Dimensioning!
Since many are using blender for 3d printing, accuracy becomes important. I have been looking for ways to dimension quickly in blender. There are several ways to assign dimensions in blender but when we move ahead in modeling, modifiers gets stacked and it becomes harder to edit models. How about introducing option to dimension quickly very similar to CAD software? Like pick edges or edge and center of circle etc.... I have shown an example with image attached. Pardon my ignorance but I have not found the way to do this quickly...
11 Comments
there is an addon for that i used it often when 3d printing. I rather would see support for real open 3d formats. Blender is great for concept designs, but walk with your .blend file to a industrial manufacturer (a plastic extruder company for example) and they start to laugh.
* @peter.boos * I would agree as technical people see this funny. But CAD software are now including subD modeling too. But organic modeling capacity with blender is limitless while on CAD side we rely either on capabilities they provide or let generative design simulation to figure out.
* @jehankothari27 * I agree with peter.boos . I'd dig into freecad or other solid cad modeler if you need those features. You can export to blender if needed, with compromises.
YES YES YES! ;-) ... can't wait to have option to export section cuts as a pdf with dimensions on it.. blenderCAD would be nice!
* @info31 * yes...if that happens...I can eliminate cad models for simple 3d modeling purpose. My whole pipeline can fit in here..
* @jehankothari27 * i thought that there are some peoplee in swiss making some cad with blender.. but yeah.. if I could export plans and stuff.. can't imagein what could be possible to work with the grease pencile.. interior designer and architects would easily already have a stylized render method.. intuitive drawing ob objects and buildings with the grease pencile... same goes for me.. my whole workflow would fit in blender. also it would be a blast if we could use ifc data...
I'm thinking of this from a practicality standpoint. The reason it works in CAD software is because we/the computer are often dealing with shapes such as circles and path objects, which allows for the computer to know where the center of things are, and even just what the things are themselves.
In Blender we work with meshes, which as far as the computer is concerned are a bunch of random points in space. While we might see a hole, Blender has no concept of that, and it would be very difficult to instruct Blender what to count as what.
It would be a nice feature to have, but I'm really racking my brain trying to think of how a user would interface with this system
* @Zach Hixson * If I understand you correctly, then I think this is the issue of relative and absolute measuring (the way we have for array). I think picking up dimension from the center will be terrible workflow because we need to move models around( I am not sure though). We can be pretty good at placing a base structure and then small features like hole etc can be place relative to the points we pick. I also don't understand the workflow behind blender but then we have curves which can spin or extrude and they are not meshes. This can be definitely think upon, in depth, provided if this is considered as enhancement because when we apply modifier our dimension will become useless ( and this is true in CAD too)
* @jehankothari27 * Not necessarily what I was saying, but that might be another problem as well.
In CAD, if I were to make a cube with a hole in it, the workflow would be: Draw square > Extrude > Draw Circle > extrude circle as hole through box. Throughout this process, the software is "thinking," of these as "Circles," and "Squares," something like
{Square(p1: 0,0, p2:1,1), Extrude: 1}, {Circle(p: 0.5, 0.5), Extrude: 1}. This means that if the computer wants to get the center of the circle, it has the individual shapes, and can process it accordingly.In Blender and other mesh modeling software like 3DS Max and Maya, it will remember a cube with a hole in it just as a bunch of points such as
(0, 0, 0), (1, 1, 0), (1, 0, 1), (0.23, 0.4, 0)...and has no way of knowing what parts of those points "belong," to the circle, and what parts belong to the square, as it's all just one mass.You are right in that we could dimension different objects together if they were booleaned, since then the software would be aware of the separation between objects, but that would make the tool very limited.
* @jehankothari27 * The only implementation that could be done would be a very simple tool where you just link it to verticies and edges
* @Zach Hixson * Ohhh Ok!!! Didn't know that. Thank you for elaborating... I had this another thought too but not sure how feasible when we consider coding...say we want to cube with a hole...then instead of performing boolean first..we make cube as a parent and cylinder as a child....Can they now behave as a system alone where parent is the center (origin) to locate the cylinder?