A polymer sample-preparation method (extended-chain condensation, ECC) based solely on molecular-dynamics simulations has been compared to a connectivity-altering Monte Carlo method (coarse-grained end-bridging, CGEB). Since the characteristic ratio for the CGEB samples is closer to the experimental value, ECC results in polymer structures that are too compact. The stress-strain relations are different in the strain-hardening regime. For CGEB samples, a stronger strain hardening is observed and the strain-hardening modulus is more realistic; for the CGEB polystyrene (PS) sample GR = 9 ± 1 MPa is found versus GR = 4 ± 2 MPa for the ECC samples. These differences have to be attributed to a steeper increase in the contributions to the total stress from bond- and dihedral angles for CGEB than for ECC samples.