kicad-developers team mailing list archive

["jean-pierre.charras@..." <jean-pierre.charras@...>]

Vesa Solonen a écrit :
> There is some refining in polygon code I'm going to do, but first I'd
> like to get some discussion about the implementation. Jean-Pierre, are
> you going to work on arc support in the near future? That's because
> taking care for dynamic p-p error in polygons is going to be quite a lot
> more work than just small enhancements and I'd like to avoid doing work
> many times...
No.
I have some work about documentation update, code cleaning and adding 
tool tips in dialogs,
before the release of the next kicad version.
I must work on zones, but mainly to optimize dialogs and automatic zones 
refilling (using the current algoritms)
(and also add tools for microwave applications, like tracks with no 
rounded ends).
This is already a lot of work.
> First, I was going to move arc approximation starting point half angle
> increment away from Y or X axis. That would give a bit more space
> between adjacent cutouts and minimise discontinuities in fill. That's
> something I don't need help at (I guess).
>
> Vertex caching would possibly give the needed boost for performance. My
> initial thought was to convert every cutout to vertex vector. In case of
> the thermals the result would be vector array or table of four vectors.
> Positioning and maybe also rotating is done outside the individual
> primitive functions along caching. Per pin thermal topology defines
> would be nice too.
>
> I've been also thinking teardrop support by polygons for via/pad to
> track connections and if there is any pointers how to get track
> direction and width easily please pass on.
>
> Why would I like these features? Because of the etching process (sharp)
> _inner_ corners are bad. They over etch and the worst thing is that they
> tend to retain process chemicals because of surface tension and what
> not. Over etching can be controlled by clever additives, but residues
> are considered harmful for reliability. I'm also quite confident that
> microwave guys would like arcs.
>
> -Vesa
For microwave applications i am thinking tear drops are perhaps more 
important than arcs ans will be a must.
this is because transitions with a track and a pad *must* be done with 
tear drops in order to avoid signals perturbations.

--
Jean-Pierre CHARRAS

Maître de conférences
Directeur d'études 2ieme année.
Génie Electrique et Informatique Industrielle 2
Institut Universitaire de Technologie 1 de Grenoble
BP 67, 38402 St Martin d'Heres Cedex

Recherche :
GIPSA-LIS - INPG
Rue de la Houille Blanche
38400 Saint Martin d'Heres