Vector Images and the SVGL stack

[Richard Gaskin]

-hh wrote:

 >> R.G. wrote:
 >>
 >> The difference between a true Bezier object and emulating Beziers
 >> with polygons is resolution.
 >
 > I see it like this, a little bit different.
 >
 > A "true" Bezier object is a mathematical model, for thinking and
 > abstract computations, but not realizable in our deterministic world.
 >
 > The graphical output for such an object is **always** a raster of
 > points, an approximation by a polygonal line ("emulated" in your
 > terms) no matter if SVG, HTML5, postscript/pdf or LC give the input.
 > If you look at the output with a microscope you still see the pixels
 > of Super Mario.
 >
 > In LC you have to give the raw list of points, the others above use
 > for short the mathematical model *as abstract description* for the
 > raster they wish, that is then 'optimally' prepared for the device
 > by algorithms of the library.
 >
 > So the difference is for me nothing more than the method of giving
 > the directive to the CPU or graphics processor, supported by a more
 > or less large library. In LC you have until now to write the
 > algorithm by yourself.
 >
 > So, of course you are right that such a library would be comfortable.

As often happens with you and I, we rarely use the same words but 
frequently agree.  Each of us has described the same rendering 
mechanics, but in different ways.

Of course whether re-rendered on resize in a script or in C any SVG can 
be made to look good.  But given how much faster C is, and since SVG is 
already provided in LC's underlying Skia, why not use it?

Mark Waddingham has already submitted the pull request, so at this point 
it's not like we need to wait much longer to have a fairly optimal SVG 
rendering option in LC.

-- 
  Richard Gaskin
  Fourth World Systems
  Software Design and Development for the Desktop, Mobile, and the Web
  ____________________________________________________________________
  Ambassador at FourthWorld.com                http://www.FourthWorld.com