[gmx-users] problem when calculating the solvation free energy of ARG in water using PME.

[chris.neale at utoronto.ca]

1. How sure are you that you have reproduced the run exactly on the  
different machines?
2. Are there any other differences besides machine? Number of cores?
3. What's your water model?
4. What else is in your system? (counterions?)
5. What are your crash/not-crash statistics? The OS difference might  
just be noise.

I realize that this could be some machine-dependent error (as you seem  
to be suggesting) but I would not be able to help you with that. Let's  
ensure that it is not some other difference and then you could file a  
redmine issue if the problem can not be resolved. You likely have a  
good small test case.

Chris.

-- original message --


I am trying to calculate the solvation free energy of ARG in water, using
GROMACS 4.0.7 (using TI).


I found that when I run the calculations on linux machines (I try few) it
run OK, but when I run the calculations on OSX it crashed in most lambda
points. I run the calculations using PME and RF, and the problem is
consisted only with PME. I also done the same thing for ALA and it works
fine. So I guess it has a connection to the charges.


  Any advices?


Best regards,