About the way torsion is saved in the qforce_fragments

xiki-tempula commented 2 years ago

In #48, I have a case where the qforce is not doing very good but it is not reproducible. So I'm trying to chase down this ghost issue. One suspicion is that the scandata is not save correctly. I noticed that for relaxed surface scan, I got the scandata in the form of

     5.930          24.93443614
    20.930          16.18528833
    35.930           7.40614033
    50.930           1.88102842
    65.930           0.04347315
    80.930           0.35472505
    95.930           0.52885336
   110.930           0.05999423
   125.930           0.59080997
   140.930           4.27734184
   155.930          12.19904202
   170.930          22.69822572
   185.930          24.57427096
   200.930          14.32332406
   215.930           5.57418543
   230.930           1.04531139
   245.930           0.01165906
   260.930           0.45166566
   275.930           0.45621612
   290.930           0.00000000
   305.930           1.20855065
   320.930           5.92901368
   335.930          14.15505403
   350.930          23.42069362

But for the one done with torsiondrive, it is in the form of

     0.000          26.22991499
    15.000          19.81758371
    30.000          10.49649813
    45.000           3.48761395
    60.000           0.28549421
    75.000           0.03214399
    90.000           0.44978747
   105.000           0.13904384
   120.000           0.00000000
   135.000           2.24429022
   150.000           8.46043367
   165.000          18.49720944
   180.000          26.24210256
   195.000          18.47860777
   210.000           8.47838421
   225.000           2.22708795
   240.000           0.00273840
   255.000           0.13837958
   270.000           0.45825996
   285.000           0.03434416
   300.000           0.26326410
   315.000           3.48145978
   330.000          10.45666930
   345.000          19.79963317

This worries me a bit as the torsiondrive always give the same angle list from -165 to 180 with a step size of 15.

I have checked that the https://github.com/selimsami/qforce/blob/master/qforce/qm/qm.py#L71, where the angle_list is the one that I would expect to have.

xiki-tempula commented 2 years ago

Ok, I see that it has been sorted to make it in the range of 0-360

selimsami commented 2 years ago

It is indeed sorted. But did you manage to figure out what was happening with the case of "qforce is not doing very good but it is not reproducible." ?

xiki-tempula commented 2 years ago

@selimsami No. I cannot reproduce it now so I will just leave it aside.

selimsami / qforce

About the way torsion is saved in the qforce_fragments #49