trace interface is not yet implemented
Problem to solve with the new feature
The trace interface existed in older versions of carp/carpentry but it has not yet been implemented in openCARP. It is a very useful feature for many reasons. A great use case would be if I'm running a very large simulation with a relatively coarse time step (e.g., spacedt = 20 ms -- usually good enough to visualize overall patterns) but I want to show high-resolution traces of Vm(t) for some subset of nodes. That's when I would use -trace!
It is also enormously helpful for interrogating ionic model variable values over time. Currently the gvec interface can be used to dump state variables, but I don't think there's any way to dump computed variables (e.g., ionic currents over time).
Those running large organ-scale simulations and wishing to closely characterize state variable behavior over time in a subset of nodes.
It is a great feature!
It is implemented in older versions of carp/carpentry but not ported over yet.
I should add one very important note, which is that in older versions I consistently experienced an issue where trace would not work reliably unless the job was running on exactly 1 core. If the node to be dumped lived on core zero, it would get dumped; otherwise, the trace file would be empty.
If the dev team implements this, I will absolutely help test and I will volunteer to write a few simple cases for creating a tutorial/example.
Links / references