Stuck on a very basic modeling problem - resolving complex poles

tomatthefarm

Hello! I'm new to the forum because I really want to improve my very, very basic 3D modeling skills. 

I've been stuck on this problem for a couple of hours so I thought I'd consult some people who might know better. I'm trying to resolve the legs of this tripod object into a single three-sided stem and there are some complex poles (purple highlighted vertexes) that appear to be distorting the mesh as soon as it reaches the stem. I don't think those triangles are helping...



If anyone might be able to suggest a nice approach to solving this, I'd be very grateful. I included a link to the .fbx below in case anyone wants to take a further look. 

FBX: https://file.io/uApB1I6TZAqX

Many thanks! 

Alex_J

whats the reference?

tomatthefarm

ah there's no reference as such, just thinking of a metal triangular-based stand.

Alex_J

reason I asked is because i thought maybe you were doing a weapon tripod or something like that, in which case all the parts would be separated. In general its best to model things same way they are constructed in real life as far as is practical.

in any case, i am not the expert on this stuff you should probably take it to the hard surface modeling threads, but I think if you split those triangles in two at the apex and let the resultant edge loop just run all the way up it should be fine. You may need to readjust the edge hardness/softness, but I only know how to tell you what to do for that in maya terms. In maya that would be Unlock Normals and then can use Soften/Harden to adjust all with a threshold angle for what becomes hard.

FrankPolygon

@tomatthefarm Welcome to Polycount. Consider checking out the forum information and introduction thread and the dedicated modeling question thread.

If the legs of the tripod are supposed to be made from square tubing then the issue is the surface produced by the intersecting shapes can't be coplanar. The joint between the perpendicular surfaces of a square cross section will either produce a seam in the middle or require a triangular gusset. It is possible to create a similar joint with coplanar surfaces but the tubing needs to be a rhomboid shape.

Here's an example of how intersecting perpendicular planes produce a joint that's non-coplanar.

Trimming this joint back to the opposing vertices produces a triangular gusset that's angled differently to all adjoining surfaces.

Changing the tube shape to a rhombus creates a joint that produces a continuous, coplanar surface between both sides of the adjacent legs.

Different joint orientations can be produced by simply rotating the triangular tubing in the center.

It's also possible to create other joint shapes by changing the size and orientation of the square tubing used for the legs.

One of the major advantages of using references is most of the concept and design work is already done. Which makes it easier to focus on learning the technical aspects of basic 3D modeling. As far as modeling goes, try using a shape based, iterative block out process. Focus on creating accurate shapes then use those shapes to help define the loop flow paths around the rest of the mesh. Avoid making overly technical assumptions. Continue trying different approaches using different shapes and compare the results to the desired outcome.

tomatthefarm

@Alex_J thank you very much for your explanation. Your answer helped me identify the issue with the phong shading happening with Cinema 4D yesterday. I'm beginning to see the importance of using real-world examples to practice on.

@frankpolygon thank you for the welcome to the forums! My apologies for having posted in the incorrect area! I can't thank you enough for such a comprehensive reply and set of accompanying examples. I see now that my example was fundamentally flawed in its assumption about the geometry of the object. The second line of your other post "Working without enough information to visually interpret the shapes tends to add unnecessary frustration" certainly rang true!

Unfortunately, I wasn't able to tell that the angle of the legs and gusset were not coplanar, but hopefully, these details will come with time and practice. I'm going to try again with the approach you mention, blocking out the rough shapes and then trying to identify the crucial edge loops. Once again, I really appreciate the demonstration and the pointers, thank you.