Simulate Early, Simulate Often... In Rhino
I read through all the previous wish list posts going back to 2010. It seems like a lot of progress has been made regarding assemblies and boundary conditions since Scan and Solve's inception.
There's quite a few wishes for a variety of features, all of them are probably very important in specific contexts. If I could cast my own 2cents out there...
I think people that are drawn to Scan and Solve probably could be classified by one or more of the following three categories:
a) they're a Rhino3d CAD user and consequently appreciate the simplicity of an integrated FEA plugin.
b) they're an FEA user and appreciate the ease of use of Scan and Solve and its affordable price.
c) they're drawn to Scan and Solve because of the inherent capabilities of its unique meshfree approach.
I think maintaining and reinforcing these three areas of strength will probably help this software stay in the spotlight and set itself apart from the large variety of other multiphysics simulation platforms out there.
Regarding item a):
I think it's awesome that Scan and Solve is available as a plugin for Rhino3d. But immediately my mind jumps to Rhino's associated Grasshopper platform and I can't help but think the power users in this community would make tremendous use of SnS related grasshopper components.
Regarding item b):
I have nothing further to recommend here.
Regarding item c):
The meshfree uniform grid based underlying technology enables a variety of capabilities that are not possible or practical with current mainstream FEA platforms. Since this is such a key differentiating characteristic of SnS, I think it makes sense to allow more user control of the grid structure than the simple resolution slider that is currently present. I see that Matteo and Vadim discussed this back in August 2010 and I would look forward to any progress that might be on the horizon regarding additional sophisticated control over the grid mesh. However, in the mean time, would it be possible to unlock the x, y, and z direction mesh densities? I would imagine a lot of simulation problems could be characterized as being more 1.5D than 3D in nature. Or likewise, more 2.5D than full 3D. In these cases I would assume the user could benefit by applying more grid cells in one direction than another. I think this might unlock a substantially larger number of real world application examples for detailed FEA simulation, since the computer memory and processing time might be reduced by orders of magnitude.