I think I have my own concept of “Valve Time”. The very last thing I officially posted on this site was September 1st of last year. Good grief! Obviously I slipped my fall (previously summer) milestone by |–| this much. And my last twitter update was in November, something about remaining work on the Direct3D11 render system (which actually, I did in fact finish). Not to mention the next AssimpNet release slipped as well, although at least with that the latest version is freely available to compile yourself. Not so much with the graphics engine.
Well, I apologize for the lack of updates, but frankly it’s because there wasn’t a whole lot to update anyways (not to mention, I think I’m only updating Abe at this point, heh!). I’m not really the blogger type anyways, and lack of content makes that even worse. But hey, I try…and I still am working on the version 2.0 redesign of the engine. Although that has been a painful process due to a lack of time and me being a stubborn perfectionist. This sole developer has been very busy shipping a product at work (last month!) and also falling in love with a fellow code monkey (a gal!) . Needless to say, personal and professional priorities have taken over and often when I do have free time, I feel like just relaxing for a change.
Hah well, that’s my excuse and I’m sticking by it. Barely a day goes by that I don’t give some thought to future work on this engine; there are times where I wish I could just stop time. Ah maybe one day. Even after I finish the redesign/port to version 2.0, there’s going to be a huge piece of functionality missing, or at least radically altered (the Scene graph). My idea with that is have something basic that can be used pretty much anywhere (the concept of an Renderable, basically the mesh data) and leave most of the scene data structure to users. And have an optional framework at a higher level that implements scene management (not so much in the form of a scene graph either). I’m more of a believer in “library” than “framework”, the core library should let you pretty much build whatever type of app you want, rather than force you into a specific path.
So that’s the general philosophy I always approached this next version. Something modular and layered, where only a small core is really needed to bind everything together, and all the various systems being mix-n-match. You take what you want, and leave behind what you don’t want.
Anyways, I just wanted to update anyone who reads this that I’m still alive and kicking (and better than ever). If you remember my first milestone post, most of Milestone 1 has essentially been completed (save for the last two bullet points). Let’s see if I can actually get something out by summer this time