Hi Xiaoquan,

This is because the "stiffness" needs to be scaled according to dimension and model. In one dimension you need to multiply with the linear system size, giving you 0.638. Next you will take into account the mass of the particle, etc. to get to the value you expect.

Matthias

On 8 Nov 2010, at 18:55, Xiao Quan Yu wrote:

Dear all,

   We were trying to test ALPS project using 1D Hard-core

Bose-Hubbard model, my commands are following:

LATTICE="chain lattice"; L=100; MODEL_LIBRARY="/u/cm/carrasqu/ALPS/lib/xml/models.xml"; MODEL="hardcore boson"; t=1; mu=0 T = 0.01; SWEEPS=100000; THERMALIZATION=5000; EPSILON=0.25 {T=0.05} {T=0.04} {T=0.03} {T=0.02} {T=0.01} {T=0.005} {T=0.003} {T=0.002} {T=0.001}

 At the end, we got the correct ground state energy and charge

density, but the stiffness which is 0.00638 at T=0.001 is much smaller than the theoretical estimated value 0.315 at T=0. Is it because the number of sites is too small for this problem ? Or something else? Thank you in advance.

Best wishes, Xiaoquan

---

SISSA Webmail https://webmail.sissa.it/ Powered by SquirrelMail http://www.squirrelmail.org/