Hi,
What is the current state of gpgpu acceleration in Inkscape? and where can I find the docs for this functionality?
Regards,
Ken
On Sun, Nov 6, 2011 at 4:06 AM, Ken O'Brien wrote:
Hi,
What is the current state of gpgpu acceleration in Inkscape?
Not implemented
and where can I find the docs for this functionality?
Nowhere :)
Alexandre Prokoudine http://libregraphicsworld.org
I see.
What do the developers think would be a best starting point for accelerating Inkscape with gpgpu? Any particular function? or component?
Also, is CUDA an acceptable technology licensing-wise?
On 06/11/11 05:21, Alexandre Prokoudine wrote:
On Sun, Nov 6, 2011 at 4:06 AM, Ken O'Brien wrote:
Hi,
What is the current state of gpgpu acceleration in Inkscape?
Not implemented
and where can I find the docs for this functionality?
Nowhere :)
Alexandre Prokoudine http://libregraphicsworld.org
RSA(R) Conference 2012 Save $700 by Nov 18 Register now http://p.sf.net/sfu/rsa-sfdev2dev1 _______________________________________________ Inkscape-devel mailing list Inkscape-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/inkscape-devel
On Wed, Nov 9, 2011 at 5:33 PM, Ken O'Brien wrote:
What do the developers think would be a best starting point for accelerating Inkscape with gpgpu? Any particular function? or component?
AFAIK, the plan was to implement SVG Filters in OpenCL.
Also, is CUDA an acceptable technology licensing-wise?
Nope. At least not as a primary way of rendering.
Alexandre Prokoudine http://libregraphicsworld.org
On 09-11-11 14:38, Alexandre Prokoudine wrote:
On Wed, Nov 9, 2011 at 5:33 PM, Ken O'Brien wrote:
What do the developers think would be a best starting point for accelerating Inkscape with gpgpu? Any particular function? or component?
AFAIK, the plan was to implement SVG Filters in OpenCL.
Also, it might make sense to look at Cairo instead of Inkscape. Perhaps not initially, as filters are more of a performance problem than rasterization, but in the long run it could definitely help (also in terms of trying to keep everything on the GPU as much as possible, instead of constantly round-tripping the data).
participants (3)
-
Alexandre Prokoudine -
Jasper van de Gronde -
Ken O'Brien