Yeah, I did ak3 month first, then 6 month, then 12 month. If you do a family plan, I think you can also get the cheaper price with a shorter lock-in.

I get cheaper on Mint because I get the 6 or 12 month price, but it means you have to have the money up front to pay for it.

The further you go, the more specialized it gets. There are people I know doing their PhDs in CS, but it was pretty much just straight math. I’m now an expert in a very specific area of robotics. But it’s only worth it if you have a specific reason to go to grad school, like for a particular career path. If it’s just because you like learning, it’s not worth it. There’s a big opportunity cost.

My older sibling did something similar - getting Ubuntu installed on my very first laptop (a 9" netbook) back in 2008 and replacing windows XP. But be warned: it is a slippery slope. At the time , I just wanted a computer that I could take class notes on (high school), and never wanted to touch programming or the terminal. Now I have a PhD in computer science. I still don’t use Arch though.

Oh boy, I keep a page just for this!.

I need to update it (for example, Arachne perimeters in PrusaSlicer now let you print extra thin perimeters), but it’s useful to have a reference for common tolerances/dimensions like screw holes.

But a couple of my little additional pet peeves:

• Don’t put fillets on the underside of prints (against the bed). The nearly-flat angle always droops and looks bad. Use a chamfer instead, or make a fillet that actually starts at 30° from horizontal.
• The weakest direction is between layers. Design your part such that you can print it in an orientation where the thin/weak parts aren’t printed where the layer lines can snap (eg, print it flat vs vertical)
• Just like the straight lines inside screw head holes, thinking ahead in your design can prevent/minimize the need for support material. The earlier you start thinking about this in you design, the easier it will be. For example, can a part be designed with a 30° slope on an underside instead of being flat? Can you think about your print orientation early in the design process to avoid overhangs?
• Chamfer of fillet inside corners, if it’s a structural part. This will greatly reduce stress concentrations.

Personally, I don’t use 3 perimeters on most of my prints. On my prusa, they look totally fine with 2 perimeters. I only switch to 3 if I need the strength (which also almost always means I’m printing in PETG, rather than PLA, FWIW).

I’m also team onshape. I have a powerful desktop, but I still end up doing CAD from the couch on my 6-year-old Chromebook, so onshape is a champ for that. It’s also nice for collaborating, which I do when working on bigger projects with my fiancee.

I got started with it entirely from the tutorials provided by Onshape itself. The learning curve was a lot less steep than I expected.

What kind of curling issue, and did the Hilbert curve help with that?

I’ve found that I’m rarely doing something where my top layer lines matter. Usually, I get my finished surface by printing things face down on a textured sheet. This works for probably 95% of the things I print whereki care about finish. The others I’ll turn on ironing, but that’s probably way slower than the Hilbert curve.