Hi Sukender, On Wed, Feb 25, 2009 at 4:57 PM, Sukender <suky0...@free.fr> wrote: > And I guess simply adding an OpenGL->D3D mapper is a mistake because D3D > convinced devs would see it as a trickery/fraudulent "fake D3D".
I would wonder if "D3D convinced devs" would consider something that is not entirely D3D centric... If that's the case then the OSG abstracting away from this would make it fail at the first hurdle anyway and they won't consider it a contender. I would suggest that such D3D centric engineers will also grate against the general ethos of the OSG community, and would likely cause far more disquiet that be productive. Personally I'm far more concerned about practical solutions that deliver value rather than ticking market boxes, so if some see it as "fake D3D" but it still works well for others, delivers a solution to those that absolutely need D3D such as for XBox ports and is maintainable then I'm happy. My question is very much, is a OpenGL subset -> D3D a technically viable solution. Would it be possible to write, would it be possible to maintain, would it perform well enough, would it be more reliable than dodgy OpenGL drivers delivered by some vendors? Would this be a good enough solution for existing OSG application developers that need to find a solution that involves D3D. As for selling the OSG to the windows/xbox games centric community, we have to be careful about trying too hard to capture a market share at the extra expenses to the OSG existing core market and other potential markets. The game market is far bigger than Direct3D, although Windows games programmers might not think in these terms. I'm looking for the best bang for the buck in terms of work we do, risk to the project and the potential benefits we can gain for existing and future users. We have to be careful about how we manage evolution as not to take on too much and end up weighed down by infrastructure that adds complexity and work to maintain. If we can provide solutions that have low risks and small developer efforts then it would be wise to go for them even if the solutions aren't perfect. This is why I raise the suggestion of a OpenGL subset -> D3D wrapper, it could well be a low cost, low risk approach and gives us a solution for some market segments. Robert. Robert. _______________________________________________ osg-users mailing list osg-users@lists.openscenegraph.org http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org
