Doug Twilleager - aka yensid - will be joining the Core team as a developer of the jME 3.0 branch.
Some quick certifications:
- Doug has 15 years of scene graphs architecture experience
- He was the implementation architect for Java 3D, and helped the guiding principles of a number of its follow on clones
- He is a lead developer of Project Wonderland, head of all jME related work there.
Thanks to Project Wonderland, he will be able to work on jME as part of his work, as a lot of progress made with either project can interchangeably benefit the other. Doug will be starting work as soon as Siggraph has ended, and he can say with certainty that he should be able to commit to six months of working with jME and most likely longer.
I recently had about an hour long conversation with Doug on IRC. For anyone interested, I put together a brief summary of the topics we touched on. As his schedule clears up, he will be more closely in touch with Kirill to discuss engine architecture, and of course the community will start noticing his presence as well.
<erlendsh> have you had much time to look into what Kirill [Momoko_Fan] has accomplished so far?
<dtwilleager> Not as much as I would like. We are about to feature
freeze our current release - and then I'm off to Siggraph next week. After
that, then I'll deep dive into it.
<erlendsh> any particular areas you see yourself diving into? I suppose primarily you will attempt to accommodate MT Game's best interests?
<dtwilleager> There are a few areas that I'd like to see improved.
- One is threading. The threading problems are a big issue for us - that's why it becomes a high priority for me.
- Then there is the seperation of rendering state from the core
scene graph.
- I'd also like to see if we can get a solid API for OpenGL 3.0.
- Also, some advanced rendering techniques are cumbersome in 2.0, so buffer management/unification would be a big help.
<@erlendsh> threading is definitely a point of issue for both you and Kirill. is this the sort of thing you might be able to collaborate on without too much conflict?
<dtwilleager> I think we'll be fine. I have a new technique in mind that solves the threading problem without duplicating the graph.
<dtwilleager> Only the proprietary engines are doing it the way we want to. I explained MT Game to the PS3 team, and they confirmed that they are doing the same thing.
<@erlendsh> now, in terms of becoming an official developer of the team, I hope you have no qualms with getting involved with some PR once in a while? ;) I mean, basically just making an effort to keep the community in the loop regarding your personal, and team-wise efforts and progress
<dtwilleager> No problem. We have already pledged our support of
jME and see this as in line with that. We have been less vocal over the
past year because we have been building out our system.
<dtwilleager> Those outlets [blogs, twitter etc.] sound great - I will be ramping up my technical blogs soon.
<@erlendsh> [Compilation of questions:] How will the rest of the WL team or Sun affiliates be involved with jME?
<dtwilleager> Other folks may want to poke around, but they would have to consult with me. We did that up until now, and I think it works. It keeps the architecture in tact.
