Bootstrapping kermit... D.G. Nova RDOS
tomj at wps.com
Sat Mar 5 02:03:24 CST 2005
On Fri, 4 Mar 2005, Jules Richardson wrote:
> At least on my linux setup here, minicom uses /usr/bin/ascii-xfr to do
> ASCII transfers (which I assume is what you want in order to initially
> transfer kermit source to the Nova). Look under minicom setup (^A O) and
> then "File transfer portocols"
> Look at the man page for ascii-xfr - the '-c' flag allows you to
> specify the pause after sending each character, which sounds like what
> you want. Or the -l flag allows a delay after each line.
> Not quite the cleverness that you mention, but it should work.
Oh, it's plenty clever enough to work first time! THanks a
million! It's a general problem, talking to old machines through
the console without any flow control protocol.
So far, I've been extraordinarily lucky with this machine. It took
some doing to get fortran4 working, but fortran5 I just copied in
and it seems to work.
Downloaded Kermit from the U. Columbia site. Made by someone at
RCA using RATFOR, which outputs D.G. FORTRAN5 source. ALl I've
got's the resulting .fr source. No documentation.
I set a 20mS delay after every character sent in ascii-xfr in
minicom. I sent the file twice and FILCOM'ed, the same. Errors
compiling on fortran4, so I drag out the fortran5... compiled
first time! (THe data collection system that was the purpose of
this machine used fort5, so it's less surprising that all the
files were on the tapes; apparently fort4 wasn't used).
Linked the obj with libraries ok. the save file runs, but is
silent; Nothing Happens. COntrol-A to abort fine.
I need to debug it, which will be somewhat ugly.
Now to go write MORE in fortran4, which doesn't really like string
variables. Then I need GREP...
More information about the cctech