6:25 a.m.
Hi,
We'll be discussing Advanced Gradients in today's SVG Working Group
meeting. As an experiment, I've made a modified version of Inkscape that
can read in the proposed syntax for Mesh Gradients and display the
meshes. You can look at the results at:
http://tavmjong.free.fr/SVG/MESH/Mesh_Inkscape.html
Tav
7:40 a.m.
Hey,
Good luck! It looks very promising. Very good to know you've still
been pursuing this!
Cheers,
Josh
On Fri, Jul 29, 2011 at 6:25 AM, Tavmjong Bah <tavmjong@...8...> wrote:
Hi,
We'll be discussing Advanced Gradients in today's SVG Working Group
meeting. As an experiment, I've made a modified version of Inkscape that
can read in the proposed syntax for Mesh Gradients and display the
meshes. You can look at the results at:
http://tavmjong.free.fr/SVG/MESH/Mesh_Inkscape.html
Tav
------------------------------------------------------------------------------
Got Input? Slashdot Needs You.
Take our quick survey online. Come on, we don't ask for help often.
Plus, you'll get a chance to win $100 to spend on ThinkGeek.
http://p.sf.net/sfu/slashdot-survey
_______________________________________________
Inkscape-devel mailing list
Inkscape-devel(a)lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/inkscape-devel
7:43 a.m.
Looks fantastic! Will have a play this weekend with them!
Sent from my iPhone
On 29 Jul 2011, at 14:25, Tavmjong Bah <tavmjong@...8...> wrote:
Hi,
We'll be discussing Advanced Gradients in today's SVG Working Group
meeting. As an experiment, I've made a modified version of Inkscape that
can read in the proposed syntax for Mesh Gradients and display the
meshes. You can look at the results at:
http://tavmjong.free.fr/SVG/MESH/Mesh_Inkscape.html
Tav
------------------------------------------------------------------------------
Got Input? Slashdot Needs You.
Take our quick survey online. Come on, we don't ask for help often.
Plus, you'll get a chance to win $100 to spend on ThinkGeek.
http://p.sf.net/sfu/slashdot-survey
_______________________________________________
Inkscape-devel mailing list
Inkscape-devel(a)lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/inkscape-devel
12:38 p.m.
On Friday, July 29, 2011, 3:25:52 PM, Tavmjong wrote:
TB> We'll be discussing Advanced Gradients in today's SVG Working
Group
TB> meeting.
It was a good discussion and your write up of your experiments was very helpful.
TB> As an experiment, I've made a modified version of Inkscape that
TB> can read in the proposed syntax for Mesh Gradients and display the
TB> meshes.
Is it possible to get hold of the modified source so that tI can experiment further?
TB> You can look at the results at:
TB> http://tavmjong.free.fr/SVG/MESH/Mesh_Inkscape.html
--
Chris Lilley Technical Director, Interaction Domain
W3C Graphics Activity Lead, Fonts Activity Lead
Co-Chair, W3C Hypertext CG
Member, CSS, WebFonts, SVG Working Groups
12:42 p.m.
On Fri, 2011-07-29 at 21:38 +0200, Chris Lilley wrote:
On Friday, July 29, 2011, 3:25:52 PM, Tavmjong wrote:
TB> We'll be discussing Advanced Gradients in today's SVG Working
Group
TB> meeting.
It was a good discussion and your write up of your experiments was very helpful.
TB> As an experiment, I've made a modified version of Inkscape that
TB> can read in the proposed syntax for Mesh Gradients and display the
TB> meshes.
Is it possible to get hold of the modified source so that tI can experiment further?
TB> You can look at the results at:
TB> http://tavmjong.free.fr/SVG/MESH/Mesh_Inkscape.html
Yes, give me a day or two and I'll put in in a publicly available place.
Tav
2:01 p.m.
On 29/07/11 14:25, Tavmjong Bah wrote:
Hi,
We'll be discussing Advanced Gradients in today's SVG Working Group
meeting. As an experiment, I've made a modified version of Inkscape that
can read in the proposed syntax for Mesh Gradients and display the
meshes. You can look at the results at:
http://tavmjong.free.fr/SVG/MESH/Mesh_Inkscape.html
Tav
Have the merits of mesh gradients been discussed over the merits of
diffusion curves?
http://wiki.inkscape.org/wiki/index.php/Diffusion_Curves
http://artis.imag.fr/Publications/2008/OBWBTS08/
Which of the two concepts is more powerful?
Joel
1:33 a.m.
On 2011-07-29 23:01, Joel Holdsworth wrote:
On 29/07/11 14:25, Tavmjong Bah wrote:
> Hi,
>
> We'll be discussing Advanced Gradients in today's SVG Working Group
> meeting. As an experiment, I've made a modified version of Inkscape that
> can read in the proposed syntax for Mesh Gradients and display the
> meshes. You can look at the results at:
>
> http://tavmjong.free.fr/SVG/MESH/Mesh_Inkscape.html
>
> Tav
>
Have the merits of mesh gradients been discussed over the merits of
diffusion curves?
http://wiki.inkscape.org/wiki/index.php/Diffusion_Curves
http://artis.imag.fr/Publications/2008/OBWBTS08/
Which of the two concepts is more powerful?
Neither :) Both have their strengths
and weaknesses. In general
diffusion curves as currently defined have more of a tendency "to do the
right thing" and are WAY easier to handle. On the other hand, mesh
gradients (especially tensor product patches) give more control. Now,
there are some ideas floating around to enhance diffusion curves to the
point where there might be little benefit to using meshes, but even so,
given practical considerations and the wide spread use of meshes it will
no doubt be useful to support meshes as well.
Also, keep in mind that diffusion curves are relatively new and there
are still some issues in rendering them (it should be possible, but so
far no renderer has been made, that I'm aware of, that is both fast and
accurate). Also, while they often "do the right thing" there are some
definite issues with non-color boundaries (the current solution is
definitely not optimal) and they can have the tendency to over-smooth
certain images.
1:50 a.m.
On Jul 31, 2011, at 1:33 AM, Jasper van de Gronde wrote:
Neither :) Both have their strengths and weaknesses. In general
diffusion curves as currently defined have more of a tendency "to do the
right thing" and are WAY easier to handle. On the other hand, mesh
gradients (especially tensor product patches) give more control. Now,
there are some ideas floating around to enhance diffusion curves to the
point where there might be little benefit to using meshes, but even so,
given practical considerations and the wide spread use of meshes it will
no doubt be useful to support meshes as well.
Also, keep in mind that diffusion curves are relatively new and there
are still some issues in rendering them (it should be possible, but so
far no renderer has been made, that I'm aware of, that is both fast and
accurate). Also, while they often "do the right thing" there are some
definite issues with non-color boundaries (the current solution is
definitely not optimal) and they can have the tendency to over-smooth
certain images.
Also keep in mind visual issues.
In all the examples I have seen, diffusion curves have had a distinct, if sometimes
subtle, unique look to them. Something akin to looking at a watercolor and being able to
say "the artists used watercolors to do this painting".
I've not seen the same for general gradient meshes.
4:56 a.m.
New subject: Re : Mesh Gradients
It would be very interesting to see what gradient meshes do with color opacity !
ivan
________________________________
De : Jon Cruz <jon@...18...>
À : Jasper van de Gronde <th.v.d.gronde@...528...>
Cc : inkscape-devel(a)lists.sourceforge.net
Envoyé le : Dimanche 31 Juillet 2011 10h50
Objet : Re: [Inkscape-devel] Mesh Gradients
On Jul 31, 2011, at 1:33 AM, Jasper van de Gronde wrote:
Neither :) Both have their strengths and weaknesses. In general
diffusion curves as currently defined have more of a tendency "to do the
right thing" and are WAY easier to handle. On the other hand, mesh
gradients (especially tensor product patches) give more control. Now,
there are some ideas floating around to enhance diffusion curves to the
point where there might be little benefit to using meshes, but even so,
given practical considerations and the wide spread use of meshes it will
no doubt be useful to support meshes as well.
Also, keep in mind that diffusion curves are relatively new and there
are still some issues in rendering them (it should be possible, but so
far no renderer has been made, that I'm aware of, that is both fast and
accurate). Also, while they often "do the right thing" there are some
definite issues with non-color boundaries (the current solution is
definitely not optimal) and they can have the tendency to over-smooth
certain images.
Also keep in mind visual issues.
In all the examples I have seen, diffusion curves have had a distinct, if sometimes
subtle, unique look to them. Something akin to looking at a watercolor and being able to
say "the artists used watercolors to do this painting".
I've not seen the same for general gradient meshes.
------------------------------------------------------------------------------
Got Input? Slashdot Needs You.
Take our quick survey online. Come on, we don't ask for help often.
Plus, you'll get a chance to win $100 to spend on ThinkGeek.
http://p.sf.net/sfu/slashdot-survey
_______________________________________________
Inkscape-devel mailing list
Inkscape-devel(a)lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/inkscape-devel
10:05 a.m.
New subject: Re : Mesh Gradients
>>>> "IL" == Ivan Louette
<ivan_louette@...48...> writes:
IL> It would be very interesting to see what gradient meshes do with
IL> color opacity !
PDF supports that, so you can test that out by playing with them.
Ghostscript can even do it in postscript using the ops which GS
uses to render PDF, such as:
.pushpdf14devicefilter
.begintransparencygroup
.setopacityalpha
.setshapealpha
.endtransparencygroup
.poppdf14devicefilter
(taken from gs/examples/transparency_example.ps).
-JimC
--
James Cloos <cloos@...1016...> OpenPGP: 1024D/ED7DAEA6
2:09 a.m.
New subject: Re : Mesh Gradients
On 31/07/11 21:26, Ivan Louette wrote:
It would be very interesting to see what gradient meshes do with
color
opacity !
ivan
Here are a couple of examples from cairo. The first example contains two
patches in the mesh. The second contains one patch that folds over on itself.
http://people.freedesktop.org/~ajohnson/mesh/mesh-gradient.png
http://people.freedesktop.org/~ajohnson/mesh/mesh-gradient-overlap-transp...
4513
days inactive
4516
days old
inkscape-devel@lists.inkscape.org
10 comments
10 participants
participants (10)
-
Adrian Johnson -
Chris Lilley -
Ivan Louette -
James Cloos -
Jasper van de Gronde -
Joel Holdsworth -
John Cliff -
Jon Cruz -
Josh Andler -
Tavmjong Bah