Implementation of chord primitives

[Peter Blomberg]

16-sided chords come in two types; 1-8, 3-16, 1-4, 5-16, 3-8, 7-16, and 2-4 are made of mostly quads while 5-8, 3-4, 13-16, and 7-8 are made of only triangles. I get that the final triangle count is the same no matter how the triangles are organized.

Nevertheless, the triangle-only solution has a lot of narrow triangles and they are concentrated to the same area. Why is that? I would understand it if the chords were created additively (previous chord + one triangle = next chord), but they aren't.

Should we do something about it?

[Magnus Forsberg]

16-sided chords come in two types; 1-8, 3-16, 1-4, 5-16, 3-8, 7-16, and 2-4 are made of mostly quads while 5-8, 3-4, 13-16, and 7-8 are made of only triangles. I get that the final triangle count is the same no matter how the triangles are organized.

Nevertheless, the triangle-only solution has a lot of narrow triangles and they are concentrated to the same area. Why is that? I would understand it if the chords were created additively (previous chord + one triangle = next chord), but they aren't.

Should we do something about it?

But they are created additively. And they should be made like that if we follow the rules for a primitive.
"Where this fraction is less than an entire circle, the primitive starts at {+x,0} and progresses in a conterclockwise direction when viewed from above {-y}."
If you make a chrd prim in PrimGen2 it is made like that. All triangles starting in the same point (1 0 0)

This however is not important for the mathematical "primitive substitution" that is peformed by some of the SW.
It doesn't care about triangles or quads. Only the radius matters in the mathematical visualisation.

All the chrd-prims with quads have been manually edited after the creation, presumably to avoid thin triangles and instead use larger quads.

Should we change them? I my opinion, no.
All the "longer" chrd prims, 5-8, 3-4, 13-16, and 7-8 should be avoided in a design. It is often better to use "shorter" chrd prims.

[Philippe Hurbain]

All the "longer" chrd prims, 5-8, 3-4, 13-16, and 7-8 should be avoided in a design. It is often better to use "shorter" chrd prims.

What problem do you see with them?

[Magnus Forsberg]

What problem do you see with them?

Let me rephrase that. I don't see a problem with the chrd prims.
Making new, "longer" edge, cyli, con and ring primitives is to be avoided.
Anything longer that a 1-4 can often be made by adding 90 deg rotated short prims.

[Peter Blomberg]

If the longer chord prims are to be avoided, could one not update existing parts with the corresponding shorter prims via script and then obsolete those longer prims?

It disturbs me that they are not made of quads

[Orion Pobursky]

If the longer chord prims are to be avoided, could one not update existing parts with the corresponding shorter prims via script and then obsolete those longer prims?

While I could, that's a lot of effort to fix something that isn't really broken.