I'm waiting on an overcast day to take a bunch of pictures to build a 3D model of my boat.
It would be cool to print out as OP suggested for his model, but I really want it to figure out a major refit of my cabin and cabin entrance as well as the interior. It would be nice to mock it all up and walk / sit / lie down etc... in VR. It would allow me to visualize in a useful way where the lines and cleats would go, how difficult is it to get in and out, or move around. etc...
I doubt it will be. That's okay, I'm not building from said 3D model, more like a cardboard mock up. Actual refit would happen onto the boat itself with rough dimensions from the mock-up.
The problem with LiDAR things I’ve found (I’m a total newb at this) is the point cloud generated by Polycam or something is ultimately only kinda useful.
It sees a bunch of rando anomalies that aren’t there and doesn’t have object or straight line recognition that would ultimately help it decrease vertices and make smoother, cleaner, models.
The ability to edit the output and /or simplify to tell the machine to Aline these 35,000 points along a straight line of an ibeam(for example) is limited.
Again, total newb. But my takeaway so far is it’s easier and faster for me to just build a clean model from scratch.
Right, licensing costs, its been a minute since ive bought my own software licenses, if youve got any secondary school affiliation, you may be able to get the education edition for free.
If you're a professional digital artist or engineer who works in CAD all the time, maybe this will be easier, but a boat is not an easy thing to start your "teach yourself CAD" journey with. ;)
I use to struggle in cad software too.
I started modeling boats 5 years ago.
I'm only a ship building technician with one failed year of ship architecture. So my design May not be good in anyway. But I have developped a method to quickly design accurate boats in blender.
It's no cad software but it has the adventage of being way more flexible.
The secret lies with subdivision surface modifier, shift+E to flatten faces that need to be flat, miror modifier (symetry, obviously better when you're designing a boat), and the boolean fonction to creat hulls and other hollow blocs using différence, and stiffeners as well as the all interior éléments with intersect.
Would take more time to explain but that's the basic of it.
If it makes you feel better, I worked on a boat where they did a 3D scan and model of the engine room to replace the engines.
Took the most prestigious marine architect firm in the US around $60k and 2 months.
The final result was almost 2" off, which is hilariously awful when aligning engines. 3D modeling is hard, apparently. This was 4 years ago and I think they are still dragging out a lawsuit, good times
Do you mind explaining what/how you’ve done it? I have scanned and measured a lot of the interior parts (cabinets, galley, head, settees etc) but waiting for the next haul out to take a crack at LIDAR scanning the hull with my phone. Given the movement of the boat in the water LIDAR has not been kind to me and was hoping for ncw its on the hard I’ll get better results.
I made a parametric model for a simple sampan once, using grasshopper/dynamo. (Means literally 'three planks' so it was generated with three curved faces). The idea was to be able to move sliders for length, width, transom position etc. and then it projected the curved faces onto flat drawings, that you could use to cut plywood.
22 Comments
hilomania@reddit
foilrider@reddit (OP)
hilomania@reddit
Ellusive1@reddit
foilrider@reddit (OP)
Ellusive1@reddit
cleverpunnyname@reddit
foilrider@reddit (OP)
-The_Nak-@reddit
foilrider@reddit (OP)
-The_Nak-@reddit
foilrider@reddit (OP)
LeewardLeeway@reddit
foilrider@reddit (OP)
strangefolk@reddit
OregonWoodsChainman@reddit
ad-undeterminam@reddit
whyrumalwaysgone@reddit
PooFlingerMonkey@reddit
yowhywouldyoudothat@reddit
PooFlingerMonkey@reddit
Cease-the-means@reddit