Aurora is realy a messy messy thing, I'm not going to lie about that, it's got a basturdised half object oriented half procedural structure (the DX code is nearly all procedural, the GUI is all OO, the internal pof handeling stuff is... nether) that even I find absurd, Aurora was suposed to be for me mostly a learning experience, todays HT&L upgrade is a _direct_ decendent of what I learned from Aurora, now it does have a realy cool (IMHO) interface, if somewhat sloppy, another thing Aurora was suposed to be was my own personal customized suited to my desires and specifications POF editor designed from the ground up for expandability (new data chuncks) to this end it has suceeded, I can probly add a new data chunk in about an hour or two even after not playing about with it much over the last nearly year, and there is sort of a drag and drop interface, and I have some half fineshed polgon editing capabilities (UV editing and normal editing)
I am quite tempted to start over with a better design for the DX code and better integration of it into the MFC GUI code, but I know it would be better for practical reasons to try and ether fix Aurora's many bugs (I think i fixed the biggest one were it would overwrite what ever was in the pof with what ever was on the screen when it loaded a new model, wich when you loaded a new model would erase the mass and moment of inerta data, link in sig, test it) or focus on FSO entierly
I don't think there is anything other than concepts that Kazan can or would want to take from Aurora (especaly becase my renderer is DX and his is OGL) about the only thing that he realy could use would be the bit of code that takes the jumbled up FS BSP data structure and straitens it out for faster rendering (HT&L) and even then he'd basicly have to rewrite it (becase it's in DX not OGL).
now it is a very cool idea thinking of PCS2.0, a PCS powered pof editor with a better interface, something closer to what I wanted Aurora to be, but Kazan's time is quite limited, a situation I know all to well, would his limited time be best spent makeing a totaly new interface (I'm assuming he isn't going to rebuild his POF handeling class) for PCS, improveing the exsisting PCS code (there are a few things I'd still like to see, there is the auto-faceting thing I was talking about before, and check the path options I made for Aurora), or makeing improvements to his FS_Open network code (wich from what I've seen seems to work quite well) (or he could take a crack at implementing OGL's HT&L code wich is currently just stubbed out)
so improvements to the PCS interface (wich I like quite a bit as it is now), the biggest improvement would be integration of the render window into the GUI (I would like something that looks sort of like Aurora, obviusly, with a render window on one side and a tab control (that would look esentaly identical to the exsisting PCS interface) on the other) a bank of render control options somwere, an option for showing the model in full window or full screen mode would be cool,
I think I may photochop something together
