delorie.com/archives/browse.cgi   search  
Mail Archives: geda-user/2014/01/15/02:52:54

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: Wed, 15 Jan 2014 08:51:26 +0100
From: Gabriel Paubert <paubert AT iram DOT es>
To: geda-user AT delorie DOT com
Subject: Re: [geda-user] More about the polygon bug
Message-ID: <20140115075126.GB23770@visitor2.iram.es>
References: <1389724747 DOT 2125 DOT 14 DOT camel AT AMD64X2 DOT fritz DOT box>
MIME-Version: 1.0
In-Reply-To: <1389724747.2125.14.camel@AMD64X2.fritz.box>
User-Agent: Mutt/1.5.20 (2009-06-14)
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: 0.5000]
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

On Tue, Jan 14, 2014 at 07:39:07PM +0100, Stefan Salewski wrote:
> The second file is the simplest test case I could come up with: a 
> > single 0603 resistor and a single line segment. It can't get much 
> > simpler than that.
> 
> Again, after applying the vim macro the bug is gone.
> 
> Even more interesting: As we have seen, the bug temporary disappears
> when we delete a line and insert it again by undo operation. Obviously
> next step: generate gerbers after undo. My result: Bug is gone. So my
> guess is, it may really be more something like a rounding bug than a
> wrong (dicer?) algorithm. Other idea: It depends on how elements are
> stored inside pcb program, i.e. undo/redo may move elements to bottom of
> a list. 
> 
> I still suffer from the GL bug -- after a few times loading pcb files
> PCB program refuses to draw something, even after restarting, and
> glxgears was extremly slow, so that I had to restart my computer. That
> may be a real difficult to detect bug. The polygon bug should be not too
> difficult for someone with some knowledge of internal PCB program, i.e.
> what happens when undo/redo occurs.
> 
> Gabriel, please test your gerbers after undo/redo.

They are fine. The problem is that, for seom projects, I have a Makefile 
based workflow in which I use pcb -x to generate the Gerber and this
just breaks my workflow. I've only hit this bug on one board, a simple
two layer board but with ground plane for EMI reasons. I've had no problem
on several 6 layer boards, but there are no planes on the external layers,
and the internal layers only have vias, not pads.

Note that it seems that one characteristic of the bug is that the
line width + twice the clearance (i.e., the "antiline width") 
is exactly the width of the pad. But the coordinate system in PCB
uses integers (a good idea to avoid floating point pitfalls).
I've tested both 32 bit (not x86, ppc) and 64 bit systems and
the bug shows in both.

	Best regards,
	Gabriel

- Raw text -


  webmaster     delorie software   privacy  
  Copyright © 2019   by DJ Delorie     Updated Jul 2019