gEDA-dev: [PATCH] GAF: Solve the "transistor problem"

[John Doty]

On Aug 7, 2007, at 11:03 AM, John Luciani wrote:

> On 8/7/07, Bernd Jendrissek <bernd.jendrissek at gmail.com> wrote:
>> On 8/7/07, John Doty <jpd at wispertel.net> wrote:
>>
>> Thanks for your feedback; I'll change to using, hmm, pinfunction= as
>> the human-readable attribute that will provide the mapping to
>> pinnumber.
>
> We briefly discussed this on Sunday. I do not remember if we came to a
> conclusion but pinlabel was suggested.

Beware that some parts have multiple pins connected to the same  
internal net. TC1411 doubles up every pin except the input! So I, at  
least, tend to have same pin label on multiple pins in cases like  
that. I'm happier with Bernd's new attribute, but I'm also unhappy  
that each pin has a growing pile of attributes.

Another way to solve this would be to allow multiple physical pin  
number assignments to one graphical pin. That would simplify some  
symbols, and allow one to switch from, say, a TO-3 to a TO-220  
footprint without having to draw an extra pin and connect it up.

Another minor annoyance is that there are times when you'd like more  
than one graphical pin to represent a single physical pin. One  
example is the ASIC pads at  http://research.kek.jp/people/ikeda/ 
openIP/openIP_2.pdf. The author drew two pins on each symbol, but  
many of them have only one connection to the underlying circuit  
(excluding global power and ground nets). His EDA system (Tanner, I  
believe) supports this, but ours doesn't. Can't have the same pinseq  
on two different pins.

John Doty              Noqsi Aerospace, Ltd.
http://www.noqsi.com/
jpd at noqsi.com