- From: Doug Schepers <doug.schepers@vectoreal.com>
- Date: Sat, 19 May 2007 04:16:19 -0400
- To: MenTaLguY <mental@rydia.net>
- Cc: www-svg@w3.org
Hi, MenTaLguY- This is something that has been talked about for a while within the group. We'd like to do it. Most of our recent effort has been concentrated on such things as the test suites, the errata, and polishing up SVGt1.2, but we've been moving toward new SVG Full features (e.g. the Filters spec). I'll read up on gradient meshes and see if I can come up with a proposal (if someone else in the SVG WG doesn't already have one). Your input is also very welcome, either to this list or to me personally. MenTaLguY wrote: > Along these lines... > >> 1. establish requirements > > For us, in Inkscape, the main requirements we have are: > > - sufficient flexibility to approximate the result of any shading function > - ease of implementation This point is very important to me. I want this widely implemented, so I can use it. Similarly, I would add: ease of authoring. While it's not as important for those using a tool such as Inkscape, hand-authorability is critical. > - potential for hardware acceleration > - some compatibility with PS/PDF shading model That's sensible, since there is a market need to export SVG to PDF, and to export SVG from Illustrator, etc. A key thing to be conscious of, though, is the need for this to be a royalty-free algorithm. If the PS gradient mesh is patent-encumbered, we will need to steer clear (or obtain a royalty-free license from Adobe, which is also possible). >> 2. identify existing uses of gradient meshes (e.g. PostScript Type 4, 5, >> 6 and 7 shading patterns) > > Something along the lines of type 4 shading patterns would meet all of our own > criteria, though the way PostScript requires the mesh geometry to be > specified is a little weird. > > What are your own thoughts/requirements? Regards- -Doug
Received on Sunday, 20 May 2007 10:24:56 UTC