5 edge vertex

[johngriz]

I'm a beginer character modeler. and I was hoping somebody could tell me
if 5 edged vertexs are definitely a bad thing ( five sidded star) when poly modeling.
See attached imaage. How could I rework this model so I don't have these ???

[chris-giusti]

They're not possible to avoid in certain areas. Even Valve's characters have a 5-edged vertice in this area (Because it gets triangulated anyways). I'd say download some characters and take a look at what other people have done to resolve this.

5-edges vertices and even triangles aren't bad if you have no other option, but just avoid N-gons (5-edged or greater polygons).

[joojaa]

Any 5 edge vertex polygon can be split  a few more 4 sided ones! You just end up with one more loops (and that may be a problem in it self).  Its not certainly as bad a s a 5 sided polygon. But still it can be worked out if you really want to put in the effort. Example:

PS: your suffering from severe case of box modelling syndrome

[chris-giusti]

Careful, because that particular method adds 2 vertices which could be just as unwanted as the 5-point vertice (situational).

Unless someone shows me otherwise, having a triangle or just leaving the 5-edged vertice doesn't make too big of a difference unless you absolutely must have a quad over having a triangle (e.g. during modelling phase). Even then, most programs will cope.

[joojaa]

Its not so much the computer that has problems. Its the artists. The loop is easy because you can control the flow in 2 directions, the parametric surface is 2 dimensional. but when you make a five sider vertex then you have a pole that behaves unontrollably. Hence best avoided if possible, or moved to a location that is not expected to deform.

In this case the char has room for the loops. you can split als the upper quads into  3 quads each but this is problematical for loop flow. So if you have time then adding a loop is no biggie. If not then you sacrafice.

PS the computer also has no problem with triangles, or pentacles, if the algorithm is done right, it just created the uncontollable pole in the center of the n-gon. Which is much much worse but if you dont need to control the pole then no biggie.