so but not sure that all are free :)
Contextfree http://www.contextfreeart.org/
Structure Synth http://structuresynth.sourceforge.net/
Blender http://www.blender.org/
Houdini http://www.sidefx.com/
http://www.paracloud.com/
http://www.bentley.com/en-US/Products/GenerativeComponents/
Groboto http://www.groboto.com/v3/

Hi andras, I'm not really that familiar with them, but I would not be surprised if those different systems may have strengths in different particular areas, so it's probably something like one kind of project with some particular qualities may be better suited for one and a different kind of project with different specific demands may be better in another.

It seems like Paracloud has more emphasis on numeric control and connecting to Excel.

Grasshopper is more graphical and probably easier to get into.

But if you just want to build a crazy form, you don't necessarily need to do it with one of these tools at all, you can just use surface modeling to construct it.

I think you need those tools more for when you're going to undertake a very involved construction of the crazy form and you know that there is going to be a whole lot of changes and back and forth involved over an extended period of time.

Using those tools to build forms I think actually takes a significant amount more time than just modeling it with say sweeps and networks and stuff like that, and it's not even guaranteed that you can just make a change and have your model update automatically in a sensible way, you really have to do a lot of planning for how it's going to update. So unless you are actually going to need to do quite a lot of updates you can actually end up losing time instead of gaining time.

- Michael

What Michael says is true. When I found grasshopper I was very exited because of the things people do with it, but its because it is good doing repetitive and progressive tasks and paneling; that makes a model look very complex. Like Michael said it needs a lot of planning and it is useful when you are going to try different things around one model, and then before you start you plan which things will be manipulable and base your definition on that.

I don't know the other programs people mentioned but If you think parametric modeling is for you I think GH is a good solution because it works in rhino and you can just copy paste things from MoI and vice versa. GH has a lot of great things inside for making complex models and for playing around in the design phase and to mention a couple more features it is easy to use compared to other software and it has a great community almost as good as this one.

Currently I'm working on a model that has a lot of parallel planes that morph from one form to another. That can be done with a sectioned loft but I'm using GH because: Im not sure what the final form will be nor i know what the final spacing between planes will be because its not constant. Using a traditional approach it can take a lot of time and I can get tired of making a new model each time i think of a change. Here GH become useful. Any way I always draw in MoI the things I need and then copy them to rhino, and in previous works, after making all I want in GH i bring the model to MoI for the final touches. In complex models MoI and GH make a good team.

I have experimented with both GC and Grasshopper. Check this link for comparison (http://designplaygrounds.com/deviants/rhino-grasshopper-vs-generative-components/). To my limited experience, Grasshopper is much easier to learn than GC .

Revit or Project Vassari from autodesk is also another tool that has some limited capability for generative design. check this website (http://buildz.blogspot.com/).

All of these applications require some knowledge of scripting if you want to do very complex forms with parametric functionality. I still prefer to use MOI for conceptual design. It is fast , easy use and enjoyable. In my opinion you can probably be able to do some fairly complex forms specially with available scripts and two new added feature (flow and twist commands)

-sharif

Hi! Thank you everybody opinion. I like your points. Actually I concentrate liveable things not that strange things from star wars :), So my aim to construct some structural elements determined upon logical geometry shapes such as curtain walls, frames etc. with many detail. For example If a screw fix changing needless to construct all of them. As well as there are many other things to exploit such as equation based graphics formula such as sun path diagrams, linear heat transfer figures and so on. That node based script engine is not just only crazy things, I believe. I just want to say it would be cool to see in moi. happy new year!