Re: gEDA-user: PCB: AutoRouter

[Anthony Blake <tonyb33@xxxxxxxxx>]

Stefan Salewski wrote:
> On Sat, 2009-11-21 at 18:32 +0100, myken wrote:

>> I was just wondering, is there any documentation on how the autorouter
>> of PCB is implemented? Restrictions, strategy, information used, hooks,
>> alternatives, etc. 

> For latest PCB snapshot 20091103 the "old" autorouter is strongly
> improved and documented in user manual.

Not only that, there is a lot of really good documentation on the 
implementation of the autorouter (thanks to Harry I think). The main 
file, autoroute.c, is explained with comments the whole way through, and 
at the top, just after the license, a paper is cited which details the 
general approach. I remember Harry saying that he does a few things 
differently than as described in the paper, but I'm pretty sure he has 
explained any differences in the comments.

And if there is anything not covered in the docs, specific questions 
would probably be answered by Harry or someone else who might know, so 
just ask away.

> The new topological autorouter is available too [...] 
> There is not much documentation for it, and it may fail

Yes, if you want to try the toporouter, be prepared to spend way more 
time hacking on the code than capturing schematics or laying out the 
board. Having said that, I have used it for a couple of small homemade 
boards with great results.. e.g., 
http://www.wand.net.nz/~amb33/acutime_interface.png - I was able to keep 
that board to a single layer with no vias.

At the top of the code there are a few papers cited which should 
probably be read and understood before getting into the code.

If there are any questions about the toporouter algorithms, or if anyone 
is interesting in contributing, just ask.

Cheers,
Anthony




_______________________________________________
geda-user mailing list
geda-user@xxxxxxxxxxxxxx
http://www.seul.org/cgi-bin/mailman/listinfo/geda-user