29 January 2025 in Brussels. Probably the next good candidate for a 1.0 release event as well. Latest blog has them down to four release blockers, though that number can certainly drift back up as well.
KiCAD; fucking amazing program. In just 4 short years it went from your standard “It’s open source so I guess I’ll use it”, to “holy crap, this isn’t just open source, it’s GOOD”
FreeCAD - in development for 20 years, somehow still trash.
I am an open source zealot basically, and I can’t use FreeCAD. OnShape, Solidworks, Fusion 360, Plasticity…all fucking amazing. I’ve used them all. FreeCAD feels like taking a flame thrower to your face. And yes, I’ve used the realthunder branch; and yes, recently. I keep trying to force myself over to FreeCAD (remember…kind of a zealot) and it’s been the same story for 8 years at least.
Hell, even Blender with the CAD Sketcher plugin (only in development for like a year) is more tolerable.
I used KiCAD to design one incredibly simple PCB (literally just traces and vias), but while there’s huge room for learning and growth, it basically went smoothly once I understood the paradigm. I have watched enough videos to get a feel for FreeCAD, and it could meet my modest needs soon, but everything is still much harder than it needs to be, and there are far too many people to keep happy to let its most popular workbenches become what they need to be.
When I was trying to simply import a DXF to extrude, and it couldn’t handle that on either Linux or Windows, I finally gave up and bought a perpetual license to Alibre. Now I just hope they continue to exist in some form sufficient to maintain the Licensing activation server.
I am rooting for FreeCAD, and I’m following it with great interest. It is undeniably better than it used to be (the Ondsel soft-fork is kinda nice), but I’m just hoping it will be competitive for simple part and assembly design by the time my version of Alibre starts to feel long in the tooth. The heuristics, stability, and UI of the commercial suites are just much farther ahead right now.
With as much shit as I talk about it - I really want someone to fork it and fix its workflow issues. It’s all kind of there, it’s just scattered, and the workflow doesn’t make any damn sense.
They need to introduce someone to CAD, watch them crazy carefully, note where they struggle, and improve those areas. The paradigm of “Draw thing on 2D, extrude/cut into 3D” should be dead simple. When you can’t even get THAT right, nobody is going to want to use your project.
They need to work on Extrusion/Cut, Dimensioning, Patterning (Mirror, Linear, Circular), Fillet and Chamfering.
Those cover 99% of the shit I need to do. If it could just do those, and let the workflow be smooth - it would be a great program. I WANT it to be usable. I WANT to switch to it!
I know you said you tried Realthunder fairly recently, but a huge push for v1 has been to bring his toponaming mitigation in, in a scalable way, along with UI improvements (but only marginally improving the underlying 1990s Catia paradigm) and Ondsel and others have done some good work on the Assembly bench. Like I said, it’s better; it’s just not quite THERE yet. I’m going to give it another try when 1.0 is officially released, and I do love having it around. The fact that it works at all is a minor miracle, but Blender and KiCAD show what’s possible.
EDIT: Just saw that Ondsel is shutting down, though several of its employees were already FreeCAD contributors.
Rather than following things that don’t work well lol like SOLIDWORKS, freeCAD could use some feedback on how things function in UGNX. Where SOLIDWORKS is barely able to handle opening a full model of a mouse assembly, NX is capable of opening a fully featured airplane model.
Another example, the extrude feature in SOLIDWORKS is different than the cut tool. In NX things are simpler, you can either create a cut or an extrusion all in one feature. That’s the level of stupidity in SOLIDWORKS that freeCAD should look at NX to avoid.
A much more important thing is figuring out the topology problem. Eliminate that