[
Date Prev][
Date Next][
Thread Prev][
Thread Next][
Date Index][
Thread Index]
[
List Home]
|
Re: [udig-devel] renderer
|
Vincent Heurteaux wrote:
I've tested uDig 0.4 and read the roadmap, and I'm impatient to use
the first stable release.
As are we all :-) A bit more work to go first - a lot of this week in
preparation for a Demo.
I'm planning to develop specific applications based upon uDig in a
way to be integrated in a Spatial Data Infrastructure.
That is great - it is exactly the kind work we are trying to facilitate.
Our developments are around spatial imagery, and for this reason,
we've tested different solutions including uDig and Seagis
(http://seagis.sf.net), based upon Martin's J2D renderer. I was very
impressed by this tool, and after testing uDig, my impression is that
J2D fits better for our needs (for performance reasons in raster
management, and fonctionnality like realtime resizing ScaleBar
etc...). I follow the differents geotools lists since few month, and I
can remember discussions about renderer stuff, but I don't know if you
plan to integrate J2D in a near future, or if your API will permit an
external effort to integrate J2D instead of Lite-renderer.
Our rendering framework is fairly agnotstic about the back end renderer
used - right now we use lite renderer. You could just as easily set up
J2D renderer. What I would like to see happen is each one of J2D
renderer's optimizations considered and generalized. As an example - I
really want the initial pass through the data when J2D transforms
Geometry + Style into a "Graphic" to render at the same time. The
immidiate response is important to me, I don't mind so much if J2D
caches inorder to optimize a redraw, but I need immidate gratification.
We also have some testing work starting up with JTS to try and minimize
Geometry overhead.
There is a Page on the Geotools wiki about rendering Performance
(http://docs.codehaus.org/display/GEOTOOLS/Rendering+Optimization) -
perhaps you could look that over and see if we have missed anything.
The page is editable ...
I've contacted Martin to know what are his plans about J2D
Development, he told me that J2D needs a refactoring to be geoAPI
compliant and some memory management modification. He told me he first
got a lot of work to do with GC. So, we plan to help Martin's
development process by allocating developer time on GC stuff so that
he will be able to work sooner on J2D.
That woudl be appricated - perhaps you should crash a Geotools meeting
(where both J2D and GC live) and everyone could bang out a schedule.
Meetings are today at 12:30 PST.
Thanks Vincent - I am looking forward to working with you.
Jody
