Comment on PSA: Try FreeCAD Link Branch (it's a big improvement!)
u_tamtam@programming.dev 11 months agoI won’t say that FreeCAD has a good UX, but it helps a lot NOT to look at it as a CAD software, but as a collection of specialized engineering tools, organized into workspaces, haphazardly put together.
First thing you need to know is which workspace you’ll need, and FreeCAD does a terrible job at explaining you that (the concept of workspaces isn’t self explanatory) AND describing what each and every one of them does. Some of which should just be disabled by default because of how fringe, unpolished or unreliable they are.
Once you’ve got that part cleared, you can learn the primitives and the jargon (what’s a body, solid, part, mesh, element, …), not great, but fair. Then, you have to learn, for every workbench, what their workflow is (e.g. create a body, create a sketch, apply transformations ; Create an analysis, define material, define loads, add a mesh, add solver, add equations, run solver, add results, tweak the pipeline so it renders, show results), and yep, FreeCAD won’t hold your hand for any of that, you’ll have to wear your explorer hat and navigate from frustration to incomprehension until it accidentally works.
But then, if you can get over that, you’ll end-up with a tool that’s more powerful and versatile than anything else, including dandy commercial offerings. It still blows my mind that nowadays anyone in their garage can do for free what not so long ago would require a full engineering curriculum and corporate sponsorship to acquire licenses. My hope is that FreeCAD would gain the same kind of visibility that Blender enjoys, with sufficient funds for a small dev team and a great product manager.
thantik@lemmy.world 11 months ago
FreeCAD could do with a massive UX overhaul and a workflow that people could use without having to relearn how to invent the wheel. The underpinnings of the software are good. They’ve done a lot of good work there. But as far as a usable piece of software, something like Blender – with a lot more complexity, and a lot more individual toolsets (2D Animation, Video Editing, VFX, Sculpting, Poly Modeling, Bone Rigging, Scripting – ALL with entire workflows associated with them) – has managed to be a wildly usable solution despite many of its individual subsystems working together.
FreeCAD suffers from the lack of proper workflow immensely, and will hamper its adoption unless the maintainers take a far more pragmatic approach to their development. They need to take this more serious, as more usage drives more funding, which drives better development. You can’t just go “It’s open source, YOLO, fuck you if you don’t like how we do things”. Because you’re just going to flounder as a long-lived project, but never a successful one.
I complain about FreeCAD because I want it to get better. I go back to it every year at least once, just to see if they’ve opened their eyes or not. I have used everything from Siemens Solid Edge, to Inventor, to Solidworks, Fusion360, OnShape, Blender, Rhino, hell – EVEN SOLVESPACE is easier to use than FreeCAD and it’s only a solver…
u_tamtam@programming.dev 11 months ago
Yup, I believe this boils down to good project management, someone has to steer those individual components so they work together better, in a cohesive manner, to make the result more than the sum its the parts. This is especially difficult in an opensource context where different contributors have different interests, and I think Blender, having managed that much, is an example to follow.
And I agree with everything you wrote, I don’t expect FreeCAD to get there in a reasonable timeframe unless it gets serious funding and expertise brought in.
Plasticity is a good project to follow, I don’t think they are comparable nor intended for the same audience, but there are product design aspects to learn from it, definitely :)