Re: Do you think polycount a dying platform?

jStins said:

Welcome to greentooth.ai please upload your image for critique!

*uploads*

thinking...
thinking longer for a better critique...

greentooth.ai: it's shit

Finally Polycount's going back to the 2007 era.

MattQ86

3 ·

Re: Do you think polycount a dying platform?

I definitely hope not! I only joined a short while ago, and this is one of the more active forums I’ve found and I really like it here.

Sure, Discord is great, but it’s more of a chat environment. It’s not the best place to post WIP projects or give detailed feedback, the format just isn’t ideal for that.

Here I’ve found that people give and receive thoughtful, constructive critique on their work, which I really appreciate.

Vastra

3 ·

Re: Do you think polycount a dying platform?

Not the same... true.. but I still know my password by hard.

I feel not like posting Art that much here anymore bc there is no real interaction going on.
On the other hand I'm probably making more art than back then.
Still got my current job through pc 5 years ago - I'd say the page still worth it!

I'm more active on modding discords nowadays though - bc the kids like to hang around there..

At least some of them put info into actual wikis which can be reached outside of discord's blackhole

rollin

3 ·

Re: Do you think polycount a dying platform?

These days most artists want to post their work in places that'll help them gain exposure (for their portfolio/game/etc.), but Polycount is only really focused on helping artists with critiques and technical questions. Artstation took over for showing portfolios (after CGHub died), and generic social media sites make more sense for advertising commercial projects.

If your an artist that likes to learn and improve then this place can be great, especially if you like the sense of community such a site can convey, but for most younger artists they'd rather just watch youtube tutorials and ask AI when they have a question.

PolyHertz

3 ·

Re: Do you think polycount a dying platform?

I am so happy that Polycount still exists and keeps the old-school raw interaction alive. After many years, I've canceled all my social media accounts (because of the noise of stupid people, AI slop, politics, Blender influencers, bots...) and found a place here. According to the current situation on the internet, I think Polycount has a very bright future.

ondrejmalota

3 ·

Re: Having doubts about taking a first job landing

you should delete their email and block them. That ain't an opportunity nor a foot in the door. That's literally modern day slavery bro. Make more art, post more art, get actual, real opportunities

zetheros

3 ·

Re: [WIP] Miss Peregrine - Eva Green Likeness (Realtime)

Legends speak of a man who was recently part of a studio closure. He returned to a 4 year old WIP thread in the ancient temple where the the greatest of the 3D wizards honed their craft for years. He hoped once more to draw upon its magical essence and wield it to complete long unfinished tasks.

Image: https://us.v-cdn.net/5021068/uploads/editor/yq/zusm28mvckgo.png

RabidRabit

4 ·

Re: Show your hand painted stuff, pls!

A small hand painted project done for practice.
Concept: Betty Jiang
Model & Texture: Alfredo Santos

Image: https://us.v-cdn.net/5021068/uploads/editor/rk/365m99daprci.jpg

frodin

5 ·

Re: Show your hand painted stuff, pls!

Recently finished this as a bit of practice for meeting deadlines, and just finishing a project without noodling too much. I didn't quite meet my first goal, but I did manage to finish this significantly faster than other, similar projects. Also my first actual post here - so hopefully more where this came from!

Image: https://us.v-cdn.net/5021068/uploads/editor/cz/w8omjcucyhte.jpg

Tom_Axler

4 ·

Re: Sketchbook: Frank Polygon

Subdivision sketch: Hemispherical helmet.

This is a brief look at a few basic strategies for routing topology around cut outs and panel lines on spherical [and ellipsoid] surfaces.

Note: there's several different ways to structure spherical geometry. Since this write-up is specific to basic subdivision modeling only basic spheres and quad spheres are mentioned.

Basic spheres, with one vertex at each pole, provide a lot of control over segment spacing but also tend to produce smoothing artifacts near the caps because all of the edges there converge at a single point. While there are a few ways to reduce or resolve the smoothing artifacts on basic spheres, quad spheres have eight evenly distributed N poles and a fairly even quad grid topology that often make it a better choice for subdivision modeling.

Quad spheres subdivide cleanly but there are still some downsides. The layout of the quad grid is fairly inflexible and most tools generate quad sphere primitives with 24 faces then increase the mesh density by subdividing. Which means the density of the mesh increases four times for every subdivision level applied. The lack of fine control over the number of segments and rings means the density will often be a bit less or bit more than what's optimal.

When modeling with quad spheres it's important to orient both the grid and poles so the topology flow matches the larger primary forms and provides support for the smaller secondary forms and tertiary details. Especially when these shapes intersect the surface of the sphere.

The example below shows how simply rotating the larger quad sphere changes the relative topology of the underlying surface. Going from a linear square grid that runs vertically and horizontally to one that runs diagonally and forms a diamond pattern. Simply rotating the underlying geometry so it aligns with the intersecting shapes can prevent a lot of headaches caused by messy topology and unnecessary manual clean up work.

It's also possible to combine the different positive aspects of both basic spheres and quad spheres in a way that avoids the inflexibility of the quad sphere and resolves the smoothing artifacts on the caps of the basic sphere.

Here's an example of how simply rotating the basic sphere so the poles are hidden beneath the intersecting shape allows it to be capped with an ellipsoid made from a quad sphere.

When modeling something like this helmet only part of the sphere is needed and there's often cut outs or protrusions around the ears. So use those natural breaks or changes in the shapes to redirect the loop flow.

It's also helpful to break complex objects down and model the parts individually. Since this allows for localized differences in mesh density and topology flow. If it doesn't need to be connected in a watertight mesh then model it separately.

Sometimes disrupting the segment spacing and topology flow is unavoidable when cutting in panel lines. While it's possible to manually compensate for this by adjusting the relative height of the new edges between the segments of the curve it often makes more sense to just shrink wrap everything onto a clean guide shape.

The example below shows how the basic shape acts as a guide and most of the panel lines follow the existing geometry but where ever they cross over diagonally the geometry is re-projected onto the clean shape below it to remove any unwanted deformation and maintain the desired surface quality.

Once the basic loop paths are established a manual bevel / chamfer operation is used to create the geometry for the panel lines and the edges in the center of those loop paths are moved below the surface to create depth.

Here's what the final high poly looks like. The editable base mesh is fairly simple, all of the support loops are added around the panel lines with a bevel / chamfer modifier and the entire surface of the helmet follows the underlying guide shape using a shrinkwrap modifier. This keeps everything in the final high poly nice and crisp but still easy to work with.

There's different ways to approach adding more complex shapes and cleaning up the topology. So it just comes down to applying the same basic principles through a bit of trial and error. Even with this less than ideal topology layout the shading is more than acceptable because the surface conforms to the guide shape.

Highly reflective surfaces have to be just right or visible surface quality artifacts will start to appear. Using a guide shape and shrink wrap modifier makes finding minimum viable a lot quicker and it's easier to clean up the topology without having to constantly re-work large sections of the mesh to maintain the accuracy of the shapes.

Image: https://us.v-cdn.net/5021068/uploads/editor/6u/j6aia2c3sxva.jpg

Recap: Choose a sphere's topology layout based on how it fits the forms and rotate the geometry so it lines up with any intersecting shapes. Simplify a complex model by creating it from individual pieces with varying geometry densities. Try to maintain consistent segment spacing between the edges that make up curved surfaces but whenever that's not possible manually compensate for unwanted deformation and consider shrink wrapping everything to a clean guide shape to preserve the surface quality.

Below are a few links to other write-ups that cover various topology issues on hemispherical shapes.

Panelizing:
https://polycount.com/discussion/comment/2769890#Comment_2769890

https://polycount.com/discussion/comment/2775511/#Comment_2775511

https://polycount.com/discussion/comment/2777813#Comment_2777813
https://polycount.com/discussion/comment/2777595#Comment_2777595

Basic helmet shapes:
https://polycount.com/discussion/comment/2771703/#Comment_2771703
https://polycount.com/discussion/comment/2806840/#Comment_2806840

FrankPolygon

5 ·

Best Of

Subdivision sketch: Hemispherical helmet.