Ben Jackson wrote: > On Sun, Mar 01, 2009 at 01:46:40PM +0100, Stefan Salewski wrote:
>> Would you suggest to modify the symbols? > > Well, in my case I am realizing that I made a symbol that led me to miss > connecting a pin. I'm trying to learn from it, by keeping in mind that > if I put a must-connect pin next to a dozen "can connect" pins (port IO > in this case) I am more likely to miss it than if I put it in with all > the other must-connect pins like power, clock, config, etc. . . . I agree. When you have many pins, having the schematic drc find unconnected "must-connect" pins seems the only practical way...a visual check can miss is the way I think about it. One way to handle that is to make split up symbols all with same refdes, with one split up section of a chip being the must-connect pins. This makes me think of creating a "must-connect" group as attribs in a symbol, and making the drc2 backend use that to check connects. I've gotten familiarized with gnetlist now, so I might find time to take a look at that. anyone see any logic flaw with the idea? What that would do for you is allow to use a layout-oriented symbol that is just like the chip pin rows and still check for must-connects. When I do that now, it means I spend a good while checking over the pin descriptions in the datasheet as I make the schematic, and put no-connect symbols on all unused floating pins. A "must-connect" group DRC could be a time saver. John Griessen -- Ecosensory Austin TX _______________________________________________ geda-user mailing list [email protected] http://www.seul.org/cgi-bin/mailman/listinfo/geda-user
