Claes Lundstrom

Have you tried exporting as OBJ instead ? Based on my experience, it seems to work best when communicating with various apps. I only use point clouds when the purpose of the scan is for use on terrain models. I don't know if Nomad supports export as OBJ but I guess it does. I use the free Scaniverse app with my iPhone. As always, use a gimble video stabilizer when scanning as it improves the quality and accuracy of the scan a lot by smoothing the phone motions when scanning. Here is an example being just under 30 meters wide and 181 square meters. I have compared it with drawings drawn based on traditional laser measuring, and the results were surprisingly similar. I actually spotted two small mistakes in the manual measuring. The mid picture shows the plans as it came out. My only objection is a slight angle drift on the right side. The rest was more or less spot on. You can't expect mm precession of course, but it is very helpful indeed. The scan was done in a single session that took 18 minutes. The last picture shows a technique I use for extracting more precise lines. I simply use the clip cube and select an area in the middle. Why? Because these scans tend to have a problem with dramatic changes in direction (e.g. floor to wall). It tends to look like a worn piece of soap. In the middle of the walls, you get less of such problems.

It's pretty easy. Just export as STL, and then import into the slicer program provided by the printer manufacturer. It converts the file into code understood by the printer. The tricky part is to know how to design functional models. When designing for printing, you have to know what works and not. The same applies if you design for later cutting, cutting with knifes, 3D milling, etc. They all have issues to consider. My suggestion would be to download a few models and see how they are made. Also start with some simple models and see how it goes. After a while, you will get the idea. It's not difficult, but it takes practice.

You don't need a 80x80 cm printer do print a one meter model. I can print up to a one meter model on my AnyCubic Cobra Max 2, with a print box of 42 x 42 x 50 cm, and it's significantly less expensive. What you do is that you split it in the middle. Besides having to glue it together, you get a cleaner print, and you generate a footprint for the model to stand on, with two flat can corresponding surfaces. I typically add a "frame" shape to add more surface area to the glue and to make it stick better to the bed. I also print just a shell, which can be say 1 to 1.5 mm thick, which means that it prints faster. For boat with normal proportions, you can easily print both ends at the same time. Having three sections could expand the model Lise to 1.5 meters and so on. It just takes longer to print. The one meter model would take 43 hours to print according to the slicer app.

Yes, I have tried both Orca and Rhino. Not for me. Sure Orca does offer some features I don't have, especially for very big boats and ships. I started with boat design way before there was such as thing as reasonably priced software for boat design, so everything was done by hand with ink on vellum. I therefore prefer a program that works in a similar way. I appreciate having the skills do to a nut and bolt design from scratch without the constraints of parametric design. I also appreciate the fact that I never sketch on paper. Everything is done directly in 3D. I can do that because the program allows me to do it very quickly. Why use VW? On smaller leisure boats, the inner structure is often much more complex than the outer skins such as hull, deck and so on. The process in many ways more similar to architecture, except that everything is curved, and highly compressed. It therefore makes sense. I also use it for compiling production drawings. Besides me, check out Laurie MacGowans web page. He is located in Nova Scotia, and uses the same toolkit as I do, including VW. He used to have a column in Wooden Boat magazine, where he did a new design for each issue based on requests from the readers. https://www.mcgowanmarinedesign.com

Just out of curiosity, I have spent a lot of time watching YouTube videos in apps like Solidworks, Fusion 360, Rhino, FreeCAD, Blender, Shapr, Freeship, Delftship, MaxSurf, etc. Can't say that I have been impressed with what I have found. These movies seem to be divided into three groups, first those who copy somebody else's drawings. Making a copy of somebody else's designs is not really boat design, it's just modeling skills, without knowing much about boat design. The second group, trying to free form design boats, illustrates how badly the standard CAD programs work. It's seldom impressive. The third group use parametric features, which can work ok I guess, but tends to generate rather dull close variations of the same theme. The biggest weakness of the standard CAD programs is that they lack marine design specific features such as proper dynamic cross sectioning, which is a key tool in hull fairing, hydrostatics, weight and center of gravity and stability calculations, practical functions of the boat, etc. All these functions are key features that you calculate and evaluate throughout the design process. Personally, I always use a dedicated boat design program, and use VW mainly for documentation, 2D drafting, and when I need solid modeling. It just works better. As for Vectorworks, yes, it fits into the standard CAD category. In that category, it does some things OK. You can have interpolated control points, which you can't do in Rhino as far as I can tell without being an expert user on it. VW's biggest flaw, as I see it, is the inability to edit more than one NURBS object at the time. In my boat design app, I can edit any number of controls in any number of objects at the same time, and that makes a huge difference when you have say 500 or 1000 NURBS surfaces in a model. Another flaw is that you for some strange reason can't nudge NURBS control points like you can with meshes. Why? Why? Why? Very frustrating, as real boat design typically consists of hours of micro adjustments.

Have about 80 realized professional boat design projects and between 1000 and 2000 built. The picture shows a 3D lapstrake printed model and it's augmented reality sister. Same model, different media.

All good points. I too use Scaniverse scans a lot. What surprises me is that I have never actually met a single contractor using these methods. Feels like leaving an essential tool while documenting a job. A recent breakthrough. I had an idea: Would it actually help if I were to use a DJI Osmo Mobile Gimbal stabilizer (or similar) ? So, I tried it. The answer is YES, the quality of scans actually goes up A LOT. Even better, it stabilizes videos with scanned models saved as USDZ too, which I also use a lot (no internal stabilization in Apple Preview USDZ files for some enigmatic reason). Ozzie2.mov

Can't say that I do, no.

The first Mac was released on January 25 1984. MacDraw and MacPaint were probably released thereabout or soon after. MiniCad came out probably 1985 or 1986, as very much as an extension, upgrade, and evolution of MacDraw, together with others like ClarisCAD and PowerCAD to mention a few. I first came across MiniCAD in 1987 (MiniCAD 2), and preferred it not because it was the fastest, but because, unlike the rest, it had 3D.

Thanks Vlado and Jesse. Ok, so not a problem then. Phew. I use them a lot for some highly specialized tasks, that speeds up the tasks dramatically. They are not complicated, so hardly not secret of fantastic in any way. I actually use Vectorscript a lot though, as an intermediate format from my other CAD program, where I do most of the basic stuff. Sure, I also use DXF and OBJ too, but I have found Vectorscript to be more accessible than DXF.

Does it mean that I can't use my own plug-ins without going through an annoying dialogues/ paper work. I wrote them for my personal use and for use within my team? That is really, really bad ! They are absolutely essential for our work. I can understand it for external plug-ins that you buy, but not for in-house developed stuff.

What a mess. Probably better than what we have now though. 😉

I usually use Apple's free "Reality Converter" to generate USDZ files from OBJ files. It can be a bit hard to find on Apple's we page, as it's defined as a beta. Works very well for my files, and I have hundreds of USDZ file on my iPhone. No GLB though. I think many rendering apps (like Blender) are able to do this too. I occasionally use Keyshot for more high rendering quality files, and it works well exporting these formats.

The problem with Multiple Extrude is that you lack control of where the starting points are located. You can flip / reverse the direction of a polygon or curve. That's good. You can step through the individual control points and edit the location of it. That's also good. What you CAN NOT do is to move the starting point. That's the problem. Let's say that an extra option would be added in the Move popup, reading "Move Starting Point", and then do the editing with the arrows below. That would make Multiple Extrude way more usable and predictable. I have had it in my TouchCAD for years, and trust me, I use it a lot. As it is now, the best way to avoid problems is to avoid using individually drawn objects. Instead make a copy of one of the parts and edit that to the new shape. In that way, you at least work with objects having the same basic DNA, and therefore avoid all sorts of problems.

Yes, CorelDraw is a nice option too, though a bit more expensive than Affinity Designer. It seems to be very popular among people working with for example vinyl and laser cutting. I have an old version, but rarely use it anymore, as they left the Mac scene for a number of years. They are back now though. The user interface feels fairly familiar to VW on the 2D side. It's not like it came from another planet, so you have a decent guessing factor. Fairly reliable communication with other apps too. One of the points where I actually do use CoreDraw is for vectorizing images. It's better than VW in that respect. In reality though, I rarely use that feature, as I mainly work with image based 3D signs and physical photo realistic 3D models, where there is not much need for vector graphics.