The current implementation of slotting in gEDA is confusing and inflexible.
It overloads the pinseq= attribute, which is also used to order pins for SPICE netlisting. It does not work when the slots are heterogeneous. This, in turn, either requires hidden power pins or redundant power pins if the user wishes to employ this mechanism to break down a slotted chip. Even if the slots are homogeneous, the designer may wish to use heterogeneous graphics for a single chip, e.g. NOR and IAND logic. The approach of getting around the slotting limitations by giving the same refdes to several instances of different symbols works fine, but users are dissatisfied with it. It's clumsy and does not lend itself well to automatic checking. I propose a new attribute, slotfile. An example might be "slotfile=2,dualopamp-2.1.slot" (the 2 refers to slot 2). The slot file format would be the same as a symbol, but "ultralight", with only graphics, pins, and attributes attached to pins. The slotfile= attribute would go into a symbol file, along with the usual other stuff. gschem and gnetlist would combine the data from the slot file with the data from the symbol to draw and connect the instance. An extended slot numbering syntax (e.g. "2.2") could be used to select alternative graphics for a slot. The old mechanism could be left in place. Many, including me, would scream if it changed. If the slot is defined by slotdef=, it's old style, by slotfile=, it's new style. I think this is a minimally invasive change that will make the slot mechanism more flexible and easier to understand. John Doty Noqsi Aerospace, Ltd. http://www.noqsi.com/ j...@noqsi.com _______________________________________________ geda-user mailing list geda-user@moria.seul.org http://www.seul.org/cgi-bin/mailman/listinfo/geda-user