/3/ - 3DCG

Can it subdivide well?

>>440656

Yeah It subdivides

I'm just wondering if it's good practice or not to use triangle polys with an extra vertice like that. They're not concave so it should be good right?

>>440655
If you resplit it like this it will use less polys, but the tris could screw up subdivisions, depends on what you are doing with it.

>>440655
You don't need to subdivide or deform stairs in most cases, so you don't need to worry about topo too much. Just apply the bevels of the size/shape you need and you're done, no smooth modifier.

>>440655
wastefull to have the bottom face of the stairs segmented since that gives nothing to the shape. The built in max stairs look like this if one forces the triangulation to show and it's what I'd recommend if you made a stair like this for a game engine.

>>440694
An even better solution might be to leave the gap disconnected so the area highlighted in red is really a hole and have the side of the stair consist of just 2 polys.

>>440664
>>440694


Not OP, but wouldn't these, especially that second one, be a nightmare to UV unwrap?

>>440700
planar unwrap would take care of that, still that looks like shit and It's not worth a few less triangles.

>>440700
Planar map the sides, and do an Unfold on the top part. Would take no more than a few minutes.

>>440694
You could save even more polys there if you'd make the bottom tris one big quad with no connection to the tris above.

>>440732
This is the best answer. Never waste time by building an unwrappable horror just to save a couple of vertices.

>>440735
>thinks any of the models in this thread are an "unwrappable horror".
>literally takes two steps to unwrap any of them perfectly
>thinks splitting the mesh like that is good modeling practice

topkek

>>440737
Give me one good reason not to split it up like that. It unwraps fine, the normal map doesn't render any errors, and it saves polygons.
>inb4 those stairs need to be animated

>>440745
It unwraps fine, but so do the other ones without any extra effort.
It can result in many different shading errors depending on the renderer you're using, it's an open seam where light can leak through.
It's also less organized by having such a simply object be split into parts like that.

You're saving so little polygons, there is absolutely no reason to be bothering with that.

OP didn't even state if this is for a game ready model, or production, or archvis. Both CE3 and UE4 would have a bitchfit dealing with lighting on that, as it would absolutely leak through the split-seam.

>>440747
>it's an open seam where light can leak through

This would be true if it was two separate models aligned side by side but since it's the same mesh it should never render incorrectly,
unless you travel so far off origin that the floating point error will cause distortions, but if your game hinges on that you're treading thin ice.

>You're saving so little polygons, there is absolutely no reason to be bothering with that.

It's insignificant for a single stair yes, but if your scene uses lots of them every polygon becomes interesting since it'll multiply the tri count.
Like a single model with a ~100 extra triangles, who cares? But hundred of these in the scene gives 10K extra tris, now that's something we'll start caring about.

>>440732
Except that's not a quad because of the verts from the stairs, it's an ngon that is gonna shade like shit.

>>440838
While I don't agree with his method, you are wrong there. He said the bottom portion wouldn't be connected to the top, thus it would just be a quad, because those triangle points aren't physically connected to he shape below, their points are merely sitting along it's edge.

>>440840
Yeah I missed that. Still would have lighting issues tho due to light leaking as mentioned above.

>>440812

>It can result in many different shading errors depending on the renderer you're using, it's an open seam where light can leak through.
>depending on what render you're using

>>441167
Fool, can you even read? The sentence you quoted says "renderer", not "render". I'm havin a serious giggle m8.

>>441172

>when you have nothing else meaningful to say about the matter so you just resort to pointing out mistakes

>>441174
What the hell are you talking about? The comment I'm replying to wasn't making any statement or asking a question, if it was, then it was highly autistic and made absolutely no sense, so please clarify.

>>441177

>>441178
>It can result in many different shading errors depending on the renderer you're using, it's an open seam where light can leak through.
>depending on what render you're using
Indeed, never go full retard.

>>441179
...are you that anon who even got b& from cgtalk?

>>441179

dude, it's just one comment. Don't let it get to your head.

>>441180
The hell are you talking about?
Dude replied with "depending on what render you're using", as if that makes any sense. A render is a result of a renderer, renderer is it's own word, which he doesn't seem to know.

>>441189

autism_speaks.jpg

>>441189
You gotta admit, being so butthurt about something that was probably a typing mistake, that's pretty autistic.

>>441193
The thing is though, even if it was a typing mistake, his comment still makes nooo fucking sense. And I'm not butthurt, I'm merely stunned by his stupidity.

>>441198

I'm stunned by your inability to let something go.

If you were so taken aback by utter retardation wouldn't it be better to explain the situation and prove how he's wrong and end it right there rather than go on a 10 post tirade on how stupid the guy is?

you were doing fine with >>440812 up until someone pointed out you ignored the part where it says "depending on the renderer"

>>441200
...even tmz writers arent as autistic as you

Speaking of staircases, is there any good tutorial on how to animate a character ascending and descending the staircase?
I have a hard time on this one.