[a / b / c / d / e / f / g / gif / h / hr / k / m / o / p / r / s / t / u / v / vg / vr / w / wg] [i / ic] [r9k] [s4s] [cm / hm / lgbt / y] [3 / aco / adv / an / asp / biz / cgl / ck / co / diy / fa / fit / gd / hc / his / int / jp / lit / mlp / mu / n / news / out / po / pol / qst / sci / soc / sp / tg / toy / trv / tv / vp / wsg / wsr / x] [Settings] [Home]
Board
Settings Home
/3/


hey guys, what is the best way to boolean and cu shapes out without fucking the topology up.
i been experimenting to see what i can do after a boolean to clean up the topology the best i can to be used in turbo smooth so far best thing is to inset and bridge

got any tips?
>>
I create the shape as a seperate object usually as a plane, and then use it as a stencil by making it transparent and placing over the designated area. I cut that shape along using vertex snapping, delete the stencil, and then just extrude down, delete face, bridge borders, and cleanup from there.
>>
>>506757

the one on left is stencile the one on right is boolean. the same result. it seems fine with simple shapes but the other day i was trying to make a SNES controller and the shapes got all type of fucked up
>>
in this particular case you can use inset
just avoid boolean alltogether and add in normal maps what is too small or impossible model
>>
>>506762
i like to make a lot of my complex shapes with splines and when i extrude them a lot of them break after adding a turbo smooth so i really want to know how i can fix shitty topology all toghater not just after boolean
>>
>>506763
show me a complex shape and maybe ill find a way to hardsurface it with minimal amount of Pisces
>>
>>506765
heres something as an example (i went back to trying to re make the SNES controller)

i tried bridging few key edges and than caping the rest. bridging each and every edge by hand wouldnt be smart/effective seeing as there arnt edges for every edge regardless so the end result was still going to be broken
>>
File: example.jpg (81.82 KB, 1149x493)
81.82 KB
81.82 KB JPG
>>506769
the way i went about it is creating 2 cylinders and copy 1 and drag it alone y to left

then i selected about 10 (or what i thought was good) faces to bridge with the same amount of faces on the other side.

added 3 more loops then slight shaped with with proportional editing
the insets are there for fun, you could have this shape with perfectly round cylinders
>>
>>506771
that looks like shit
why are you on this site trying to give tips
>>
>>506776
what? i can make an hyper realistic snes controller from this shape alone, you are literally dumb
>>
>>506779
do it, id like to see that from someone who managed to fuck up a cylinder
>>
File: snes.jpg (64.41 KB, 1146x432)
64.41 KB
64.41 KB JPG
>>506782
100% you are autodesk user, because you can't recognize the feature that did it
>>
>>506779
>literally dumb
>an hyper realistic

Just stop you autist.
>>
>>506786
don't you have any advice to give to him? or its xmas shitposting special
>>
File: fgt.png (102.51 KB, 577x690)
102.51 KB
102.51 KB PNG
>>506784
still looks like shit

heres the shape OP was looking to get
>>
>>506789
that's exactly the same thing i did (the third)
are you fucking dumb
why you even bother arguing
>>
>>506790
ill still wait for you to make a hyper realistic controller tho
>>
File: NotAComplexShape.jpg (509.47 KB, 1727x756)
509.47 KB
509.47 KB JPG
>>506784
>100% you are autodesk user

oh shit you really are Autistic

This isn't a complex shape.
>>
>>506791
you literally copied my method only to make a slightly better example of it and claim its a completely different.
only reason why i didn't post a snes shape is to show op the required method he should make in order to achieve it, he didn't ask me to model a snes controller which i didn't
>>
>>506790
>>506784

This is the same as >>506789
?

you are completely fucking retarded.
>>
>>506794
two cylinders into edge bringe into inset
yes its 99% the same except i didn't baby him into the perfect controller shape
please resort to /b/ or other containment zone you are from
>>
>>506793
>literally copied my method
There's that word 'literally' again. The telltale sign of an illiterate 12 year old who doesn't realise pointless hyperbole doesn't emphasise their point.

>copied my method
Yours doesn't look anything like that. you just bridged the sides of 2 cylinders.

>my method
Your method resulted in pure shit. you don't own any method of 3D modelling.
>>
>>506795
>demonstrating where supporting loops and key edgeflow should go is "babying".This coming from the guy who claims he'll turn that disproportionate piece of shit into a hyper real snes controller.
>>
>>506797
what are you even talking about? op wanted a method on how to create a snes controller so i gave him a method to use.
which by the way >>506792
>>506789
used the same method that i did the only difference is the created poles to prevent shading errors (on flat surface mind you), im really being mindfucked by idiots right now only because i wanted to think for himself on how to option shape from given method
>>
>>506798
>used the same method that i did the only difference is the created poles to prevent shading errors

>what are you even talking about?

Ok i'm out. Third time you've proven you're a LITERAL illiterate. It's like conversing with a wall.
>>
>>506801
yes you used the same method? tell me exactly why you did different
>>
>>506797
>supporting wire
wow the mona lisa of modeling, you closed that ngon like a champ
>>
>>506802
it was me that did >>506789
hope u know there are like 3 people that think you're retarded, no one stole your method
>>
File: snes.jpg (68.19 KB, 1146x449)
68.19 KB
68.19 KB JPG
>>506804
here you go, i was right so there is no need to sperg now
go to sleep
>>
>>506805
still shit lmao
where are the supporting edges, where is the extrusion, where is the turbosmooth or w/e inferior smoothing blender uses
>>
>>506807
>using bevel when autosmooth does it for you
you need economy lesson my friend
>>
>>506808
babys first 3d
>>
>>506808
>using automated full mesh smoothing operations without having supporting loops or pre existing bevels

Just another item to the list of evidence that you're autistic.
>>
>>506810
in this particular case you could have doubled the polycount
>>
>>506756
The only way to boolean without fucking up the topology is to boolean two objects whose topology perfectly aligns.

But if you're the kinda lazy shit who uses boolean then you're not going to do that.
>>
File: images.jpg (6.16 KB, 259x194)
6.16 KB
6.16 KB JPG
>>506813
would boolean be feasible to do something like this

or you should sculpt and remesh
>>
>>506813
that isn't completely true. For instance, if some verts are close enough after the bool you can just select them and merge by radius. But when this is the case it isn't usually mechanic type modelling.
>>
>>506814

Nah, that's nothing but math and chamfering the vertices.
>>
>>506814

just doo it
>>
all you anons don't know the true power of floating geometry
try it sometime
>>
>>506898
This anon doesn't know the power of intersecting geometry jaggies. If you want something to look good, then parts that are actually connected in real life, should be connected on your mesh so the shading looks proper.
>>
>>506756

You're making too much work for yourself. Keep the complexity of the two objects roughly similar. Your shape has way more faces than the cylinder you're cutting it out from, which will make cleanup close to impossible.

Either use a high poly shape to cut a high poly object and then manually chamfer the edges if necessary, or use a low poly shape on a low poly object, clean up the topology and subdivide the whole thing.
>>
>>506756
>>boolean
>>without fucking the topology up

lol
>>
>>507206
Agreed. What I do is use as many parametric modifiers as possible. Smooth, Shell... Max 2016 has QuadChamfer now. You can put a Smooth modifier below it, with the desired angle treshold, then in QuadChamfer you can just choose "chamfer all unsmoothed edges".

This is actually three settings:
1. All Edges
2. Chamfer only Unsmoothed
3. Min/Max angle checkboxes disabled.

As for booleans, no clean topology can come out of this business. Instead, make excellent operators, ProBoolean them however you like, and use NO mesh-dependent modifiers above Pro Boolean, such as Edit Poly, or UVW Unwrap. Instead, when you want to edit something, go down in ProBoolean's own stack; and if it's awkward to edit from the stack, pick an operand and extract it (as an instance). Isolate, edit, ProBoolean magically updated itself while you worked.

Only parametric stuff should then go above ProBoolean, such as Smooth, Material, Normal UVW Map, Bend, etc etc.

Just an observation. If you plan to put much turbosmooth on it, then it looks to me like you arent modelling for performance/gaming. In which case, why are you worried about the topology? Just worry about the shape - and how easily you can get there, optionally edit it. Thats why I proposed this ProBoolean approach.
>>
>>507206
But seriously, a boolean operation is a creative decision, and generally speaking computers should not make them. Do it by hand. Also, to save youself trouble, don't do it at such complex geometry. Do it at lower res and save yourself the anguish.
>>
>>506898
you cant have floating holes nigggaaaaaaa
>>
>>507394
You float the face that you want to boolean, saving you having to fix the geometry across the whole mesh.



Delete Post: [File Only] Style:
[Disable Mobile View / Use Desktop Site]

[Enable Mobile View / Use Mobile Site]

All trademarks and copyrights on this page are owned by their respective parties. Images uploaded are the responsibility of the Poster. Comments are owned by the Poster.