Page 1 of 1

machine roundoff error

Posted: Mon Dec 15, 2014 12:08 am
by sonyadave
"The increment in the independent variable is less than machine roundoff error"

I always get the he above error when I am running my code. I am using NEuron 7.3. The error applies only to two of my channel, at certain high conductance values. I am solving with "derivimplicit". I tried 'Cnexp", and it gives the same error.

I started with a precompiled model from modelDB, but I have made lots of changes, and incorporated channels from various existing models.

Thanks,
Sonya

Re: machine roundoff error

Posted: Mon Dec 15, 2014 11:59 am
by ted
You might try uninstalling NEURON, then install from the most recent alpha installer for your platform (Mac, PC under MSWin, or Linux) from http://www.neuron.yale.edu/ftp/neuron/versions/alpha/, or get the latest development code from NEURON's mercurial repository (see http://www.neuron.yale.edu/neuron/download/getdevel) and compile that. If that doesn't fix the problem, we'll have to ask for your model source code and try to reproduce the problem in order to diagnose it.

Re: machine roundoff error

Posted: Mon Dec 15, 2014 4:35 pm
by sonyadave
Thanks. I'll try that.
Sonya