On niedziela, 30 marca 2008, John Griessen wrote: > DJ Delorie wrote: > > Imagine, if you will, a DIP package with a trace centered between two > > pins. The user should be able to just push the crosshair between the > > trace and one of the pins, and have the trace moved over just enough > > to make room, resulting in two traces between pins. > > > > The simple case is just nudging traces aside. The harder case is > > adding bends in the traces, such as what the global puller results in, > > or what liquidPCB does. > > A short description of the above is "radial force field from a point" > such that more force enters into a balance that decides where a point on a > trace lands. The other part of the balance would come from a model of > surface tension, a force along the length of all traces that is equal > everywhere, such that curves end up having a sideways force on them... the > second part describes "global puller". A third part could be approximate > friction -- a level you must go past to > get motion, but which adds no further drag once moving. > > With those attributes active, you could have two tools that would change > shapes of traces, one a zone of action that reduces friction some, (maybe > the mouse thumbwheeel could command how much..), the other DJ's point tool > that exerts a pushing force. > > And you could make a "broom" or "dusting-brush" from a line of point force > tools. > > Now how to code all these physics models? Hmmm...?? > > John Griessen
It sounds good:) I think that a good idea in creating this tool may be recalling the intuitioncoming from physics models, but I will try to avoid implementing it straight as a phisics model, because it may cause some complications. Moreover, it seems to me that creating depicted model with huge number of connections may be too expensive in the terms of computation. (I think that similar effects can be obtained by making this computation locally and using something which acts like this model). -- Bartlomiej Wolowiec _______________________________________________ geda-dev mailing list [email protected] http://www.seul.org/cgi-bin/mailman/listinfo/geda-dev
