Compute RDF strange behaviour

Dear all,

I have noticed strange behaviour of compute rdf while simulating of argon-carbon system. Here I reproduce this issue simply simulating argon bulk. So then I use hybrid (lj + airebo) potential and compute rdf I get explicable large number in the first bin. But then I use just lj - everything is ok.

lammps-31Mar17, Installed YES: packages KSPACE, MANYBODY, MOLECULE, VORONOI, USER-INTEL.

Input scripts:

hybrid(lj + airebo):

Dear all,

I have noticed strange behaviour of compute rdf while simulating of
argon-carbon system. Here I reproduce this issue simply simulating argon
bulk. So then I use hybrid (lj + airebo) potential and compute rdf I get
explicable large number in the first bin. But then I use just lj -
everything is ok.

​thanks for reporting this and sorry for the delay. i only now found the
time to try and properly reproduce it and make additional tests.
it looks like this is an "undesired feature" resulting from the recent
neighborlist refactoring. it applies to several conditions beyond what you
are reporting and i've recorded on github at:

until we have tracked this one down properly and implemented a bugfix, i
would suggest you try out version 17 Nov 2016 and see if the issue was
present *before* the refactoring.

thanks for reporting this,
      axel.