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: Fri, 27 Jun 2014 13:05:47 +0200
From: Gabriel Paubert <paubert AT iram DOT es>
To: geda-user AT delorie DOT com
Subject: Re: [geda-user] pcb: Patch for arcs with different radii for x and y
 on screen
Message-ID: <20140627110547.GA21723@visitor2.iram.es>
References: <ojbt218fi23of8gryausfclb DOT 1403353431756 AT email DOT android DOT com>
 <201406211627 DOT s5LGR0aR004148 AT envy DOT delorie DOT com>
 <1403799508 DOT 25929 DOT 8 DOT camel AT pcjc2lap>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <1403799508.25929.8.camel@pcjc2lap>
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
  0.8 BAYES_50               BODY: Bayes spam probability is 40 to 60%
                             [score: 0.5011]
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 Thu, Jun 26, 2014 at 05:18:28PM +0100, Peter Clifton wrote:
> On Sat, 2014-06-21 at 12:27 -0400, DJ Delorie wrote:
> > > I "think" if gerber supports them, then getting arbitrary axis
> > > directions will still be awkward / impossible without approximation.
> > 
> > Gerber doesn't support them.  At least, not reliably across fabs.  We
> > approximate them with line segments.
> 
> Speaking of Gerber - at some point, we should re-validate our output
> choices against current UCAMCO recommended practice.

Indeed.

> 
> I would suggest that we re-enable arc features in gerbers at some point,
> and see if we identify any fabs which still have issues. (They ought to
> fix their end, not be a reason to reduce the fidelity of our output).
> 
> 
> I know we do polygons "wrong", with stitching of pieces without holes. I
> vaguely recall that we used to do it the "right" way, but changed for
> sake of compatibility with bad fabs.

And this actually caised me issues, I had to keep an old version
of pcb while I finished a project.

> 
> 
> UCAMCO have recently issued a white-paper stipulating the "proper" way
> is to alternate positive and negative layers, I can see how this would
> help CAM software to process things nicely, and as a benefit - it would
> speed our output quite a bit for complex boards.

I completely agree.

> 
> Similarly - do we stick to their recommended best practices of ONLY
> creating pads using a single flash of a single aperture? 

No. In the Gerber files I produce, the only items which are flashed
are vias (and perhaps square pads).

> The white-paper
> citing this recommendation explains that this structure is used to allow
> the CAM tool to extract pad locations more reliably, and that ATE flying
> probe testers will use this extracted data to identify probe points.
> 

Agreed.

	Gabriel