Well, to be fair it had some good features, but also some bad ones.
The good is there are a lot of helpful people on the forums, and there is any feature you could possibly want as far as doing stuff goes (not so much graphics features). The bad is that the documentation is just nonexistent almost, if you get the shader version. It's also really complicated to do simple things and even going through the torquescript is a bit of a nightmare. I can't even imagine going through the source.
If the looks and performance were better I could forgive it, though, but while it is 'pretty good' the torque advanced engine is a dead end that will never get serious feature upgrades. It's already not the greatest and it will get worse in time. If I ever wanted to do a follow up then I would have to throw out all my code - torque 2 will be completely incompatible.
Overall I think you are better off with torque if you are not a pretty strong coder, but I feel like C4 is the best bet out there so far. Unfortunately it has some goofiness I have discovered. It does a lot of stuff in CPU like the sound for instance. That can make things a bit jerky, and I am not sure it's multithreaded (which would be a ton of help now that duocores are pretty standard). It's also not really geared towards GUI style games it seems, and had kinda poor mouse support as yet, of all things. But, I think it will work out OK. It's all in C++ but even so I have found it a lot easier to work with than torque.
I guess there is just no good all around solution that covers all angles. Torque 2 might turn out well. Sounds like they have a good plan for it, anyhow. However they are diverging into 3 separate dev paths. I don't think that is something that inspires much confidence in me because even their old 1.x and advanced models have serious outstanding bugs that have not been addrssed for years on end.
