I suppose there’s an element of preference as well. If !myinterest@instance exists and is limping along with 80 subscribers and a post once a month, is that less discouraging? Maybe 300 subs and a post every other day is adequate? At the risk of scope creep, maybe the answer lies in more data and options to account for the preferences of those new to the Fediverse. I concede I don’t have answers though, and I’m obviously putting less work into it than you are.

Fight the good fight, friend. I need more posts about old TV shows and niche hobbies, so we just need more decent people, however they arrive. :-)

IMHO, the APIpocalypse resulted in too many communities that died on the vine and discouraged their creators and few visitors. Funneling that energy into fewer, more general communities to build up views and conversations strikes me as a a necessary forerunner to a massive “Cambrian Explosion” type of thing. Subreddits, for the most part, naturally evolved because there was already a critical mass of users interested in the topic, not because the sub existed first and attracted the users. What would you think about a different approach to collect various subreddits and file them under healthier lemmy communities that are not one-for-one, but still relevant?

Sub : Community

• askreddit : asklemmy
• amitheasshole : asklemmy
• explainlikeimfive : asklemmy
• gaming : gaming
• pcmasterrace : gaming
• minecraft : gaming
• etc, etc.

Yup. At this point, “locally installed, reliable, parametric modeling on Linux” = “FreeCAD, including Ondsel, and SolveSpace”. That’s it. Well, there’s code-to-CAD as well, which obviously retains parametric history, but goes about it very differently than a design tree.

For non-parametric modeling, BricsCAD and Plasticity enter the discussion. For parametric on the web, OnShape works very well but I hate their licensing scheme and the huge doughnut hole in their pricing model.

I think the guide I did at !cad@lemmy.world is still in pretty decent shape.

I actually settled on Alibre Design. Permanent license at half the cost of a year of OnShape for a slightly dated but very capable parametric modeler, and the free trial made Parametric modeling click for me in a way FreeCAD didn’t. It comes with a renderer of a similar class, though I haven’t tried that yet. Changing colors of parts has been enough for my needs.

FreeCAD has apparently fixed the topological naming issue, one of the big things that was keeping them so far behind the commercial suites. It’s already in the weekly builds, along with several other enhancements pioneered in the Realthunder fork, including UI enhancements and a default Assembly workbench. Version 1 is going to come out in the late summer or early fall, I think. Ondsel is FreeCAD but they have some venture funding to pay developers to work on the main project and to bolt-on an optional paid PDM system (download from their GitHub and you don’t have to sign up for anything). I had some crashing issues on both Windows and Linux when trying to import DXF files into either flavor, and as you say, there’s still that learning curve, but I can get a part done in it now if I need to.

SolveSpace can do some nice things and will teach you good techniques.

I had the same issues as you with OnShape, particularly since their free licensing is very weird, and in the worst case it implies that while YOU must use your designs non-commercially, no one else is similarly bound. It’s sloppy legal drafting, and that annoys the little black kernel of lawyerness still sunk down in my heart. Fusion has become the poster child for free feature erosion and price hikes.

BricsCAD Shape is a basically an AutoCAD clone warped and twisted to act like SketchUp, and it works on Linux. It’s meant to be the tease to get people into their full-suite ecosystem, but I couldn’t find any legal limitations on the free version.

Depending on what it is you’re scanning, the people mentioning Blender may have a good point.

Shapr3D at $300/year might also be a good option.

Finally, for your particular workflow, Plasticity at $150 permanent license may hit the exact sweet spot. Definitely try their free trial. Some of the other programs I tried are also interesting.

You could also probably use Inkscape to get DXF files for both sides of the coin, making sure the size is right and that the path accommodates the width of your letters/strokes. Then, hypothetically, you should be able to import the files as drafts and then convert them to sketches with the press of a button (in reality, this crashes FreeCAD for me lately, but it could be a quirk of my setup). If you’ve already modeled the coin itself and use the new sketches to pocket into the existing solid, IIRC it should work okay. FreeCAD would not extrude a non-contiguous sketch into multiple solids, but I think it’s fine as long as the end result is still a single body.

Excellent, reader cases always seem to cost way more than they should, espeically with how fragile they are.

Although, you should know you’re supposed to move over to another room to pretend your workspace is cleaner than workspaces ever are, LOL. Then, what are we rocking for the keyboard?

The nozzle is 2.4mm.

https://www.youtube.com/watch?v=r7kMwPnydkk

The nozzle itself is 2.4mm. https://www.youtube.com/watch?v=r7kMwPnydkk

First, a little plug for !cad@lemmy.world because more traffic is welcome. The pinned post there is a fairly comprehensive list of viable 3D mechanical CAD suites. I’m a rank amateur with actual designing, but if you want someone to drone on at length about their business models and licensing terms, I’m your guy. (short version: “removed you, removed you, removed you, you’re cool, and removed you.”)

Now then… I also came from TinkerCAD, and I actually think the grouping and alignment tools lifted from vector art programs are super intuitive, and they almost provide a sort of design history if you use them right, but there are so many things that can’t be done quickly in TinkerCAD, and Autodesk also nerfs it for reasons that are commercially sensible but not technically necessary.

Almost all parametric tools , and also most “grown up” (for lack of a better term) direct modeling tools can do the Boolean addition and subtraction that is at the heart of TinkerCAD’s “solids’n’holes” paradigm, often in a couple of different ways. For instance, to make your orange part there, I’d draw a 2D silhouette of the vertical view, then extrude (or “pad” or “pull”) to the height. Then I’d draw on that top surface, possibly with a reference plane set up first to avoid having the model too far up its own ass (i.e. the toponaming issue), making the shape that needs to be extracted. Then you can cut or extrude down into your solid. Most tools will know what you mean, but some might make you do use a distionct tool or manually do the boolean “difference”. You can then do the same with your hex grid, setting up a new sketch for that. Later, if (for instance) you wanted to have 12 holes or bigger holes, you’d just edit the one sketch. Your red part would be similar, but doing the back of it would involve extruding out from the new sketch. The power of sketch and extrude is, apart from the ease of implementing a parametric history, doing several things at one that would each have to be a manual hole in TinkerCAD.

Finally, there’s the simple matter of fillets and chamfers, which TinkerCAD doesn’t support as an independent function. Manually adding them gets tiresome real quick and is the “killer feature” that made me realize I needed to move on. Other tools like loft sealed the deal. TinkerCAD is capable of some really interesting parts, but not efficiently.

I’m guessing vase mode is your friend with a nozzle wider than your filament.

You mean the thing for hosting galleries because Lemmy can’t do that yet and Imgur won’t work on Mobile without an app? Yes.

As a social network? No, and even when I tried it wouldn’t let me switch my post’s visibility from unlisted to public, so I shrugged and used it as an image host as I’d originally intended.

Per the edit to the main post and in the interest of full disclosure, I had a filament break about 1/3 through, but that’s okay because the partial PLA print was nowhere near strong enough along the layer lines. There are plenty of decent 3D printed keyboards out there that just need some assembly and post-processing, so if this one doesn’t work within the constraints I’ve set, there’s not much point to it.

Learned some stuff for future designs, though, so we’re all good. 😊

Most of my projects lately have been hand-wired keyboards. I like messing around with completely custom layouts and seeing what my home shop can handle. They almost all use the 3D printer somehow, but this is only the second one with a case that’s completely 3D printed. I’ve done one that was just rewiring a board with a broken PCB, three on laser cut aluminum I sent out to have made, then a few more with 3D printed plates or Masonite that my 5W diode could cut without too much trouble. There are plenty of 3D printed orthos and splits out there, as well as several nice 60% and 65% board that you print in multiple pieces, but my (admittedly somewhat arbitrary) goal for this project is to see how close to a “traditional” board you can manage as print-in-place on a garden variety Ender 3 bed.

Possibly, and now that you mention it I might round off the two at the front, but the sides are chamfered, and the main angle towards the front is quite shallow. This is also a very shallow “case”, like your average gamer mech board circa 2016. Reasonably, you shouldn’t really be touching it at all.

Absolutely! It sounds really interesting, and even if I don’t use OpenSCAD type stuff myself, many do and it would be super cool to see how the development progresses.

I’m no professional, but it does seem that many of the ways to design around TNI, e.g. defining reference planes for sketches that are associated with geometry via shared variables rather than a face earlier in the tree, are ripe for automation and/or are very sensible to obfuscate under a layer of abstraction. TNI doesn’t strike me as inherent to the way designing a solid object needs to be, but rather that it’s difficult or impossible to avoid based on the code that actually makes a 3D solids kernel work. To my mind, it seems like awareness of its persistence is sufficient if you have a mature set of workarounds and heuristics that make the software act predictably. I suppose it does promote lock-in and precludes a truly complete and portable file format though; the solution is clearly for one of the big 3-4 industrial suites to open-source itself!

As for that github table, I’m all ears about expanding it or correcting anything that might be wrong with it. The “real” one on wikipedia is rather daunting and necessarily for a broader audience.

I think some people do take them off if they have CRTouch and decent spacers, but the wheels still work as nuts, and then, neither CRTouch nor an Ender style bed is perfect, so if something is real wonky it can be good to make a few adjustments before probing. The Bambu was likely designed with ABL in mind, and at the price they’re charging I’d also expect a little more fit and finish, LOL.

The Bambu will have some sort of auto bed leveling. The simplest method is a limit switch connected to a small plunger style probe mounted next to the nozzle.

Whatever the specific method, the idea is something that lets the printer sample multiple points on the bed and use the Z axis dynamically to adjust for the small irregularities. If you don’t have that, even half a millimeter can ruin print quality, so the leveling screws are there to handle it manually from the bed side.

it doesn’t take very many orders to make buying your own printer cost-effective.

That’s what I saw years ago when I looked for US options. So instead, I just have two very cheap printers. My hobbies don’t make any money whether I’m efficient or not, so I’m more willing to trade time for cash investment.

Ahh, gotcha. I will not opine further then. :-)