So you played only the demo and now you want to judge the technology behind it?

First of all, the price is very cheap for a technology like this with full source code.

Regarding the tool set: The tools are great, import is not as easy as in Unity but still much faster than any pro engine that needs Max or Maya plugins and special conversion routes for textures, materials and whatever. Even UDK is more complicated in terms of import.
C4 reads Collada files very good, with or without animation. Textures will be converted to a C4 format that can use GPU compression similar like DDS.

A big advantage of C4 is it's great zones and portal system. It allows to create really big interiors and renders only what you see. All features are rock solid and well thought.

The voxel terrain is impressive, you can sculpt caves, bridges, whatever.

I am not the most advanced C++ programmer, but I can read the source of the C4 samples. The demo is understandable but not the best example for a beginner because of all the features.
And yes, the other samples are also not very short. The reason is, that there are so called controllers that have to be written and have to be registered to be available in the editor. So it needs some code to set everything up. But in the end, when you are though it, it will help a lot to write more code based on it and it helps to have these features available in the editor.

Many if not all of the coders I met talk good about the structure and organization of the C4 source code and of the API design (especially people coming from Torque or Ogre).

I personally prefer to code in a language without pointers and references and type-casting and all these barriers. I can indeed achieve more with less code in other tools. But this is my own preference. I wrote a little physics game in Unity in an evening and a memory 2d game only with the GUI in another. So I can create results faster with their more simpliefied API. But on the other hand my performance stress testing indicates that Unity has not the fastest renderer currently.

So in the end it is the same story as always: Use the tool that fits your needs best! But dont judge it only from a demo! Do some stress tests, some import tests, some playing around with the GUI, with the functionality to move, translate and rotate objects, whatever. In the end, the programming language should not really matter, when the technology fits perfectly to the desired game. We can learn a new programming language faster than the company behind an engine will implement the features that we miss.