Indexing Bug & Additional Stars

[webbjj]

Hi Everyone

I have stumbled across a potential indexing bug that I wanted to bring to your attention as well as the existence of extra stars in the dataset. I believe it is related to dynamically formed binaries. I am running Nbody6++ with the following input file:

1 1000000.0 1.E6 40 40 640
12500 1 200 1434 50 1 10
0.015 0.02 0.05 20.0 20.0 3100.0 1.0E-2 8. 0.6
2 0 4 0 1 0 1 0 0 0
0 3 0 3 2 1 0 1 3 7
1 1 0 0 0 2 2 1 0 1
0 0 2 2 1 0 0 1 0 3
0 0 0 0 0 3 8 0 0 0
1.0E-07 0.001 0.1 1.0 1.0E-06 0.001 1.0
2.3 50.0 0.1 0 0 0.001 0.0 5.0
0.5 0.0 0.0 100000.0
1.5e10 5.5e10 10.0 4.0 220.0 8.5 0. 0. 0. 10.0 0. 0. 0. 223.68796485010824 0.
0.0 1.0 1.0 1.0

While there are no primordial binaries, by the time it gets to nbody timestep 1680, there are 4 binaries. Here is bev.82_1680

   4   6512.5

1.68000E+03 1 2 5 14 14 14 0 9.96024E-01 9.85456E-01 5.01143E+00 4.51095E+00 2.58489E+01 1.75669E+01 -1.00000E+01 -1.00000E+01 -3.96019E+00 -4.12794E+00 3.24187E+00 3.32575E+00
1.68000E+03 3 4 4285 8660 0 0 0 8.00694E-01 8.02674E-01 8.79045E+00 6.40767E+00 4.09644E-01 1.78897E-01 -1.49514E+00 -2.17685E+00 -4.32391E-01 -6.99782E-01 3.60419E+00 3.56746E+00
1.68000E+03 5 6 4689 4899 0 0 0 1.84442E+00 9.89642E-01 8.35474E+00 6.15014E+00 3.77889E-01 3.61061E-01 -1.58298E+00 -1.62675E+00 -4.60340E-01 -4.75449E-01 3.59620E+00 3.59281E+00
1.68000E+03 7 8 1 3 14 14 0 2.57839E-01 4.45500E-01 7.48501E+00 6.10300E+00 1.48394E+01 1.44455E+01 -1.00000E+01 -1.00000E+01 -4.20122E+00 -4.21290E+00 3.36239E+00 3.36823E+00

It is important to note that at this step sev.83_1680 has NS=1242 and there are 1242 stars in that file. This makes sense, since NS+2*NB=1250, the initial number of stars in the dataset. Surprisingly conf.3_1680 has 4 extra stars in it that are not in sev or bev, with IDS [12505 16785 17189 12501]. These stars have non-zero masses, so not sure what they are doing here.

Even more confusing is when we go to the next output step bev.82_1700 now reads:

   4   6590.0

1.70000E+03 1 2 5 14 14 14 0 9.70733E-01 9.85456E-01 5.01143E+00 4.51095E+00 2.58489E+01 1.75669E+01 -1.00000E+01 -1.00000E+01 -3.96019E+00 -4.12794E+00 3.24187E+00 3.32575E+00
1.70000E+03 3 4 4285 8660 0 0 0 7.76344E-01 8.02534E-01 8.79045E+00 6.40767E+00 4.09644E-01 1.78897E-01 -1.49510E+00 -2.17684E+00 -4.32378E-01 -6.99780E-01 3.60419E+00 3.56746E+00
1.70000E+03 5 6 4689 4899 0 0 0 1.80253E+00 9.89638E-01 8.35474E+00 6.15014E+00 3.77889E-01 3.61061E-01 -1.58295E+00 -1.62672E+00 -4.60329E-01 -4.75440E-01 3.59620E+00 3.59282E+00
1.70000E+03 7 6786 1 3 14 14 0 2.60372E-01 4.46390E-01 7.48495E+00 6.10296E+00 1.48394E+01 1.44455E+01 -1.00000E+01 -1.00000E+01 -4.20122E+00 -4.21290E+00 3.36239E+00 3.36823E+00

AS you can the fourth binary underwent an exchange. However the troublesome part is that I think for that binary J2 and NAME(J2) are wrong. When I read in the conf.3_1700 file the first 8 star names are 5,14,4285,8660,4689, 4899, 1, 6786. I would have expected the last NAME to be 3 given the bev file. Also at this time step sev.83_1700 still has NS=1242 but only has 1241 stars in the file. There is no single stars with ID 3 or 6786.

Finally, at this timestep there are still 4 stars in the conf file that are not in sev or bev - [12505,16785,17189,-1]. I am not sure where the -1 comes from and am still confused as to why these extra 4 stars exist in the conf file.

Any insight or help that you can provide would be most appreciated. Its not clear to me if the indices output in the bev and sev files can be trusted or if I should just assume the order of the stars in the conf file and then the bev+sev files is the same. Either way, the 4 extra stars in the conf file are an issue is they have no SEV/BEV information.

Thanks

Jeremy