what's difference between mode_compute6 and 4 in dynare

[blackeif]

what's the difference between mode_compute6 and 4 in dynare when using bayesian approach for estimation? which one is better? many result cannot be repeated even using the same mode_compute type,it seems bayesian method is not reliable in dynare.and another thing,when using mode_compute4,it usually shows error message as follows,but mode_compute 6 does not:
-------------------------------mode_compute6 error message-----------------
Log data density [Laplace approximation] is -6855.383601.

??? Error using ==> chol
Matrix must be positive definite.

Error in ==> metropolis_hastings_initialization at 52
d = chol(vv);

Error in ==> random_walk_metropolis_hastings at 43
[ ix2, ilogpo2, ModelName, MhDirectoryName, fblck, fline, npar, nblck, nruns, NewFile, MAX_nruns, d ] = ...

Error in ==> dynare_estimation_1 at 1061
feval(options_.posterior_sampling_method,'DsgeLikelihood',options_.proposal_distribution,xparam1,invhess,bounds,gend,data,...

Error in ==> dynare_estimation at 62
dynare_estimation_1(var_list,varargin{:});

Error in ==> Liu_replica at 270
dynare_estimation(var_list_);

Error in ==> dynare at 132
evalin('base',fname) ;