john@ghostwheel.unm.edu (John Prentice) (11/26/90)
Regarding my posting considering f2c, let me follow up before I get a thousand mail messages sent to me. Several people have tried the C version of the code on their systems and are having no problems. So, as I suggested might be the case, I must be making a dumb mistake. Also, I wish to correct any misconceptions left by my message that I regarded the converted C code as poor. I would not expect f2c to produce the cleanest C code possible. However, I still don't think you can write a C version of the Fortran root program that is as simple and clear as the Fortran version was. That was really the point I was going for but poorly stated. I am about spent out arguing the virtues of Fortran versus C for numerical work. I will continue to use Fortran for a variety of reasons until someone with significant experience in large scale scientific computing gives me a very, very good reason to change. So far nobody has done that. I appreciate all the responses I got and the time and care people took in sending them. John Prentice