Carrara is simply a one stop shop for me which is why I've continued to use it (plus, I don't want to go through learning a new program).

I have used Amapi for certain modeling needs (almost none).  The Amapi interface has a few aspects to it which are overwhelmingly maddening for me.  Where the XY coordinates of the cursor are is one of my most hated things for Amapi.  I have yet to be able to figure out how the heck that works.

When C became capable of importing Poser/DAZ figures, the user base made an interesting shift.  From my perspective, the original C user community were a small niche that liked to model and render and try all kinds of weird rendering stuff.  Most of the discussions in the forum were technical (not only for newbies which is always great but also from heavy hitting users that were trying to do really whacked out stuff) and were answering questions about how to get some kind of new type of result.  A perfect example is the tutorial I recently posted showing how to do the planet halo.  I cooked that up to solve a user problem a long time ago.  The forum was chock o block full of VERY cool stuff like that.

The forum was a lot of fun for me then.  Now that the user community contains a lot more people who are much more interested in importing already made models and textures to create scenes, we don't see as much of the bigger trickier weirder stuff.

The gallery clearly shows this aspect.  We hardly ever had people (human figures/models) in any of the renders.  Now we are dominated by them.  The reason is that people like to render people and when you make it easy to do that, it will happen.

My job has taken away a lot of my time from C.  Other forces have reduced my attention to my art as well (like the fact that they redid Star Trek with all new Enterprise scenes and they didn't screw it up and it's totally amazing so I don't feel the need to generate my own stuff as much).

From my point of view, the next BIG place for C to go is by making a whole new set of animation toolsets which could push the package into the Lightwave arena.  If you really wanted to make the code worthwhile for an upgrade (which 6 to 7 was more of a bug fix deally and functionality wasn't much of an upgrade interest for me), then we need to get to where C could do big animations really easy and the user community would be making things like commercials and little movies that we'd be seeing on Youtube a TON more.  I know animations are out there and that it's not too crazy to get them made but C could be a lot better for animation.  I think that's the next step for it.

One last dig, why didn't anybody every take my recommendations to get the metaball modeler fixed up?  Being able to enter the XYZ coords for an individual metaball with manipulators we have in the assembly room would be a HUGE addition.

I'm going back to work on the Millenium Falcon.

Speaking of strange and weird technical problems to solve, here's one.  C wouldn't take any shader I applied to an object which I have used 3D boolean on.  In order to overcome this weird problem, I had to apply a new shader domain to the object (a new complete name) so C would allow me to apply a shader other than Default.  I could even take an object into the browser and back into a brand new file and the problem would persist (dragging the original default shader along).  Weird.  An easy solution once you know it and something I'm sure the forum community could have solved if I hadn't been able to bash it out myself.

-Kix