On Tue, Jun 21, 2011 at 1:50 PM, Andrew Poelstra <as...@sfu.ca> wrote: > On Tue, Jun 21, 2011 at 01:29:02PM -0700, Russell Dill wrote: >> >> I've always worried about this too, especially when doing a design in >> metric and sending it to an imperial board house. My guess is you need >> to calculate the worst case error and add that to the DRC. The >> difficult part would be finding all the possible sources of roundoff. >> >> Perhaps if you could translate the design to a mil grid with a given >> precision and *then* run DRC. >> > > We are working on moving pcb toward metric base units -- then > a mm would be 10^6 nm rather than "about 3937.00787 cmils" > like we have now. > > mils would also be an integer (254) number of base units, so > hopefully a lot of precision issues will go away.
If one point is on 228600nm and another on 126988nm, it would seem to make a 4 mil (101600nm) trace/space rule (101612nm). But when converted to mils: 228600nm = 9 mil 126988nm = ~4.99952756mil If you output to a X.YYY gerber format, that will round to 5mil and your features will then only be 4 mil apart. _______________________________________________ geda-user mailing list geda-user@moria.seul.org http://www.seul.org/cgi-bin/mailman/listinfo/geda-user