Erikas - Bevel Sucks Video


Active member
Erikals - Bevel Sucks Video

I subscribe to your channel - a lot of really great information and interesting experiments.

On your latest video, I completely understand the sentiment behind it. Rounder as a tool leaves a lot to be desired.

But I took a different approach and I think I got the results you were looking for, although only you would know that for sure.

Some pics.

Starting Cube

First Edge Bevel

Rounding first edges

Delete polygons and chamfer edges


Final result

I mean, I know it isn't using Rounder or edge bevel to achieve the rounding; but the final result appears to be what you are looking for, with significantly fewer steps.

No? Yes? Usable?
Last edited:

no, unfortunately not,

it creates a box with a rounded shape.

could be of use it certain scenarios, but mostly not.

the polyflow is also inconvenient.

so the "solution" is LWcad at the moment. I advice everyone to buy it.

thanks for testing, it might be of use in other scenarios.


  • preview.png
    20.7 KB · Views: 219
Last edited: won't be of help here tho'
an alright tool, but used for other operations, quite unrelated.

I was just responding to John's mention of LW's destructive workflow. There are some ways to deal with it, like with this plugin from the full TrueArt modelling pack. It can also undo bevels from Bevel++ (not just LW's native bevel tool).
I don't have TA Modeling pack so i can't say how good or not it is and how up to date his plugins are. I find Bevel ++ to be much better than native and whilst it's not totally no destructive, its certainly more predictable.
Quick link to LWCAD 2020 demo for round tool (~12:30)

You appear to be reinforcing Erik's point much more than refuting it: LW's native bevel/extrude/shift-type tools lack usability/quality, and obtaining reliable, usable ones costs extra.

IMO, LW11's "Chamfer" should have been a catch-all superset tool of like "C4D Bevel"-level capability. IMO, it made zero sense for Ikeda to produce another "subset tool", whose subset was even SMALLER than existing LW tools. I find it makes even less sense that Viktor went further, producing a BUNCH of "subset tools" in LWCAD broken up along LW-similar role lines. WTF?!

Honestly, it's not the UX mistakes that bother me, it's the hell-bent dedication and focus on repeating the same ones over and over that really sets me off.
Last edited:
You appear to be reinforcing Erik's point much more than refuting it: LW's native bevel/extrude/shift-type tools lack usability/quality, and obtaining reliable, usable ones costs extra...

You appear to be misunderstanding this conversation if you believed anyone is here to debate or refute something. We're just here to discuss workflow solutions. Opining & blaming is pointless, but you're welcome to carry on if you find some value in that. :D

a follow-up to that other video >

buy LWcad here >
That's pretty much how I would approach it for the most part. Native Rounder is powerful, but it has many undocumented rules that took me a long time to figure out. There is one bevel I wish it could pull off, and that's a bevel on a curved surface.

I've attached two shots:

The first is a Boolean subtract which isn't uniform.
The second is using Rounder which gives the intended effect, but . . .

If you need to bevel a larger hole and the hole is a larger diameter that crosses the lines that make up the curved surface, Rounder will generate a mess, requiring a lot of tedious manual clean-up to get everything looking proper. This type of beveled hole can be applied to a sub-d surface and patched into a hard surface object, but again, that's more effort.

Anyone know of a faster, better, easier way to pull this off for a hard surface object? (If I have to buy a plugin, so be it.)

Thanks in advance.

Hard Surface Boolean Subtract.jpg

Hard Surface - Rounder.jpg
Modeling operations on curved surfaces and fillets/rounds..cad tools handle that better with solids, not really any issue or limit to perform fillets, then again you will be limited in other areas depending on the purpose, and you may still have to wrestle polygon clean up if you aim to render it rather than use it for print or construction.
The intent is for 3D printing RC submarine hulls, (4mm thick) some of which contain external hull openings like this DSRV:


The intent is for 3D printing RC submarine hulls, (4mm thick) some of which contain external hull openings like this DSRV:

View attachment 150265

View attachment 150266

I am not an expert in Designsparks, but know a little a bit about it, it´s free if you sign up an account, and you should be able to apply array circles or cylincers in arrays around a dome or shape as for the hull of the submarine, use various commands like combine, then split then use the pull tool to fillet/round as you like without having to mess with polys.
Provided that the resulting stl or export would be good enough to send straight away for the 3D print that is.

I did try the above so it works on curves surfaces, or cylinder shaped, domes etc, and to some degree you can import obj objects, convert to solids, so you could perhaps start the model in lightwave.
But ..doing it all in designsparks let you start with everything parametric mostly, and changable and with a fair amount of accuracy with dimensions and easy to change fillets or dimensions when needed without rebuilding it all.

If you have to render or continue to model inside a polymodeler, then it would be more hassle though, but for straight 3D prints..perhaps an alternative.

Thank you for your replies; I really appreciate the advice.

The Circle Band technique works pretty well - enough that I can manually make point adjustments to get the mesh smooth enough for rendering and 3D printing while minimizing faceting, which involves a lot of sanding, though I have been having pretty good luck using XTC-3D to fill in layer lines and provide a uniformly smooth surface for later painting.

I suppose I could output the hull sections in a compatible format to Fusion 360, modify them, and save back into LW, but I'll make another go of it using the above video. It's a little more work in sub-D to get super-accurate symmetrical holes, which is why a hard surface modeling technique is preferred. When the thruster propellers are installed, the clearance tolerances need to be tight for the blades in the ducts and mounts for the bushings that support the thruster motor shafts.

Thanks again!
Top Bottom