Astro Hacker News - Akse3D – open-source 3D modelling anyone can master

jonatron |next [-]

Clicking "Try Aske for free" takes you straight to it, no sign up! I usually avoid clicking these buttons because they usually take you to a sign up form.

thenthenthen |root |parent [-]

Yeah, i was also confused and made me scroll down searching for the ‘real’ try for free option lol

lucasoshiro |next |previous [-]

Really nice!

Do you have plans to release it for desktop (maybe using Electron)? One of my main complains about tools like Tinkercad is that they are browser-based, and it's easier/faster to have everything local. That's one of the reasons that I moved to OpenSCAD

prmoustache |root |parent [-]

Browser based is not opposite to being local. Given it is open source I guess you can self-host locally on your computer.

Towaway69 |next |previous [-]

Does this work with quest headsets?

Are there “simple” tooling for doing 3D sculpting using quests? Any recommendations?

serf |next |previous [-]

it's a valid idea to try to make an 'easy' modeling/cad system -- the problem is that every such version of software that takes on that challenge makes the suite near useless or incredibly hard to use/clunky for experts. glaring holes, omitted features, zero render or post-process support, poor drawing/print support, etc.

i'm convinced this will always be the case and that the right approach, if there is one, is to take an industrial grade cad/modeling suite and attach a guided 'novice' interface onto it rather than making a novice-based cad system from scratch.

I haven't seen an example of that, either -- but it feels like an easier approach.

darksim905 |root |parent |next [-]

I am a very basic beginner/familiar with tools like Solidworks that have every feature under the sun, but I never really took a class or learned beyond experiencing it at work for a company that made cosmetic packaging. What you're saying is somewhat right, because unless you're a trained engineer who even understands what the hell GD & T means, yeah, you're going to be confused as hell from a design perspective.

CAD/CAM software with Cartesian planes are already confusing as is for most folks. Once I started watching some videos discussing tolerances and such from an engineer perspective, the layout, tooling and concepts made a lot more sense to me.

At that point, it's essentially understanding the intention of the tools and if you're performing additive manufacturing or subtractive. A lot of CAD/CAM software is geared toward machine shops and setups and mindsets like that and not necessarily 3D printer-esque communities. I think these solutions are great for the 3D crowd and not so much the engineer.

There are some in-between things that break the mold and do things in unique ways for people who are product designers like Rhino. The node editing in that is so cool. I look forward to seeing what more people can do with AI now and scripting.

btown |root |parent |next |previous [-]

The problem, of course, is the moment you have some aspect of the model not representable by the basic primitives - do you make it impossible to switch back to the beginner interface?

I'm reminded of the concept of "ejecting" from e.g. Create React App a few years back - the idea was that if your beginner-friendly interface is actually built on the same underlying engine (in this case bundler and deployment assumptions) you can have full fidelity when you need customization, albeit with a one-way transition.

In the JS world things moved more towards build systems where beginner-friendly-DX and full-configurability could coexist. I'm not sure that CAD has the same dynamic.

Perhaps something like nested layers could work: you can use a complex model as a layer, but only opaquely, and build things around it with solids-and-holes; you can then lift that to itself a complex model, do things with professional CAD, and then treat the result itself as an opaque complex model if you switch back? That gets complicated fast.

desdenova |root |parent |next |previous [-]

This is closer to an open Sketchup than a Blender. I think they fill different niches.

dofm |root |parent |next [-]

Much closer to Tinkercad in operation (negative solids etc.)

thenthenthen |root |parent |previous [-]

Agree, this is very much needed.. is there a repo? Would love to add fillets at some point

zuzululu |root |parent |previous [-]

I didn't get that impression it was trying to be anything other than a beginner friendly tool, did you ?

Why can't we have both ?

WillAdams |next |previous [-]

Nice!

I use BlockSCAD a lot --- maybe an option for more math-oriented students?

Some quick testing:

I really wish the dimension objects around the editable dimensions were draggable (the "nudges" (up/down spin arrows) are nice).

I couldn't get "Hole" to function as I expected (assumed it would remove itself from all other objects).

No panning of the 3D view?

stiray |next |previous [-]

Was preparing a rant about CAD solutions being free as long as people didnt get hooked and then closed behind paywall with ever increasing price.

Then I see that I can self host it. This is interesting novelty.

But there were so many bad apples in CAD area, that I will stick with FreeCAD and after 20 years, if this will evolve and prove it is just not another scam, happily give it a try.

darksim905 |root |parent [-]

Bad apples?

contingencies |root |parent [-]

Many major companies switched local forever-licensing to monthly SaaS payments. Others which looked promising (SketchUp) were effectively sold/abandoned.

danbruc |next |previous [-]

Bug report. In blueprint the selection rectangle does not align with the mouse cursor except at the origin.

majorchord |next |previous [-]

Now if only they could master a name everyone can pronounce.

krapp |root |parent [-]

I've probably heard half a dozen pronunciations of "Godot" at this point, it's fine.

bethekidyouwant |next |previous [-]

is there a modifier to move an object on the Z axis? the draw a shape is neat, would like to be able to create straight lines.

joachimhs |next |previous [-]

I run a small makerspace for kids and teens in Norway, and we kept hitting the same wall: the 3D tools we tried were either too fiddly for a 9-year-old or locked behind installs and accounts. So I built Akse — a browser-based 3D modeller for beginners, where the path from idea to a printable STL is as short as I could make it.

It's deliberately limited, but with a rather powerfull 2D Blueprint mode where you draw an outline on millimetre paper and extrude it to 3D. You build by placing and combining primitive shapes (box, cylinder, sphere, cone, pyramid, wedge, torus), set any shape to "hole" mode to cut it out of another. Everything is in real millimetres, so what's on screen matches what comes off the printer. Output is a single STL. That's most of it — no parametric constraints, no assemblies, no fillets. For teaching beginners that's intentional, not a gap.

The obvious comparison is Tinkercad — same space (primitive-based, browser, education-oriented), and I'm not claiming Akse is better. The differences are that it's open source, embeddable as a Svelte component, works in Norwegian as well as English, and is even more stripped down. It mostly exists because I wanted something I could shape around how our workshop actually runs, and put in front of Norwegian-speaking kids without an account or install.

Under the hood it's a Svelte 5 component using Three.js for rendering and three-bvh-csg for the boolean operations; storage goes through a small port interface so it's backend-agnostic, and the standalone version just uses localStorage. It's early (v0.1) and has rough edges. I'd really value feedback on where it trips up first-time users, since that's the entire point of the thing.

Source (AGPL-3.0, with a commercial option): github.com/joachimhs/akse3d

phkahler |root |parent |next [-]

I have similar hopes for Solvespace - that every middle school student can pick it up and design things. We have a couple issues that keep me from recommending it too strongly though - bugs in the boolean code are IMHO the biggest blocker for kids.

Any chance you could have the kids make comparisons between the two? Solvespace is completely constraint-based, so it may be a bit harder to learn but also more flexible.

It's a single exe, but there is also an experimental web version: https://solvespace.com/webver.pl

dofm |root |parent [-]

> that every middle school student can pick it up and design things.

The Solvespace UI is a long, long way from being the sort of UI a contemporary kid has any kind of comfort with, I'm afraid, and will be obtuse even to teachers (many of whom, with subjects that concern technology, do not have time to develop expertise in an obtuse UI and may indeed only be confident they understand the meaning of all the lessons they are teaching and not much more).

I don't think bugs in your booleans are your biggest problem at all.

I think Tinkercad has weaknesses as a classic CAD package, and there are things I would like to see done better, but as a package to teach younger people how core concepts in 3D modelling (rather than the ontologies of bRep) actually work, it is the standard you are working against.

WillAdams |root |parent |next [-]

A UI which seems a bit more in-line w/ contemporary expectations is Dune 3D:

https://dune3d.org/

which I find a little less confusing than the traditional 3D CAD packages I've tried (and failed to learn) --- at least for Dune 3D I've made it through the tutorial successfully.

dofm |root |parent [-]

Dune 3D is OK — and is after all dependent on Solvespace.

phkahler |root |parent |previous [-]

>> The Solvespace UI is a long, long way from being the sort of UI a contemporary kid has any kind of comfort with...

Respectfully, I disagree. Even adults have used the word "fun" to describe using solvespace. But I don't actually have feedback from kids, hence the question to OP.

dofm |root |parent [-]

Don't get me wrong, I think Solvespace is really important and I am glad you are keeping it ticking over, but it is the least fun UI I have ever encountered, I'm afraid (kind of obtuse on a trackpad too)

Not saying any of this is easy — Shapr3D on an iPad, expensive and marketed on its extraordinary usability, is just utterly perplexing!

SecuredMarvin |root |parent |previous [-]

Group transform is out of order. It does not transform the group but the elements. This leads to the suspicion that position and rotation are not transformation chains on object trees but attributes. That would be the wrong architecture.

I am very sorry, but please explain. Why is this a nice looking Svelte / Three / CSG app, but the basics are wrong?

felixyz |root |parent [-]

Someone built something useful and you complain that it's "nice looking"? Consider reframing your feedback.

master-lincoln |root |parent |next [-]

No, they are complaining that a basic functionality is not working and hinting at not having the right architecture to deliver correct results in more complex scenarios.

bethekidyouwant |root |parent [-]

but this is clearly not meant to be onshape, its for kids who want to 3d print something. the 'hole' mode does not seem to subtract like I would expect.

SecuredMarvin |root |parent |previous [-]

Yes, sure. I will try to be more blunt, as I was not trying to complain about anything.

I think you have invested with agreeable results, but you missed fundamentals. Adjust course to "architecture first" and I expect a great product.

Gracana |root |parent [-]

I think that's a lot better. "Please explain your wrongness" really had me scratching my head. Hopefully they can make improvements to the 3D architecture or find someone who can help.

palmblock_dev |previous [-]

[flagged]