search

anderkve / FYS3150

https://anderkve.github.io/FYS3150
26 stars 14 forks source link

Disagreement between analytic and numerics for L=2 #95

Closed ivarlon closed 1 year ago

ivarlon commented 1 year ago

Hey, I'm getting kinda weird results for problem 4 when comparing analytical to numerical results. I get heat capacity and susceptibility to be lower numerically than the analytic values. problem4_chi.pdf

Am I right in computing mean(E) = \sum_{N+} d(N+) exp(-E(N+)) E(N+) and equivalently for M?

FWIW, the means computed numerically and analytically aren't much different (like less than 1% or thereabouts), but the heat capacity and susceptibility seem very sensitive to these small differences.

ivarlon commented 1 year ago

NVM, I figured it out, had some wrong analytical values.