after a long time of absence I'm back, only for a short report. My old user name was erin, but I forgot my account data. So I created a new account with the new user name erintwo.
I was really astonished about seeing that there is still some traffic on this forum. All the old developers have disappeared and the ETR procject seems to be dead. But the few posts signalize that the Tuxracer idea is not dead yet. I think there are still a lot of persons who are interested in this game.
Someone asks if the development is going on. No, the complete ETR team has gone and the development of ETR has finally stopped.
And I can't see any chance to continue ETR, it was too long time that the development stagnated. And to say the truth, the code of Tuxracer is quite out-dated. A lot of things could be done much better by using actual possibilities: graphic, objects, collision etc.
In another post cousteau reported that he has detected my project website about Bunny Hill. That's right, this site is still alive and after disappearing from ETR I did some work on the Tuxracer code and a level editor. But then I got some expressions of other modern games (not one-person-shooters, I hate them). That moment it was clear that I first have to learn a lot about game development by using the available engines and libraries. So I stopped my work on Bunny Hill and looked around, and up to now I was busy with testing such engines:
- Blender game. Wonderful, very nice scenes are possible. The character can be controlled and animated in a comfortable way. It's not necessary to write any code, all is done by a system of logic bricks. But Blender game is too slow for Tuxracer. Really too slow.
- UDK and Unity. Only for Windows, hence unsusable. BTW: These engines are optimized for indoor scenes and one person shooters, that doesn't fit to Tuxracer.
- Irrlicht. Nice graphic engine that allows a very good appearance of the winterly environment. There are some imports for simple character animation. Irrlicht does a lot, including an event implementations that makes SDL dispensable. Irrlicht is good, but good enough? It lacks a real LOD for large terrains. An advantage: it is quite easy to implement own C++ code, e.g. for character controlling.
- Ogre. That's a graphic engine, like Irrlicht, but with more options. Ogre is not without problems since this has a huge number of dependencies. It requires an up-to-date linux system and the installation under Windows is not easy. Nevertheless it seems to be a good solution.
- Crystal Space. The best engine in my opinion, just with all needed options. Unfortunately it is extremly difficult to install. Till now I wasn't successful und I gave up.
- Bullet or ODE. It's nonsense not to use a physics engine like Bullet or ODE though the physical simulation in classic Tuxracer is really good. The aim is to implement a collision detection, and therefore we need a physics engine. Bullet seems to be better than ODE.
You see, I don't force the development of my Bunny Hill project. At the moment I can't say how to go on. There are some reasons to start from the scratch by using Ogre, Bullet and OIS. The character modelling and object modelling will be done with Blender, and I hope do find some suitable export scripts to integrate the blender meshes into Ogre and the own code.
This should be clear: The new Bunny Hill will be another program, and I'm not sure that I'll be able to finish that work. That's not a task for only one person. It would be nice to keep Tuxracer alive, but the most urgent concern is to found a new team. That's something I'm not able to.
So, I've told enough. Nice to read some posts on this forum again.
erintwo alias erin