GabiJack.com

While at the partner’s pavilion during SolidWorks World 2009, I watched a very nice demo of a 3D modeling product known as SolidThinking. I must say, with my very limited knowledge of surface modeling in SolidWorks,  I was very impressed by how easy it was to create and modify all sorts of capricious shapes starting from very simple geometry and/or dragging around, pushing and pulling the already existing elements of the model. I remember thinking of it as virtual clay being shaped by an artisan. In a matter of minutes, a stylish armchair was created before my eyes, and even rendered in a nice dark brown leather texture, and positioned in the middle of a room where an invisible wall was made part of the surroundings, simply to catch the shadow of the chair as if it were really projecting on the image in the background, all to make the whole scene appear more realistic. A pair of sunglasses was also created with the greatest of ease, and modified into three different fashion styles, again making use of very simple geometry.  Amazing!

So, a bit after coming back home from  SolidWorks World, I went and downloaded a free 30 days trial version of SolidThinking from their website in order to give it a try.  I must say it is kind of easy to use, although not as easy as it looked during the demo, which is reasonable to expect from someone who has never before tried it and is used to work with SolidWorks, I guess, given that the whole interface (graphics area, tools, etc.) looks dramatically different from one application to the other, and, in general, there are also quite a few differences in how the 3D model is created.

A few quick examples of what is mentioned above:

In general, and although it also offers tools for solid modeling like prisms, spheres and such, the models you generate in SolidThinking  will be made up of  a group of NURBS surfaces. This is important to keep in mind. A model created this way may appear like a solid, but if you save it as an iges file and take it to SolidWorks, what you’ll see in your feature manager is a group of imported surfaces that make up the model.

As I said before, I really don’t have much knowledge on surfacing and didn’t have a lot of time to test it thoroughly, but I wonder if it’s precisely because it works with these kind of surfaces that it’s so flexible and makes it easy to shape the model practically any way you need or want to, simply by editing the surface and dragging control points, like you would for a spline.

Another big difference that takes some time to adjust to is the interface. By default, you are watching your model in four different views (Top, Front, Right and a perspective view); although you can change the layout at any time to include different views of the model.  Moving the model around is also done differently. Instead of simply dragging the model with your mouse, you have to drag while pressing a button in the graphics area. There’s one for panning and one for rotation.  I’m not sure I like it; maybe I just need to get used to it.

Also by default, you’ll be assisted by a series of grids that are there to help you place points and objects into the scene. You can define the grids, each with its own origin and spacing. It takes you by surprise at first, if you are not used to working with a grid, but it’s actually kind of useful. Having to define an origin at the beginning of many commands (like whenever generating a NURBS curve ), as well as finishing the command by pressing the spacebar also took a bit getting used to.

There is no feature manager, like in SolidWorks, but there’s something called the construction tree that keeps track of the history of your model and all the steps involved in its creation, all the curves, surfaces or parameters that have been either created or modified will be stored there and can be accessed  and modified at any time.

To tell you the truth, I find the way many of the tools in SolidThinking work both cool and confusing all at the same time.  Cool because in order to do something very similar to what we would call a sweep in SolidWorks, you don’t even need to have the profile, path and/or guide curves in touch with each other. No, you can draw one or a couple of NURB curves somewhere in one of your views, say the right view, and your profile, say a circle or ellipse, somewhere in the same right view, not even near the NURBS curves, and not even in another plane perpendicular to the curves, then use an operation called pipe or bi-rail to sweep the profile along the curves and create a cool shape.  That is so easy, it’s cool, but what’s confusing is how is it that by doing just that you end up with a shape like the one in this image, where the surface highlighted in red of this fountain pen was created by “sweeping” that green circle along a couple of curves that were drawn on the right view.  The two NURB curves on the right view should provide information about the shape of the surface as seen from the right, but what I don’t get is where does the shape as seen from the top comes from.

Rendering in SolidThinking reminds me a bit of PhotoView 360, if only because it gives nice results and is fast and easy. However, the renderer in SolidThinking allows the use of lighting and cameras, as well as a few tricks like that of the invisible wall to catch the shadows of an object or the use of panels of light to simulate the light coming from a window, for instance.

I wish I had had more time to explore the software, but all in all from what I’ve seen I think it’s a nice tool for conceptual design. I’m not so sure about how useful it may be for engineering applications, but it seems like a nice tool for the product designers to help bring their ideas out of their head and into life faster. Like I said before, it makes me think of virtual clay…

3 Responses to “SolidThinking: an interesting approach to 3D modeling”