Hi all! I now (after seeing poll) completely understand that the auto-retopo is what more people that were voting want to see.
However... In my efforts to better understand Blender's mesh/data representation, I feel that doing an "easy" 3d_to_2d rasterizer for compo. tools would be better for me to start with. I don't believe doing the compo. part will take longer than a week, and it'd sure help me to get a better idea of how Blender stores data. After 2 decades of graphics programming I don't feel that either project is "outside of my ability" to do, but I am still somewhat new to Blender's codebase. Implementation of the Double Edged Matte node took 1 day, and that was my first time ever coding for Blender. Maybe I'm wrong, the mask rasterizer could end up taking longer than 1 week, but I really doubt it. As soon as I have that working correctly, I will switch development over to an implementation of the greedy mixed int. quad. retopo system. ** in reply to the widget idea ** I do think that's a great idea, I will get it working as a node first, then let someone more familiar with the other areas of blender get the code working in the other areas. If nobody picks it up and gets it to a good workflow state, then perhaps I shall take a look at it after the retopo. On Sun, Apr 10, 2011 at 6:30 PM, Daniel Salazar - 3Developer.com <[email protected]> wrote: > Pete I tough you where already working on the geometry node? What happened? > > This is a disappointing way of measuring the development priorities. > Why not asking in a BlenderArtists poll if they prefer an "improved > DAG" or if they rather prefer a "Realistic Spit Simulator!" (insert > awesome promotional videos from other software packages). If this was > the way we steer development I'd just loose all hope in improving > workflows and tools for specialized communities like the growing VFX > community in Blender. Do a google search for "blender rotobezier" to > see how much real interest from VFX artists is there. > > cheers > > Daniel Salazar > 3Developer.com > > On Sun, Apr 10, 2011 at 4:48 PM, Tom M <[email protected]> wrote: >> Did a poll at blenderartists.org >> >> http://blenderartists.org/forum/showthread.php?214657-fast-automatic-retopologizing-or-better-masking-tool-for-the-compositor-foe-live-film >> >> So no need to reply on list as to preference - discussion and voting >> can take place there. >> >> LetterRip >> >> On Sun, Apr 10, 2011 at 12:45 PM, Tom M <[email protected]> wrote: >>> For the artists - this image will make more sense to them, then the >>> technical name of the process >>> >>> http://www-i8.informatik.rwth-aachen.de/typo3temp/pics/dc5508c320.png >>> >>> LetterRip >>> >>> On Sun, Apr 10, 2011 at 12:43 PM, Tom M <[email protected]> wrote: >>>> On Sun, Apr 10, 2011 at 12:17 PM, pete larabell <[email protected]> >>>> wrote: >>>> >>>>> Now, what I'd like to ask is: >>>>> >>>>> Given that it would probably not be best to try two additions at the >>>>> same time... which one is is more desired in blender first? >>>> >>>> More than 90% of Blender artists use the modeling tools, less than 1% >>>> currently use the compositor for doing masking and other tools for >>>> compositing of live action footage.. >>>> >>>> So the user view would be pretty overwhelmingly for the 'mixed integer >>>> quadrangulator' (although to them it would be called automatic >>>> retopology). >>>> >>>> LetterRip >>>> >>> >> _______________________________________________ >> Bf-committers mailing list >> [email protected] >> http://lists.blender.org/mailman/listinfo/bf-committers >> > _______________________________________________ > Bf-committers mailing list > [email protected] > http://lists.blender.org/mailman/listinfo/bf-committers > _______________________________________________ Bf-committers mailing list [email protected] http://lists.blender.org/mailman/listinfo/bf-committers
