what dort of architecture problems? How is db3 architected differently? how easy would it be to switch architectures?
Things are just very... flat. Non-compartmentalized. The graphics dig in to the physics and the DNA and bot memory and all that. It's like a complicated phone operator's switchboard. eg:
picture of a switchboard. Makes the code hard to maintain because a bug in area A might be caused by a bug in area W.
why are we currently working on db2 from time to time, if db2.5 is almost done?
DB2, for all it's faults, is a relatively stable code base. Whereas the C++ port was basically feature complete but still needed bug fixing. And as I've come to learn, the last 10% of a project take 90% of the time.
So given that I wanted to start over in C# and do things right from the ground up, it just made sense to abandon the C++ code.
oh also, now that I have the code, how do I execute it?
It's just C++ code. Nothing fancy. Stick it in either Visual Studio (I used VS6 for this, but it should work in later versions) or GCC and hit compile.