Developing i-pi interface to local MD program

[Karthik Ganeshan]

Hi all, 

I have a local MD code we use to test some of our hypothesis on written in fortran77 (yes, its ancient), and I'll like to test PIMD's effect on it. 

Unfortunately, the modifications haven't been ported to lammps yet. 

How would one go about developing the interface to i-pi? Are there any walkthrough examples or a generic guide available for this? 

Thank you! 

Sincerely,
Karthik. 

[Michele Ceriotti]

The driver.f90 program is a good example of how to develop a driver. 

It should be easy to hack the MD loop of your legacy software to serve as the communication loop.

[Venkat Kapil]

You can check out i-pi/drivers/driver.f90 as a template. It is a simple fortran code that implements a number of model potential energy surfaces and interfaces with i-PI. Feel free to adapt it to also run your potential.

[Karthik Ganeshan]

Okay, the example does make sense. I think I should be able to add it in. Thanks again!