Running Dynare from Matlab

Question

I am new to Dynare++ and I have really quick question I cannot seem to find the answer too.

What is the difference between these two commands and why is the output different?

!dynare++ --per 20 --sim 3 file_name.mod

dynare file_name.mod

In the first command its unable to find steady state values based on my initial values and in the second it can. Why?

Here is the mod file of the basic RBC model I want to simulate with the first command. I can do so with the second, but not the first.

var c, l, k, w, r, y, i, z;
varexo eps;

parameters beta, delta, alpha, a, rho, rbar, wbar;

beta=.99;
delta=.025;
alpha=.33;
a=1;
rho=.9;
rbar=1/beta+delta-1; 
wbar=(1-alpha)*(alpha/rbar)^(alpha/(1-alpha)); 


model;
//Inter-Euler
1/c=beta*(1/c(+1))*(1+r(+1)-delta);
//Intra-Euler
a/(1-l)=w/c;
//H.H. Budget Constraint
k=w(-1)*l(-1)+(1-r(-1)-delta)*k(-1)-c(-1);
//Firm FOC Capital
r=exp(z)*alpha*exp(z)*(l/k)^(1-alpha);
//Firm FOC Labor
w=exp(z)*(1-alpha)*exp(z)*(k/l)^(alpha);
// Production Function
y=exp(z)*(k^alpha)*(l^(1-alpha));
i=y-c;
z=rho*z(-1)+eps;
end;

initval;
k=((wbar/a)/(wbar*(1+1/a)-(r+delta)*(alpha/rbar)^(1/(1-alpha))))*(alpha/rbar)^(1/(1-alpha));  
c=wbar*(1-((wbar/a)/(wbar*(1+1/a)-(r+delta)*(alpha/rbar)^(1/(1-alpha)))))/a;  
l=(wbar/a)/(wbar*(1+1/a)-(r+delta)*(alpha/rbar)^(1/(1-alpha))); 
w=wbar; 
r=rbar; 
y=(k^alpha)*(l^(1-alpha));
i=(k^alpha)*(l^(1-alpha))-(wbar*(1-((wbar/a)/(wbar*(1+1/a)-(r+delta)*(alpha/rbar)^(1/(1-alpha)))))/a);
z=0;
end;

shocks;
var eps;
stderr .02;
end;