W3C home > Mailing lists > Public > www-svg@w3.org > June 2007

Re: [Inkscape-user] Equivalent to Illustrator's Gradient Mesh?

From: Rikkert Koppes <rikkert@finalist.com>
Date: Mon, 25 Jun 2007 09:55:48 +0200
Message-ID: <467F7504.9050109@finalist.com>
To: MenTaLguY <mental@rydia.net>
CC: Chris Lilley <chris@w3.org>, adam@emiasys.com, www-svg@w3.org, bulia byak <buliabyak@gmail.com>

I am not quite aware of what gradient functions do exist, but maybe a 
potential field approach might be useful.
This would be equivalent to say calculating an electric field potential 
when a charge distribution is given. This can directly be translated to 
an intensity distribution. This should also result in "smooth" results 
as is the case with an electric field.

Rikkert Koppes (mophor)

MenTaLguY wrote:
> On Sat, 2007-06-16 at 14:17 +0200, Chris Lilley wrote:
>   
>> I would be interested to see your results. What artefacts are you
>> seeing - sharp edges (I would not have expected that from Gouraud)?
>>     
>
> The source of the problem is second-order discontinuity at the edges of
> the triangles (i.e. the color is continuous, but its rate of change is
> not).  The general visual effect is of "rays" connecting the vertices of
> the triangulation, making the triangles fairly evident:
>
>  http://www.rydia.net/~mental/trymesh.png
>
> So, unless we want to force the SVG author to manually create very
> finely divided meshes (which bulia rightly vetoed), we will need to
> select a different shading function which takes into account neighboring
> triangles and can give us higher-order continuity.
>
> The "friendlier" function can be then approximated in OpenGL or
> Postscript by automatic subdivision.  [Actually, with OpenGL on modern
> hardware, the shading function could probably even be implemented
> directly as a fragment shader.]
>
> Here's the current state of the test harness (shown above in the
> screenshot); I'm not too proud of the code, but I may as well post it so
> people can start playing with it.  The code is public domain.
>
>  http://www.rydia.net/~mental/trymesh-20070622.tar
>
> Planned still are:
>
>  - an "eyedropper tool" for picking node colors
>  - the ability to load a background image
>  - miscellaneous UI improvements
>  - alternate gradient functions
>
> [I do have a couple alternate functions buried in the code already, but
> they're toys for testing rather than being useful artistically, and the
> automatic subdivision code still needs work.]
>
> I'll post newer archives periodically, but the very newest version will
> always be available via git:
>
>  git://git.inkscape.org/projects/trymesh.git
>
> Patches are very welcome.  README follows at the end of the email.
>
> -mental
>
> ------------------------------------
>
> REQUIREMENTS
>
> "trymesh" has the following requirements:
>
>  - a C compiler (gcc)
>  - Ruby 1.8
>  - Ruby/Gnome2
>  - Ruby/OpenGL
>  - Ruby/GtkGLExt
>  - libgts
>  - RubyGems
>  - RubyInline
>  
> If you're on Ubuntu, these can be satisfied by installing
> the following packages:
>
>  gcc-4.1
>  ruby1.8
>  ruby1.8-dev
>  libgnome2-ruby
>  libopengl-ruby
>  libgtkglext1-ruby
>  libgts-dev 
>
> If you have a recent enough version of Ubuntu, you should
> be able to install RubyGems via the 'rubygems' package; otherwise
> you will need to download RubyGems from:
>
>  http://rubyforge.org/frs/?group_id=126
>
> Unpack the archive, and run the following in the unpacked
> directory:
>
>  sudo ruby setup.rb
>
> At that point, you can install RubyInline via:
>
>  sudo gem install RubyInline
>
> RUNNING
>
>  ruby bin/trymesh
>
> USAGE
>
> Left-click to select a vertex; drag it to reposition it.  Double-click
> to create a new vertex with the color of the most recently selected
> vertex.  Right-click to delete a vertex.  The sliders at the bottom
> control the color of the currently selected vertex.
>
> There are three toggles at the top of the window that control what is
> displayed:
>
>  [G] - displays the color gradient
>  [T] - displays the edges of the triangulation
>  [V] - displays the vertices
>
>
>   
Received on Monday, 25 June 2007 07:55:33 GMT

This archive was generated by hypermail 2.3.1 : Friday, 8 March 2013 15:54:37 GMT