[ 3 / biz / cgl / ck / diy / fa / ic / jp / lit / sci / vr / vt ] [ index / top / reports ] [ become a patron ] [ status ]
2023-11: Warosu is now out of extended maintenance.

/3/ - 3DCG


View post   

File: 752 KB, 1237x906, how-do-I-3d.png [View same] [iqdb] [saucenao] [google]
744753 No.744753 [Reply] [Original]

So I added the vertices circled in red in order to fit the inscribed circle with quads, which is good. But the issue is that this effects the subdivision surface, making this particular section slightly flatter. How is this avoided?

>> No.744755
File: 515 KB, 989x832, how-do-I-3d.png [View same] [iqdb] [saucenao] [google]
744755

Example of what I mean by "flatter". The orange line is a perfect circle.

>> No.744766

>>744753
Add more sides to the cylinder, so the edge is always there, and not only when you place the hole (or just add those cuts manually everywhere). Alternatively, try positioning the hole differently, without adding the extra edge to the cylinder.

>> No.744770
File: 170 KB, 1237x906, Untitled-1231231.jpg [View same] [iqdb] [saucenao] [google]
744770

Why not use existing geomtry? Just move the hole over so you don't have to create new cross sections.

>> No.744771

i didn't understand the question but i hope this helps

https://streamable.com/pxhkcn

>> No.744776

just nudge them manually who gives a shit

>> No.744780

>>744770
Yeah. Seems obvious now.
>>744776
That sort of "just eyeball it, whatever it's good enough" artist mentality drives me up the wall.

>> No.744851

>>744770
>Why not use existing geomtry?
Why not use existing questions thread, anon?

The answer is the same, OP's a retard. Just one of many corona migrants.

>> No.745385 [DELETED] 

ÞumÞ

>> No.745502

>>744780
found the enginerd

>> No.745506

>>744780
>whatever it's good enough" artist mentality drives me up the wall.
If nobody can see it, then it doesn't matter. You can do everything to the most autismal standards but the only person who will care is you.

>> No.745518

>>745506
craftsmen traditionally have pride in doing things right despite not being held to external scrutiny. Modern industry is all about efficiency though, so I understand your perspective. Compare Japanese manufacturing vs their geographical neighbors.

>> No.745537

Just use booleans like a real man

>> No.745555

If you're gonna subdiv it don't go so high as 12 edges in your circle. Go for 6. Same goes for other geometry. Makes no sense why it's so polygon heavy. Just go lower, you have a lot of edges that are redundant.

>> No.745556

>>744780
>That sort of "just eyeball it, whatever it's good enough" artist mentality drives me up the wall.

This is an ironic statement considering you fucked up your subdiv by having irregular polygon sizes. The edge where the hole is located is actually straighter and has a different arc than the rest of the model.

>> No.745600

>>744753
your arrangement of vertices on a flat surface really doesn't matter.

Where topology matters is animation. If you want a knee to bend correctly or face to deform correctly then you need to watch your topology.

When the video card renders your model it converts it to triangles from quads anyway.

>> No.745601

>>744780
>That sort of "just eyeball it, whatever it's good enough" artist mentality drives me up the wall.
fuck you nigger

>> No.745696

>>745600
>When the video card renders your model it converts it to triangles from quads anyway.

No it doesn't, you retard. Converting to triangles is about tangent normals, not topology. If it looks bad with quads it will look bad with tris.

>> No.745800

>>745696
I think his point was to not worry about getting quad only topology and just focus on getting good geometry

>> No.745847

>>745800
Good geometry is quad only geometry. This is especially essential when you plan on SubDiv your model.
OMG, you people are so retarded...

>> No.745878

>>745537
big retard aura in this

>> No.745884

>>745878
Nothing retarded in that my dude. Place modifiers in correct order and you can get great results with subdiv + bool workflow. Obviously don't do that on stuff that has to deform though.

>> No.746048

>>745847
I don't think you know what the words you're using mean lel. for subdiv workflow, ngons in your topology are fine as long as you keep them and their immediate neighbours planar. Just having quads won't necessarily fix subdiv issues bevause star singularities in edge flow may cause incorrect subdiv behaviour, which is exactly what we see in OP's case. so in fact, you are categorically wrong, and I find your self-confidence quite amusing.

>> No.746057

>>746048
>ngons in your topology are fine as long as you keep them and their immediate neighbours planar

What other pearls of wisdom do you have for us? Don't trust people who lie? Blacks are good people as long as they obey the law? Jews are trustworthy as long as they're honest? Bricks float as long as they don't hit the ground?

>> No.746073

imagine not using a parametic modeller for this. i could have this modeled, no kidding, in about 90 seconds

cad gang

>> No.746140

>>746057
I don't see your point. just a rapidly moving goalpost. sorry. Since you clearly just want to argue so that you can feel like you're right and I'm wrong, I'm sorry to have gotten in the way of your satisfaction. let me help you out:

I love blender, am transsexual, read reddit all day, and believe that that on a manifold mesh, faces - edges + vertices = 3.