Fix the link?

Well then: splines.  3rd order Hermite is a nice easy starting point.

Its a simple 'using delta time equals jitter problem'
I kinda dont like delta time and try to avoid it because of this
Our latest game does a low pass filter over a coupke of collected delta times and so on

My advice: either make sure you have a rock solud framerate,  which is good anyway; or read up on delta timing / time steps in video games

Learning stuff in java2d is a useless tool. Sure maybe you get better at coding but that comes with any form of coding.

And thanks for the correction but I think I got the point across. 

Relminator, OK, is there example code or pseudocode out there for using a K-D Tree to store regions?  It seems like a pretty large divergence from the way K-D Trees are presented around the web.  Handling dynamic entities that are moving in and out of tree nodes frequently and overlapping multiple nodes is a problem that I don't immediately see the solution to.

Riven, if my question annoys you, you can go somewhere else.  Trees are used for collision detection in professional games, this is not a purely academic question, and optimization of such trees is a problem that has been well studied, I just can't find the publications that discuss that optimization.  Asking how best to optimize a spatial tree is a perfectly on-topic question for this forum.  You don't know the answer and rather than just not comment, you insist that I shouldn't want to know the answer myself.

various flavors of k-d trees can be nice.  But you really can't make sweeping statements like this about any data-structure without a ton of details.  Why a kd-tree instead of say a vantage-point tree?  Or a non-region quad-tree?

So I'll reiterate my question: How best to
optimize a quadtree for large numbers of
dynamic rectangular entities arranged in
an uneven distribution across a vast
space, beyond the simple textbook
implementation presented around the
web?

By fun you mean like taking nutcrackers to ones testies. *whisper* Masochist *whisper*

Thank you :-)
My goal is to make a simple, very basic tetris-like game, but it should have a solid, efficient fundament which can be used to build more complex 2D games. I used OpenGL many years ago and have never used it since then, so I think it would be quite hard to implement. Most people don't recommend to use OpenGL for 2D, some do, but I don't have enough experience to evaluate the pros and cons. I also haven't found any tutorials on 2D OpenGL games in Java.
Doesn't Java2D use OpenGL under the hood? OS window managers are also 2D, but sometimes based on OpenGL.

Isn't that the whole issue at hand? In 2D you're expected to have every sprite nicely translucent at the edges, whereas in 3D you can get away with your typical antialiasing, defining shapes not by transparant pixels in sprites but by geometry.

That's the single reason good 2D games are prone to be suffering from overdraw, caused by mandatory back-to-front rendering.

I sample 5 textures, transform normals from tangent space to object space and write to two RGBA16F textures per pixel. 3D games usually require much more complex shaders, much more bandwidth (in the case of deferred shading), more anti-aliasing and higher screen resolutions. Even with early-Z optimizations, we still get massive overdraw for lighting and post-processing.