Re: gEDA-user: DRC vs shorted nets?

["harry eaton" <bumpelo@comcast.net>]

Bob Paddock wrote:

>> Netlist shorts  are NOT DRC errors.

>I expected the error to be a DRC because the optimizer moved traces, and
the
>error was not there prior to moving traces.  The connectivity of the traces
>did not change, only their location.

Their location changed in such a way that the connectivity was changed. In
your
sourceforge example, it placed a track right across the two pads of C101.
This
is a nice solid short, not a tenuous one, thus there is no DRC, only a
short.

>> There are connectivity errors which are checked by the rats nest
function.

>Why do these connectivity errors  only show up when the design is
completed?
>At that point there are no rats.  Should such short not show up all of the
>time?

Shorts should be reported whenever you check for them with a ratsnest
command
*if they exist* if you don't get a report of shorts with the ratsnest
command, it
means there are no shorts; When they are reported it's because they are
there!
(excepting for the possibility of bugs in the code, which at the moment no
valid
examples of false positives or false negatives have been given).

harry