search

jrhaberstroh / cGromCorrFMO

Correlation computation package for gromacs
2 stars 0 forks source link

f1_2SpectralCorr Correlation functions disagree #16

Closed jrhaberstroh closed 10 years ago

jrhaberstroh commented 10 years ago

Several implementations of the correlation function disagree. There are many t's to cross and i's to dot, and it is important to get them all right.

jrhaberstroh commented 10 years ago

For diagonals, there was a problem with the FFT method due to complex conjugation.

For off-diagonals, there is an important time-asymmetry to the correlation functions, and the convention was resolved to be Corr(f1,f2) = <f1(0) f2(t)>.