Is there any difference if I write a constraint A=0 as the nonlinear inequality constraint or nonlinear equality constraint when use the genetic algorithm?

The relative code is:

if true

[x,fval]=ga(@objfun,Nvars,A,b,Aeq,beq,LB,UB,@nonlconstr,options);
[c,ceq]= nonlconstr(x);
%the function nonlconstr is written like this:
A=0;
for i=1:n
Vjzb(i)=function(x);
  if Vjzb(i)>1.05
      A=A+(Vjzb(i)-1.05)^2;
  end
  if Vjzb(i)<0.95
      A=A+(Vjzb(i)-0.95)^2;
  end
end

end

In my project, the Vjzb(i) need to satisfy the constraint 0.95≤Vjzb(i)≤1.05, so A should equal to 0.

My question is: In this program, is there any difference if I write this constraint as c=A or ceq=A when I use the genetic algorithm? In fact, I have tested them in MATLAB, and it seems that writing this constraint as c=A is better, but why?