I am a Mechanical Engineer and avid CAD user. I've worked in a variety of industries including heavy industry and consumer product design. I've used SolidWorks, CREO, Inventor, SolidEdge, and Fusion360 in production environments. I dislike them all, below are some of the reasons:
- Exorbitant Cost. Seriously. I can't even believe that a tool engineers have been using for 40 years can still cost upwards of $5k PER LICENSE for mid-tier capabilities. That doesn't even include the very expensive file management tools needed to support the package in a production environment. It is outrageous to me that in the software engineering fields, the most cutting edge technology, amazing software, can be had for almost nothing. Sure there are expensive software development tools in existence, but their need is less pronounced with the advent of open source tools and code. In many cases, they are really a "nice to have" rather than a brute requirement. Even as you get into very specialized fields such as machine learning and robotics, the capabilities that free open source software provides is astounding.
- Locked Down File Format. I'm tired of the file format game. Each CAD vendor has their own special proprietary file format. Sometimes they are compatible/importable on a model level, but especially once you get to the drawing level there is no intercompatability or opening of other CAD systems drawings. If you want to open a file from another vendor, you either purchase their software (which is somehow always different from yours) or you import via STEP, which trashes most of the 'smartness' of the other CAD system anyway.
- Computer requirements. It does make sense that your computing needs grow proportionally to the size of your model, but there is no reason you need to have an NVIDIA QUADRO XXX that costs 10x more and is 10x slower than a "consumer card". Something is wrong here. Further, these requirements are often somewhat arbitrary and don't always emphasize the correct specifications for what you actually need to accomplish.
- PDM hassle. It really bothers me that software engineers have amazing, fully functioning, free tools via git, meanwhile, we are stuck with outdated, difficult, expensive tools to do any sort of data management. I'm not even looking for advanced functionality like merging features, but how hard is it to implement a simple check-in/check out system without spending $10k on a full fledged database system?
- Restrictions in how you model. You have to do exactly what the tool wants you to do, whether or not it is actually an efficient way to produce geometry. I have found CREO is the worst offender in this regard.
- Operating system restrictions. I want to be able to work on more than just Windows. MacOS, Linux, ios, android. We live in a multiOS world, why is every CAD system (except Fusion360) not compatible with more than Windows? Why is every CAD system built on a layer of code that shows it's age and inflexibility at every opportunity (still looking at you CREO)
I've been thinking I want to make some open source CAD modules related to robotics to build on some of the existing amazing software out there, but I know that I don't have the desire to support a whole fleet of CAD systems and I don't want to limit their use to such a small slice of the world, especially one with such high walls as current commercial CAD systems do. I also appreciate the deep access that FreeCAD gives developers to the kernel. Many other CAD systems trap you behind a walled garden API that limits dearly what you can do and dramatically slows down operations with the overhead of the CAD system.
My initial experience with FreeCAD on MacOS was a little crashy, but I'm not attributing that to any fault of the underlying technology systems. Do you think that with the proper development support, FreeCAD and the OpenCASCADE kernel have enough oomph to compete with SolidWorks as the current industry leader of CAD systems? The Linux to Microsoft's Windows? I look at other OSS projects like Blender, which is now at a point to be a full Maya replacement and gain confidence that we as creatives want tools that allow anyone to use them but still maintain exposure to all of the deep tools for expert users/developers. If a tool like Blender, which primarily targets the Creative side of the spectrum can become so dominant, I wonder why FreeCAD couldn't do the same in an arguably more technical domain.
I linked up to the FreeCAD source code, but was struggling to get everything I needed to build on MacOS. It appears the directions are outdated since Mavericks. Has anyone built on MacOS recently?
Is there a FreeCAD development road map?