How can I solve delay differential equation (using dde23) in a broad range?

Dear all, I am trying to solve a delay differential equation. It is very simple, but I don't get enough accurate answer from these set of equations. The dde equations are as follow:

dA/dt=K*exp(0.5(1-a1i)G(t-T)-0.5(1-a*1i)Q(t-T))A(t-T)-K'*A(t)

dG/dt=D-Gamma*G(t)-exp(-Q(t))*(exp(G(t))-1)*|A|^2

dQ/dt=C-Q(t)-S(1-exp(-Q(t)))*|A|^2

1i in the first row of equations is imaginary number. K, K', D, C, and a are constant coefficients. A, G, Q are time dependent.

I have one basic questions which is that how can I define the arguments of dde23 to get better response? Do I have to use ddeset or options in my dde23? How can I understand there is a jump or discontinuities in solving these equations? How can I recognize that I need other options in defining dde23? I think as I extend the simulation range I get worse answer. I think the truncation error increases as the range increases.

The initial values for A, G, Q are 0, 4.18, 3.20, respectively. The developed code is as follows:

clc

A_0=0.0000; % Initial Condition of optical field

G_0=4.18; % Initail Condition for saturated gain

Q_0=3.20; % Initial Condition for saturated absorption

T=5;

% opts = ddeset('RelTol',1e-6,'AbsTol',1e-10);

options = ddeset('RelTol',1e-6,'AbsTol',1e-8,... 'InitialY',[0.001+0.001i;G_0;Q_0]);

sol = dde23('DDE',T,[A_0;G_0;Q_0],[0, 600],options);

plot(sol.x,(abs(sol.y)).^2);

title('Figure 1. Pulse Propagation.')

xlabel('time t');

ylabel('y(t)'); -------------------------

function v = DDE(t,y,Z)

gamma=39.15;

kappa=0.55;

alpha_g=0.2;

alpha_q=0.2;

Gamma=0.08;

S=4.65;

q0=Gamma*4.18;

g0=3.20/S;

T=5;

ylag1 = Z(:,1);

v = zeros(3,1);

v(1) = gamma*sqrt(kappa)*(exp(0.5*(1-alpha_g*1i)*ylag1(2)-0.5*(1-alpha_q*1i)*ylag1(3)))*ylag1(1)-

gamma*y(1);

v(2) = g0-Gamma*y(2)-exp(-y(3))*(exp(y(2))-1)*(abs(y(1)))^2;

v(3) = q0-y(3)-S*(1-exp(-y(3)))*(abs(y(1)))^2;

end

Thank you in advance for all your help! I am looking forward to your answers!