Kinda late jumping in this thread.
Let's see... Thing's I'd like to see in Uru:
Visual- HDR support
- Shader materials - - Can you say specular and bump maps?
- Render to Texture - - (aka Myst V's 'DynamicReflectionCam')
- Improve the wave-sets - - using the previously mentioned features
- Improved shadow techniques - - Uru's sucks so badly
- Video / streaming textures - - I second the use of Ogg Theora
Interface- KI Interface improved (plus maybe 'DniNet'
- Nexus Interface (seriously, how could this interface be worse?)
- Books (maybe make it so they can be more easily added, and act more like a physical object)
TechicalEntity SetupI'm not much of a programmer, but it seems the very organizational scheme of Uru is somewhat.... strange. I mean, the Quab's are set up as .age's. Everything is set up as ages. And the whole way avatars are setup. Jesus what a mess. Don't the vast majority of other games use entity systems and have different types of entities, some being animals or characters, and the AI is linked up to them? Nothing (aside from the Quabs, iirc) even has AI. Say you want to add in a cool physics system where you can pick up objects like in HL2 or something, it would probably have to be an enormous hack in Uru. Not flexible at all as a real game system.
Avatar SystemWhere to even start with this? lol We all know it needs a boatload of improving.
Paging Geometry and LODI think Paging geometry and good LOD support would be a great feature for Uru to have. You could have a much better way to deal with enormous scenes, like an honest representation of D'ni.
But again, I don't know how easily you could actually add it, especially taking networking code into consideration.
Honestly, we might be better off making a new system using ogre and other good open source libraries. Leverage as much other OS stuff as possible, and cannibalize Plasma when they finally release the code. Take whatever is good, and integrate it into the new system. And then port the existing ages to work with it. And our tools would be 100% compatible with a system of our own making. Probably a much better game plan in the long term.
Best part about it is that
- No waiting for Cyan; you can START NOW!
- Fallback in case Cyan never releases the code
- Ogre can already do freaking awesome stuff, and will be freaking awesome in the future, with no effort of our own
- We can include other libraries that do awesome stuff, whenever we want (if we set up a good architecture)
- There is nothing really extraordinary about Plasma, other than Uru already uses it. And we all see how dated and rigid a game system it truly is.
- It could be an experiment to see how tenable it is. If it doesn't really work out, the original Plasma system is always there to go back to.
Although I realize how monumental a task it would be, and it's really not fair for me to say that this must be done. I wouldn't really be able to contribute to making such a system, as much as I'd like to.
But it is my general outlook on that the game systems sucks. And we only consider NOT doing a complete overhaul because it would break older ages, and because it's a huge amount of work. However, I would imagine the collective total amount of work to overhaul the system and port stuff would be less than the work needed to make hackish 'improvements' to the existing architecture, and continue using it in that form. We would pay for it in the future with the frustration everyone would experience trying to make some pretty basic stuff work in the game, and then finally decide an overhaul is necessary, after all that effort.
EDITHere's a good link for somewhere to start investigating for building a new system, for anyone thinking the same things I am here.
http://www.indiezen.org/