> On 2017-12-28, at 07:52, didier derny <didier@aida.org> wrote: > > I'm making a replacement for the 64k extension for the Commodore 8032 > > this board is connected on the 6502 socket and the 6502 is partly over the 2 row of pins to connect to the 6502 socket > eagle 7.5 seems unable to route that, I have to route manually most of the connections between the 6502 socket and the 6502 > in fact the tracks are turning around the socket instead of making direct connections pin to pin First of all having to route traces manually is not to be called a problem. In fact it is probably still the only way to a "decently" routed board. Unless what you do is a really simple, quick'n dirty POC type of a PCB you will have to manually route it anyway or you get "wild" results. That is, unless something has dramatically changed since EAGLE V6.x, which is what I am still using while trying to move to KiCAD. And it is not only in EAGLE. There are more expensive pieces of software that can do even worse in this field. Nothing to be surprised about. > any idea to this routing problem ? I guess I'm using the wrong library element for the 6502 socket... My guess would go to DR/Net classes. Do a DRC after routing your trace(s) manually and it might show you that it doesn't like something. Whatever it finds being a DR violation will be something the built-in logic will try to avoid. Another place to check is the autorouter settings. This is somewhat related but you might have some settings there set to values that prevent the routing the way you expect. LBL check your net classes matrix. This is the third place I know, which affects how things are routed (DRC will show manual violations of those too). In any case - as we discussed some time ago here - manual routing is where you end up anyway. Better sooner than later ;-) -- SD! - http://e4aws.silverdr.com/ Message was sent through the cbm-hackers mailing listReceived on 2017-12-28 09:01:22
Archive generated by hypermail 2.2.0.