My problem with Leadwerks is the same as Ogre. It looks like set of libraries you build your own framework around, not a complete game engine.

If you go through the pdf tutorials section it seems to indicate that you're going to have to do a fair amount of work to build a real framework around the basics. Pretty rendering is all good and well, but I am not keen to do that much work before I can even start making a game. Not again :P
 

For me, it depends.

Is it more of a pain to build a framework from an existing high-level library, or to comb through a dense game engine trying to re-write it to meet your needs?

I've done it both ways now, and while I still lean in favor the latter, that only applies if your game concept and the engine are a reasonably close match. Some days with Torque I feel like I am doing nothing but fighting the engine. But then I stop, count to ten, and try to remember all the stuff that DOES just magically "work" for me in the engine, which helps.
 

Sure. But then again, when I wrote my own framework/engine, I STILL had days when I felt like I was doing nothing but fighting the engine. At least this way the engine writing part is already done.

But I definitely hear you, every now and again "grass is greener on the other side" syndrome hits and I go and spend a day or two "evaluating" other engines. Which means playing with their demos and looking at their screenshots instead of doing the boring grunt work on SoW.
 

Yep, I'm the same way. I've built my own from scratch before (more than once, but Void War was the commercial result). And vowed never to do THAT again. But then time dims the pain...

Spent a bit more time playing with them last night before getting on to the "real" work. Enough to satisfy myself that out of the candidates that I looked at, there was no "miracle cure" that was going to solve all of my problems overnight.