Irrlicht Engine

I would put money on it being a C-style one for two reasons.

First (& foremost) there is still ALOT of code written in straight C. It is relatively simple to wrap C-code into an object oriented form & to integrate C-code into other languages. The same cannot be said in reverse for most other languages.

Secondly, OpenGL is primarily an API over a device driver. Device drivers are in general written in Assembly &/or C. With C, one can control the bloat to an almost exact degree. Again, most other languages cannot say the same.

I do wonder though why everyone seems to differentiate between coding in an "object oriented" manner and coding in C. Coding object oriented applications/libraries is just as possible in C as it is in C++, Java, etc. The only difference I find is that C doesn't force you too (niether does C++, but if you're not using it's OO or template facilities - you are pretty much writing straight C anyhow).

--EK

Hmm, now that the first API descriptions are out for OGL 3.0 I'm pretty sure that Irrlicht won't support it for the next month (who says years?). I doubt that there will be too much performance loss when staying with the old procedural interface which laready offers the geometry shaders etc.
Anyway, now that we know how it will look like in the future we can make our plans for 2010...

What are the plans for 2010? Also, do you need one of the newer graphics card to take full advantage of OpenGL 3.0?

Yes, there are some requirements on the card. At least for the next step in OpenGL after the already planned 3.1. But due to the drop of many long-standing techniques (suc as the fixed function pipeline) older cards won't do anything with OpenGL 3 anymore.
2010 was just a rough estimation when we'd be at a point for full support of OpenGL 3.0. No official date at this moment

Oh OK, so the Irrlicht doe's plan to take advantage of the graphic features.