www.neuron.yale.edu

Hello,

We're working on porting NEURON models to some non-standard hardware, and we were wondering if you could point us to some documentation on how exactly NEURON implements its implicit Euler solver. We're interested in knowing details like whether you use Newton-Raphson,how you determine convergence, etc.

Thanks,
Michael

The method is the same as explained in
http://www.neuron.yale.edu/ftp/neuron/p ... ffic84.pdf

No Newton, no convergence test.
For that the variable step solvers are used.

The single step starts in nrn/src/nrnoc/fadvance.c:fadvance()
The NEURON Book has a section 7.3 devoted to it.