X-Authentication-Warning: delorie.com: mail set sender to geda-user-bounces using -f
X-Recipient: geda-user AT delorie DOT com
X-Envelope-From: paubert AT iram DOT es
Date: Thu, 27 Aug 2015 15:28:24 +0200
From: "Gabriel Paubert (paubert AT iram DOT es) [via geda-user AT delorie DOT com]" <geda-user AT delorie DOT com>
To: geda-user AT delorie DOT com
Subject: Re: [geda-user] Antifork
Message-ID: <20150827132824.GA11105@visitor2.iram.es>
References: <alpine DOT DEB DOT 2 DOT 00 DOT 1508230609370 DOT 6924 AT igor2priv>
 <55D9BDC7 DOT 4000608 AT jump-ing DOT de>
 <alpine DOT DEB DOT 2 DOT 00 DOT 1508231450350 DOT 6924 AT igor2priv>
 <CAM2RGhSC6UfCr8ixF5iuffmUhcdgVuz9nA+YjiZNk8-f7Q5r-A AT mail DOT gmail DOT com>
 <alpine DOT DEB DOT 2 DOT 00 DOT 1508231935040 DOT 6924 AT igor2priv>
 <CAC4O8c8PyLQNgxz1SP-98ZhyH5KPDmnNpiULVvdL+TMo-DhN4w AT mail DOT gmail DOT com>
 <alpine DOT DEB DOT 2 DOT 00 DOT 1508240407100 DOT 6924 AT igor2priv>
 <CAC4O8c_HdSL-3+etrvPozAZuqUX-Rqx4FGACkchttRkZsFhYPA AT mail DOT gmail DOT com>
 <CAM2RGhQMUdE_Z6OnJPTgTbkcQ4fWp6rGnrXib48S71tsGmMMSw AT mail DOT gmail DOT com>
 <mrl8ds$ee7$1 AT ger DOT gmane DOT org>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <mrl8ds$ee7$1@ger.gmane.org>
User-Agent: Mutt/1.5.21 (2010-09-15)
X-Spamina-Bogosity: Unsure
X-Spamina-Spam-Score: -0.2 (/)
X-Spamina-Spam-Report: Content analysis details:   (-0.2 points)
  pts rule name              description
 ---- ---------------------- --------------------------------------------------
 -1.0 ALL_TRUSTED            Passed through trusted hosts only via SMTP
  0.0 URIBL_BLOCKED          ADMINISTRATOR NOTICE: The query to URIBL was blocked.
                             See
                             http://wiki.apache.org/spamassassin/DnsBlocklists#dnsbl-block
                              for more information.
                             [URIs: delorie.com]
  0.8 BAYES_50               BODY: Bayes spam probability is 40 to 60%
                             [score: 0.4999]
Reply-To: geda-user AT delorie DOT com
Errors-To: nobody AT delorie DOT com
X-Mailing-List: geda-user AT delorie DOT com
X-Unsubscribes-To: listserv AT delorie DOT com
Precedence: bulk

On Wed, Aug 26, 2015 at 10:43:40PM +0200, Kai-Martin Knaak wrote:
> Evan Foss (evanfoss AT gmail DOT com) [via geda-
> user AT delorie DOT com] wrote:
> 
> >> My fantasy when I heard about ruby scripting in pcb was that perhaps
> >> Stefan's router might someday see the light of day afterall.  Oh well
> > 
> > I keep seeing mentions of this legendary router. I have been these
> > gEDA lists +10 years but I don't remember it. What was the big appeal
> > if it?
> 
> It comes up with cool looking layouts. When done properly, the technology 
> promises to be very efficient and find clever solutions. There is a little 
> demo on youtube from back in 2012 when the toporouter sort of worked:
> 	https://www.youtube.com/watch?v=NqT4ZYGB3VY
> 
> As fancy as the result looks, the toporouter never lived up to its 
> potential. The state shown off in the video has several significant 
> limitations.
> 
> * The algorithm is all-or-nothing by design. It can only applied to the 
> whole layout. You cannot tell it to route just this subset of connections. 
> 

Quite a limitation, especially for boards with analog RF signals. I've
done quite a few boards with analog RF switches: the RF signals have to
be routed very carefully (controlled impedance, nothing coming too
close, etc). But I would have liked to have an autorouter for the
control signals, which are never critical (they are quasi static,
changing them every 15 minutes would be considered extremely frequent).

> * The algorithm restricts it self strictly on one side of the board for 
> any given net. That is, when in a dead end, it is not able to insert a via 
> and switch to bottom and work from there. With thru hole components this 
> is often less problematic. But SMD without vias tends to be unroutable.

That's quite an understatement. 

> 
> * The router ignored any pre-routed copper, too. You had to start it on a 
> footprints-only layout.

At least it would need keep-out areas for where you don't want the
toporouter to affect the later parts.

> 
> * As seen in the video, more often than not, there are a few unrouted 
> nets. Typically these are hard to fix manually.

Even with vias?

> 
> The developer who did the toporouter left the project when the Google 
> summer of code he was financed by was over. Unfortunately, there was 
> insufficient documentation. And the code style was such that the router 
> was assessed as working but impenetrable by those who took a closer look 
> after the fact. Even more unfortunately, there was a major shift at the 
> core of pcb going on about this time -- the internal precision of geometry 
> was pushed from 1/1000 mil to 1 nm. Ever since this move, the toporouts 
> straight out fails.

Actually the previous resolution was 1/100 mil.

    Gabriel