii))+2*r_vel(ii)*dt*theta_acc*cos(theta(ii))+r(ii)*theta_acc*cos(theta(ii))-r(ii)*(theta_vel(ii)+theta_acc*dt^2)*sin(theta(ii))
   2*r_vel(ii)*theta_vel(ii)*sin(theta(ii))+2*r_vel(ii)*dt*theta_acc*sin(theta(ii))+r(ii)*theta_acc*sin(theta(ii))-r(ii)*(theta_vel(ii)+theta_acc*dt^2)*cos(theta(ii))+g];

X=inv(A)*B;
r_acc(ii+1)=X(2); %r acceleration 
F(ii+1)=X(1); %force
T(ii+1)=dt*ii; %time
theta_vel(ii+1)=theta_vel(ii)+theta_acc*dt; %theta velocity
theta(ii+1)=theta(ii)+theta_vel(ii+1)*dt; %theta position
r_vel(ii+1)=r_vel(ii)+r_acc(ii+1)*dt; %r velocity
r(ii+1)=r(ii)+r_vel(ii+1)*dt; %r position 
end

%%
%%% Animation
%%% Block
Xb = [-25 -22 -22 -25]; 
Yb = [-1 -1 1 1];
%%%Holder Outside
Xout = [-39 3 3 -39]; 
Yout = [-1.75 -1.75 1.75 1.75];
%%%Holder Inside
Xin = [-38 -10 -10 -38];
Yin = [-1 -1 1 1];
%%% Base
alpha = linspace(0,2*pi,20);
X0 = 1*cos(alpha);
Y0 = 1*sin(alpha);
%%
close all
%%%% Animation Set up
h = figure;
filename = 'Project1Animation.gif';
for ii = 1:length(T)
    clf
    for jj = 1:4
        Xbp(jj)=(Xb(jj)+r(ii))*cos(theta(ii))-(Yb(jj))*sin(theta(ii));
        Ybp(jj)=(Xb(jj)+r(ii))*sin(theta(ii))+(Yb(jj))*cos(theta(ii));
        Xoutp(jj)=(Xout(jj))*cos(theta(ii))-(Yout(jj))*sin(theta(ii));
        Youtp(jj)=(Xout(jj))*sin(theta(ii))+(Yout(jj))*cos(theta(ii));
        Xinp(jj)=(Xin(jj))*cos(theta(ii))-(Yin(jj))*sin(theta(ii));
        Yinp(jj)=(Xin(jj))*sin(theta(ii))+(Yin(jj))*cos(theta(ii));
    end
    patch(Xoutp,Youtp,'b'); hold on
    patch(X0,Y0,'g');
    patch(Xinp,Yinp,'w');
    patch(Xbp,Ybp,'r'); 
    axis([-45 45 -45 45]); 
    axis('square'); 
    axis('off')
 %%%% Animation to GIF
 frame = getframe(h);
 im = frame2im(frame);
 [imind,cm] = rgb2ind(im,256); 
 % Write to the GIF File
 if ii == 1
 imwrite(imind,cm,filename,'gif', 'Loopcount',inf);
 else
 imwrite(imind,cm,filename,'gif','WriteMode','append');
 end
end

[DGM]

Unless something changed, imwrite can write multiframe gifs with different colormaps per frame. The problem is with imread. Imread cannot properly read multiframe gifs with unique local color tables.

Historically, there were workarounds, but after R2018b, the process of using imread on a multiframe gif will destructively remap all frames to use the calculated color table for frame 1.