I have a question, in an area I know less than nothing about, so I fully expect rolled eyes and sighs of dismay from those in the know...
Given that the Plasma engine is as complicated as it is, and even Cyan is having to reverse engineer it as part of open sourcing it...
Would it make more sense from a time/energy/money investment sense, to start from near-scratch with an Open Source engine, like OGRE or Crystal Space?
6 comments:
Yep.
Yes.
Indubitably!
(actually I think no but couldn't interrupt the trend :-D )
Easier to pull teeth out of a cow.
Given that it might be, and again keeping in mind that I'm a utter lay person, what would be the challenges and barriers to doing so?
Not easy at all.
Plasma has a very distinctive "style" of rendering. You'd have to rewrite an awful lot of the very core rendering and blending techniques to get it to look the same. The texture blending algorithm mf implemented is possibly the best I have seen in any engine to date. Lots of these techniques aren't documented anywhere else other than in Plasma itself so you'd have to reverse-engineer it anyway to find out just how they did certain things. Every engine has different features, so you'd have to recode a bunch of features just to maintain compatibility parity. Witness the effort it took Cyan to change just the physics engine, keeping everything else the same. :P
Short version - if you want it to still look, behave, and act like Uru... hell no it'll be FAR more difficult to shift over from something else than to keep on doing things as they go. If you don't care about losing Uru's unique visual style and having it inherit another game's style, go for it. But not even the latest CryENGINE engine (FarCry/Crysis) is capable of producing the same visual style... it'd look different.
Post a Comment