3run what is the exact reason you are looking for a new engine? You have started quite a lot of projects and I enjoyed a lot of those prototypes from a gameplay perspective. Why don't you try and find a game idea/ gameplay prototype which you want to transform into a real game? I think this would give you enough of a motivation boost again.
Just plan the game so you have a fast way of content creation, choose a solid graphic style (which can and will influence your gameplay, too) and you are good to go, and why not try to make it a commercial project to be sold on Steam or some other platform.

Yes 3run i totaly agre! I've been here over than 10 years, and i do not remember that the forum was so much inactive. Take me for example, for over 10 years with two accounts (first i lost), i make only 650 post. Or better example, guys who made Supercan, i think they are not registred, but still work on next project with Acknex engine

I agre with Superku, i think you have enought expirience to make great commercial game, and with Acknex, i am sure you can do it.

I don't want to change the engines yet.. I just can't work with Acknex any more for now. The lack of support from Conitec side.. I've reported few PH_CHAR bugs, yet they aren't even fixed.. Acknex is using physX 2, which is pretty limited in comparison to newer physX 3 (and I don't like the way it's implemented, which makes it pretty useless to use without any pain in the ass).. And I hate facing this strange limitation/bugs that I don't understand.. F.e. I had this very strange bug/limitation lately (I'm not sure, maybe OBB suppose to work this way?), I was creating a simple game prototype for one of the forum members.. The idea was, hitting the moving box, with a ball. I wanted to keep it simple (as examples from AUM etc, so that man I was creating this stuff for, could learn from it as well), so I used 'c_move' for boxes and balls as well. They were moving towards each other, and I used events to delete the box, when ball hit it (from the box's event).. I used different events (EVENT_SHOOT with ACTIVATE_SHOOT in the ball's c_move, EVENT_IMPACT, EVENT_ENTITY for ball.. used even EVENT_PUSH with IGNORE_PUSH for ball), but still the ball was passing throw the moving towards it box... I couldn't understand why this happens.. (and yet I don't know why). I've read a log time ago, in older AUMs that if object moves too fast, it will pass throw other models and blocks, but the thing is, that box and ball were moving pretty slow, I couldn't say that they were moving this fast, so they pass throw each other.. Another thing is the OBB ellipsoid hull (I don't like native physX, it's pretty useless to create something complex), but yes there are lot of workarounds (for physX as well), and I'm using them, but yet I'm not satisfied by results, and I'm really pissed already off by the simple thing, that I have to use this workarounds in order to get something stable.. :<

Thank you guys for link and support as well, maybe I'll try to go on with Acknex to create a game, but I'm not sure yet. I'm lack of ideas (not for the game, but how creating some snippets for game as well), so I'm not sure, if I'm experienced yet, to create anything stable with Acknex yet (maybe not even with other engiens, I don't know)..

Greets

I can understand your frustration and think that the native collision detection can use some work because of ellipsoid hulls, however it is rocksolid even at extreme unit speeds. You must have done some mistake (which is only normal) and/ or overlooked something.
Only with physX you have to take special care when dealing with fast moving objects.

Maybe, and I guess, more likely I did have something missed or so.. But I'm pretty fed to the teeth with Acknex for some reasons, maybe I just need to take a break..

This is good advice. 3run, you should try to create a project that plays to your strengths. Try to figure out what you know you can do really well, and use that to make a game that's interesting and has a good presentation. If you do that well, your game will sell itself on its own merits.

imo it is fun to discover other engines, but you can get frustrated by any of them. I tried those I mentioned above, and currently only Esenthel 1 and 2 are on my pc to play with them sometimes (and UDK because I was lazy to uninstall it). you can get inspiration from them, and you can decide whether acknex is the best for you or not. you can also find that your programming or artistic knowledge is not good enough for some of them. it is important to have proirities of your own requirements, what you expect from a game engine, depending on what kind of games you want to make. but it needs a lot of time to get deep into any engine except if you make a game that the engine offers by sample projects, like Esenthel's shooter, multiplayer and rpg projects or Unity demos. maybe you could take a sheet of paper and write down the things you want to do, in the way you would like to create it, what features you need, and assign a priority to each of them. or wait for Rayne

Well, the main thing that I wanted to see, is support for old AABB or at least CAPSULE hull instead of ELLIPSOID for native collusion detection..
I want to see support for physX, cause it's pretty useless in it's current state, if you want to create some complex movement with it.

Greets

I have to say Rayne promises what a game engine should be, so i would start learning c++ now and play with a c++ API like irrlicht or a free game engine using c++ to get up to speed in c++.
Rayne looks like it is going to blow everyone away, but will require c++ knowledge.
After hating c++ for some years and trying C# i reverted back to c++. It's boss.

The Skyline Engine also seems to come along nicely:
http://www.chi-ad.com/Skyline/SDN/skyline-news

Might be worth a look...