MODELLING and SCULPTING
Blender is a professional, free and open-source 3D computer graphics softwaretoolset used for creating animated films, visual effects, art, 3D printed models, interactive 3D applications and video games. Blender’s features include 3D modeling, UV unwrapping, texturing, raster graphics editing, rigging and skinning, fluid and smoke simulation, particle simulation, soft body simulation, sculpting, animating, match moving, camera tracking, rendering, motion graphics, video editing and compositing. It further features an integrated game engine.
Bryce is a 3D modeling, rendering and animation program specializing in fractal landscapes. The name is taken from Bryce Canyon—a rugged region with many of the same landscapes that were first simulated with the software.
Carrara is a full-featured 3D computer graphics application featuring figure posing and editing, as well as nature modeling, in addition to traditional modeling, animation, texturing and rendering. The software is also capable of dynamic hair and fur simulations, particle effects, soft body and rigid body dynamics. Carrara is now owned and developed by DAZ 3D. Carrara is compatible with Poser and DAZ format 3D figures and props. It is further supported by a number of 3rd party plug-ins and add-ons.
Daz Productions, Inc., commonly known as Daz 3D, is a 3D content and software company specializing in providing rigged 3D human models, associated accessory content and software to the prosumer market. It was originally part of Zygote Media Group, a general purpose, application-agnostic 3D content broker, and split off as “Digital Art Zone” in 2000 to focus on supplying content for the Poser market.
Hexagon is a subdivision modeler owned by DAZ 3D. It was originally developed and published by Eovia and was acquired shortly before the release of version 2.0 by DAZ in 2006. The software drew heavily on Eovia’s other modeler, Amapi(it shared the same developers), though it omitted the NURBs and precision measuring tools. The main focus is Subdivision modeling but it includes Spline tools and surface tools. Because of the two omissions, it is not as well suited to product design as Amapi, but is aimed more at illustrative and character modeling with an eye to making it accessible for those new to working in 3D.
LightWave 3D is a 3D computer graphics software developed by NewTek. It has been used in film, television, motion graphics, digital matte painting, visual effects, video games development, product design, architectural visualizations, virtual production, music videos, pre-visualizations and advertising.
Makehuman is an open source 3D computer graphics software middleware designed for the prototyping of photo realistic humanoids. It is developed by a community of programmers, artists, and academics interested in 3D modeling of characters.
Maya /ˈmɑːjə/, commonly shortened to Maya, is a 3D computer graphics software that runs on Windows, macOS and Linux, originally developed by Alias Systems Corporation (formerly Alias|Wavefront) and currently owned and developed by Autodesk, Inc. It is used to create interactive 3D applications, including video games, animated film, TV series, or visual effects.
Maya LT modelling and animation software is built for professional indie game makers. Bring creations to life with powerful game development tools.
MeshLab is an advanced 3D mesh processing software system that is oriented to the management and processing of unstructured large meshes and provides a set of tools for editing, cleaning, healing, inspecting, rendering, and converting these kinds of meshes. MeshLab is free and open-source software, subject to the requirements of the GNU General Public License (GPL), version 2 or later, and is used as both a complete package and a library powering other software. It is well known in the more technical fields of 3D development and data handling.
Modo (stylized as MODO, originally modo) is a polygon and subdivision surfacemodeling, sculpting, 3D painting, animation and rendering package developed by Luxology, LLC, which is now merged with and known as Foundry. The program incorporates features such as n-gons and edge weighting, and runs on Microsoft Windows, Linux and macOS platforms.
Modo indie packs powerful, industry-leading 3D modeling, sculpting, and animation tools into an affordable package that’s perfect for developers and freelance game artists.
OpenSubdiv is a set of open source libraries that implement high performance subdivision surface (subdiv) evaluation on massively parallel CPU and GPU architectures. This code path is optimized for drawing deforming surfaces with static topology at interactive framerates.
Poser is a 3D computer graphics program optimized for 3D modeling of human figures. The program has gained popularity due to allowing beginners to produce basic animations and digital images, and the extensive availability of third-party digital models. Originally designed for illustrators.
Rhinoceros (typically abbreviated Rhino, or Rhino3D) is a commercial 3D computer graphics and computer-aided design (CAD) application software developed by Robert McNeel & Associates; an American, privately held, employee-owned company, that was founded in 1980. Rhinoceros geometry is based on the NURBS mathematical model, which focuses on producing mathematically precise representation of curves and freeform surfaces in computer graphics (as opposed to polygon mesh-based applications).
Silo is a polygon/subdivision surfaces 3D modeling application created by Nevercenter. It has a focus on quick editing, a customizable interface (all mouse buttons and keyboard shortcuts can be assigned to any function), and a flexible workflow.
3D-Coat is a commercial digital sculpting program from Pilgway designed to create free-form organic and hard surfaced 3D models from scratch, with tools which enable users to sculpt, add polygonal topology (automatically or manually), create UV maps (automatically or manually), texture the resulting models with natural painting tools, and render static images or animated “turntable” movies.
TopoGun is a stand-alone resurfacing, and maps baking application. The resurfacing functions in TopoGun will help you modify and/or recreate the edgeflow of your digital 3D models. The maps baking functions, will help you bake various types of texture maps from your high resolution 3D models and then allow you to apply them to your newly created optimized meshes. These texture maps contain information that will help you recover the appearance and features of the original high resolution mesh.
Transpose Master is designed to help you pose a model that is composed of SubTools. It works by creating a low resolution combined mesh of all the SubTools which you can pose. The pose can then be transferred back to the original model. The Transpose Master sub-palette is in the Zplugin palette.
Vue is a 3D scenery generator software package. It is used for the creation, animation, and rendering of natural 3D environments, in particular outdoor landscapes. It is used by visual effects studios for this purpose. For example, Industrial Light & Magic used it to make backgrounds for the movies Indiana Jones and the Crystal Skull and Pirates of the Caribbean: Dead Man’s Chest, and DreamWorks Animation used it in the 2008 film Kung Fu Panda.
Wings 3D is a free and open source subdivision modeler inspired by Nendo and Mirai from Izware. Wings 3D is named after the winged-edge data structure it uses internally to store coordinate and adjacency data, and is commonly referred to by its users simply as Wings.
WorldBuilder Pro a stand-alone scene builder and renderer for laying out and rendering 3D outdoor environments. WorldBuilder is still one of the only commercial products that allows users to render scenes that could be millions of polygons.
ZBrush is a digital sculpting tool that combines 3D/2.5D modeling, texturing and painting. It uses a proprietary “pixol” technology (see below) which stores lighting, color, material, and depth information for all objects on the screen. The main difference between ZBrush and more traditional modeling packages is that it is more akin to sculpting.
RETOPOLOGIZING PROGRAMS: 3D Coat, Topogun, MeshLab which is open source
- Mentor Insight Jonathan Rush, Lead Character artist at Battlecry Studios
- CHARACTER MODELING IN ANIMATION, WITH SHANNON THOMAS
- ARTIST INSIGHT: CALEB NEFZEN (DAVID MOLINA)
Modelling is the first stage of the Production pipeline, as models are required for texturing, shading, lighting, rigging, and animation. Some modellers specialise in Characters & organic models, others in environments, others in Hard Surface models. Modellers are responsible for taking the concept art from 2D to 3D. Along
with highly trained skills in modelling techniques, the usage of modelling tools & software, a Modeller must also have a solid comprehension of solid meshes, correct
geometry, volume/weight, good vertex distribution for lighting & rigging, how light may react to different shapes, topology termination techniques, and especially good edge flow.
The Subdivision Modelling section covers the basic theory of how. However, the Techniques section does explain how Normals interact, followed with how that’s the same as smoothing without a 0 degree border-line, with images to demonstrate. It also explains why Bevels will not work for your situation, and why you really have to manually insert edge loops or use similar tools to the same result.
Researchers have demonstrated that it’s possible for machine learning algorithms to capture the “persona” and create a digital model of a well-photographed person like Tom Hanks from the vast number of images of them available on the Internet.
PIXAR IN A BOX – collaboration between Pixar Animation Studios and Khan Academy. Sponsored by Disney.
Raul Ramos is an Oregon-based illustrator and creature designer currently working in the gaming industry. He has a passion for all aspects of art and art making with strong interests in biology, paleontology, and comparative anatomy.. all of which have given him a great appreciation for paleo art, wildlife illustration and creature design. Below you will find one of the most thorough articles we’ve featured yet, which will give you an in-depth look for how to sculpt an anatomically-correct dinosaur, as well as inspire new techniques and methods for any and all creature designs.
Zbrush Creative Techniques – Twitch TV, frameworkVFX
My aim for this Channel is to provide a resource of quick digital creative Demo’s
– Tips & Tricks related to facial modeling for Animated Productions
TRANSPARENCY and ZBrushCentral The thread OP talks about putting each object on a layer and turning up the transparency in the Mat – then with 4R6 you need to uncheck flatten layers in the render drop down.
Introduction to MODO 701
Zbrush Polypaint importer for Maya / Free + Source code
Official ZBrush Summit 2016 Presentation – Laika
Blender 3d / Zbrush: Asset Massive Library and Kitbashing using Prebuilt Parts
Edge of Tomorrow – Creating the Mimics Shot Build
Creature Modeling & Sculpting Techniques with Autodesk® Mudbox®
Meet The Experts: Freedom from UV’s, Ptex Painting in Mudbox 2014
MODELLING PROCESS DIARY
3D Modelling is based on the Cartesian coordinating system in three dimensions, a point in 3D space defined by width, height and depth.
The zero point for each of these axes is the origin and the intersection of all three axes, providing the starting point for calculations and defining a point in 3D space or World Coordinates Position, (0,0,0). This defines the World Space, origin and the position properties indicate how far away the object is from the origin.
Each object is also defined by its own object space or local space. If the geometric centre of an object is not centred at the origin and is centred at its own local space even when placed at origin it will be off-centered from world space.
Maya uses the right-handed Cartesian coordinate system, where the X axis points left to right, the Y axis is up and down and Z points forward and backwards.
Modelling uses many techniques including polygonal modelling, NURBS and subdivision surfaces or SubD. They have pivot point which is where objects rotate and scale from and is the placement point for the X, Y and Z coordinates. It can be the object or another node’s pivot point within a network of nodes and hierarchies.
Maya uses nodes to carry attributes such as coordinates, geometric descriptions, colour values. The transform node or DAG node (direct acyclic graph) has the values for translation (position), rotation and scale and also hold hierarchy information about any other child or parent nodes that are attached. Another node defines the shape, creation node or shape nodes, how the object was created with attributes such as radius and the shape nodes are the children of the transform node.
Topology: whatever type of geometry you use it will either be created by NURBS, or points, edges, and faces. The way these components are connected together and the flow around the 3D object is the topology. You can think of topology as the type of polygon faces, the type of vertices and the flow of the edges.
Retopologising to create an animatable mesh. To create a mesh which is suitable for deformation, so all the edge loops will be carefully considered in pursuit of this.
When scaling objects consider the difference between using the scale tool or changing the radius to adjust the size.
Rotation, Heading, Pitch and Bank.
Cinema 4D: converting primitive state, paramedic object, into polygonal object, moving from mathematically defined geometry to object that user can define, such as repositioning the axis. Make Editable. Enable Axis Modification. Scale Document function can quickly adjust the scale without messing up the scale settings in the coordinates tab. The Scale Project function works on everything in the document, can isolate objects by copying and pasting them int a new empty file, make the changes and copy and past them back. Edit > Current Scale > Target Scale.
NODES hold specific information and any actions associated with that information such as a surface, light or camera material.
The components of polygon surfaces are made up of faces that share three or more edges and vertices or points along the edges which are used to create hard-surface models. A mesh is is a collection of vertices, edges and faces that defines the shape. It is a closed plane figure, enclosed by lines that form many angles with the lines creating edges. A polygon mesh is the surface that is composed of points or vertices that form an edge that shares at least two adjacent polygons.
A polygon mesh’s orientation is defined by its normal vector, normal which tells the computer whether the polygon is facing outward or inward.
When three polygons share the same edge a non-manifold surface is created which can lead to rendering and animation problems, having a configuration that cannot be unfolded into a continuous flat piece. There is also a problem where two polygons share a single vertex but not a complete edge.
Polygon Faces have two surfaces with normals determining a surface’s orientation affecting such things as visibility, rendering, dynamics, lighting and shading.
Splines are lines or curves with control points (control vertices) for modification which can either approximate a curve or interpolate it being influenced by adding more control points or vertices, spline curve weighting. Linear, Cardinal, Bezier, Hermite, B-spline (basic spline), Nurbs (Nonuniform rational B-spline) and spline patch.
MODELS for the INTEGRATED PROJECT
Design Visual Thinking
ROSE the APRON
I started by thinking about the box size, do I include the head or start with the basic body. Where will I need the edges, the vertices, the faces and how many subdivisions will be required to create the shape? I was planning this for the arms, legs and character detail at the beginning and not approaching it from a basic cube and building from there. Starting with all of that and forgetting to “keep it simple for as long as possible”.
By doing this I found it difficult to control the components of the model. Adding more and more was not the solution, the edges were not connecting, I was not working out how to create the quad faces and how to create the character’s shape.
After making the basic box shape, creating the band around the body then extending the arms I had an odd arm shape. This was not how extruded faces were meant to look, I could not work out why the smoothed model was this odd shape. It turned out to be a nonmanifold surface, too many faces hanging out together. I later had the same problem with the hair, working it out and improving my extrude technique then there was hair.
I could not work out where to extrude the leg from. Deleting the faces and bridging with divisions, like we did for the penguin was not working. The number of edges I needed to extrude for the legs was not lining up with the edges on the body. I went back to 4 faces and redirected edges to create a face to extrude for the leg. Still I was extruding the leg each time there was a change in the shape of the leg. Eventually I extruded once making the shape, later adding edge loops and moving and scaling to make the shape.
I made a separate box model for the foot with the same number of edges as the leg. Then combining the mesh and merging the vertices with merge vertex tool.
The first time round was more about getting it wrong and learning some basic modeling techniques and tools.
Now I understand more about having edge loops that connect and go around the shape, making it easier to add edge loops later and placing them to give more definition and sharper edges. This gave more control over the vertices including when and how they are moved to create the shape. A model can be built by both extruding faces and deleting faces then extruding the edges giving more choices and levels of control while keeping the geometry simple.
The band around the character works well with the arms extending from there. I like the face as the top part of the apron. Next time I would prefer a flatter character making it more apron like with hair ties that are not so wide and more like ties.
Starting with a box shape and four subdivisions for width and depth and two for height. When making the shape first and sloping the top it was difficult to extrude to make the rounded shape along the top rim. Going back I kept the basic box shape and extruded creating the faces and shapes for the round lip shape at the top of the basket opening. Then I shaped and moved the vertices to create the rounded shape for the top of the basket being aware that the inside shape needed to stay inside and not protrude through the outside faces.
Smoothing the model’s surface, the edges, can be done with smooth preview or converting the polygons in the Mesh menu.
Smooth Mesh Preview affects only the display of the polygon mesh in the scene view, when rendered it will appear in its original un-smoothed state in the final image, using Mental Ray the geometry appears smoothed in the render. Converting the smooth preview to a smoothed polygon mesh using Modify > Convert > Smooth Mesh Preview to Polygons or to a subdiv proxy using The Smooth operation subdivides the geometry, quadrupling the number of polygon faces in the geometry and increasing the number of vertices available for modelling unlike the smooth preview where the vertices stay the same.
Using two shapes, a box for the top and a cylinder for the handle I’m not sure what to do about the triangles at end of the cylinder shape. With the cylinder I know the top faces will be deleted when joining the two shapes. After moving the vertices to make the shape I realised that scaling the sides together makes a more consistent shape. Starting with 8 subdivisions this was not working to enable me to make and join the shapes. I redid the cylinder with 12 realising how many vertices were needed for making the shapes and later join them. Looking back I think it might have been possible with 8.
I started moving the vertices to make the shape of the front and back of the spoon and found it was not working, then using extrude for the faces I experimented and found using scale and move I was able to adjust the shape and faces to get the rounded shapes. After getting the basic shape I then added some extra edge loops around the indented, inside of the spoon to clearly define the ridge on the edge of the scoop shape.
When lining up the vertices I was unable to work out which ones belonged with each other to make the joined shape. The first attempts did not make a symmetrical shape, twisting on one side. Undoing and lining up the shapes again I needed to add two extra edge loops, one centre front and one centre side to help me line up the vertices and shape. Then combining the two meshes and merging the vertices I was able to establish a clear, symmetrical shape. After this I realised I could delete the two extra edges I had made to help me line up the shape evenly and still keep my original shape while adjusting the vertices along the join.
The corners of the original box shape have become a bit lost in the spoon shape. They are still quads and I’m not clear if what I have done was the most appropriate place and way to move the box corners. The other thing is that they have impacted on adding and adjusting edge loops.
The adventures of making a Dalek: in class we did get further with this than the image shows.
Final project sequence, click here and password is marm108
After working in the greys of the Maya interface it is exciting to see the environment start to come alive with textures, lighting, colour and the animation of the mechanical arm introducing some personality.
My Mechanical Arm Test
password : mArmTest
Starting the mechanical arm base and having a bit of a test with the base. Working out how to select faces, edges and vertex selectively and not collecting many others not required to change the shape. Select > convert selection and going inside the shape were very helpful. After starting the model I remembered how helpful it was to make it straight along only the y axis and to keep it central to the grid.
As I worked up the arm parts I was experimenting with working out how to make the shape that could be fit inside each arm block so that it could extend and retract the length of the arm. Thinking that a lattice deformer would be the way to go I soon discovered it was not very helpful and went back to component mode to make the adjustments.
The joining of two parts for the construction of the elbow was challenging. Working out the shapes with the edges and vertex being able to join and merge took several attempts.
The wrist section was a challenge in working out how to make a joint that would work both in the movements to pick up the box and the skill to model the shapes. One of the ways I explored was a ball shape like the base, shoulder joint and there was not enough range of movement and I could not work out how to attach the finger shapes. I changed the pieces of the joint and settled on a flat circular shape that would rotate and more suitable to attach the finger parts so they could move to pick something up.
While working on ideas to attach the fingers part to the wrist joint I explored a few options and came across many problems with my modeling. How do I to make the connecting shape? Which parts do I delete, how will it support the parts, how will it move and do I connect the parts making it workable and not too chunky? I become so caught up in making it work I forgot to think about other alternatives and thanks to Simon he suggested breaking it up into three sections when making three parts of one object was not working. Bless his cotton socks.
Back to the base, I was working on the legs to support the circular structure for the rotation movement of the shoulder section and I explored extrude, lattice and component options and could not get close to the shape. Again Simon had a workable solution and the soft select on the poly extrudes worked like a treat and home I went to make the shape. It took several attempts, I was not able to work out how work on the extrude to make it symmetrical. I adjusted the vertex points and need to work more on that area.
Class Mechanical Arm included exploring curves.
While making the class mechanical arm one of the concepts we developed was making a pattern. From a series of nurb shapes placed along a line and using surfaces > loft we created the initial shape that became the pattern. Had a big surprise when the shape was deformed due to choosing the nurb shapes in the wrong order.
This decorative shape for the mechanical arm changed the nurbs circle into a square with linear degrees and four sections as part of the inputs in the CB. Finally worked out that after placing the first nurbs shape on the cv curve freezing transformations sets the CB to ease the rotate of the duplicated shape on the curve.
Also managed to get the C, snap to curve happening. After doing a test loft it was interesting to see how the shape could be changed by moving the nurbs circle and adjusting the CV points to improve the shape.
Other areas we explored were deformers for shapes together with duplicate special to rotate 7 rods around the cylinder. The bend deformer has been used to make the twisting shape of the rods around the base shape. Not sure about the relationship between lattice and its base.
Discovering Image Planes, they display images for modelling, textures, reference concept drawings and images sequences for live action which may or may not be attached to a camera (cane be any type of camera), can be resized and may or may not be included in the final render. Here we are using a live action background. When including the image as part of the render be aware of the Colour Space options for Colour Management. Consider using RGBA Display Mode when the image will be rendered and used for compositing shots. Adjust Colour Gain for brightness and Colour Offset for contrast. Alpha Gain adjusts the transparency of the image, adjusting the opacity. There will always be some lens distortion from photographs.
Revisiting the ufo model and animation on live action footage. When opening the image plane and the legs are missing, possibly something to do with different versions of Maya.
– the legs were missing though the names were still in the outliner. When remaking and animating the legs, grouping not parenting has a transform node. To animate the The tool setting needed to be on object.
The lights needed to be adjusted to work with the image plane and legs then added a third light.
ufo movie on vimeo, click here : password ‘ufo108’
Discovering Modelling from the Autodesk books.
Solar System video – click here – from the book, many hours, like heaps of hours. Please to say a few years later and looking back on this, things have improved, including my wordpress skills.
Still going with the locomotive engine – using polygon modeling, CV curl tool, and the outliner grouping and using layers. Learning about to approach the whole and how elements join together to form the engine. Suggested things to model : dining room table and chiars, computer monitor with it s angles and overall surface details, desk or floor lamp, pen, computer mouse, chess set, fruit, car.
Questions – where to find some of the numbers in the CB. AE for extrude, adjust vertex. Evenly scaling and extrude on a poly cylinder.
– used poly for cylinder and cube, cv curve, extrude and scale and move to form shapes, shift pivot point to set bolts, rotate and duplicate
– set up layers in the channel box to hide groups when working on other sections
NURBS MODELLING – Non-Uniform Rational B-Splines
Nurbs modelling is suitable for models that benefit from smooth surfaces and contours, they are four-sided patches and are automatically converted to triangles at render time. Tessellation, how the surface is covered to triangles, settings can be changed at any time thus being able to change the resolution.
Nurbs are made up of curves, curve points, which are lines defined by points with movement defined by its U-coordinates, the location of a point along the length and the position of the point is defined by its U-parameter. The V-coordinate specifies the location of a point on the surface, U and V are perpendicular to each other.
Curves can be made up of multiple spans with edit points marking the connections between two spans. NURBS curves and surfaces are controlled with the points or control vertex (CV), the CV’s are handles used to edit the curve’s shapes. Hulls are straight lines that connect the CVs, a visual guide and groups of CVs that follow the curves that define a surface. Parameterisation is the way the points along the curve are numbered, Uniform or Chord Length.
Nurbs surfaces can be edited by moving the position of the CVs or hulls of a surface. There are seams where the end of the surface meets the beginning.
CV curve default is will a degree of 3 creating smooth, rounded curves and needs a minimum of four CVs for the curve.
Starting with NURB modelling they can then be converted to polygons for further modelling. The tessellation methods for converting NURBS to polygons can be Control Points, General which is based on the number of U and V values or the number of contiguous edges that will wrap around your model, Count where I can enter a target value for the number of polygons and Standard Fit which has a criterion for subdividing the model.
A Boolean operation creates a new surface from two surfaces to create a new surface and works best when the geometry is kept simple. Union which adds two surfaces, Difference subtracts two surfaces Intersection which overlaps parts of two surfaces.
Booleans can leave n-sided polygons and overlapping edges which gives us bad topology.
Modeling a Nurbs Pump, from Autodesk Maya book.
On the duplicate cylinder the pivot point was different for the three objects – the cylinder and the two ends. When I scaled the second cylinder the book had the rescale for x and y changing, leaving z and this distorted the cylinder and shortened the length. I needed it to stay as a cylinder, the same length and reduce the diameter, I changed y and z and left x.
It was interesting to choose the hull in the marking menu to close the ends of the caps instead of attempting to select the cv’s. I needed to redo one cap the lack of isoparms changed how it shaped.
When it came to lofting the two cylinders together I had them too far apart and could not work out how to move one cylinder closer to the other. When I selected the whole group or made it a group it did not move as a group. At this point I had connected the caps to the cylinders by using loft>attach surfaces>detach surfaces. Not sure I was clear on which isoparm needed to be selected to detach surfaces. When attaching surfaces one was going to the wrong place so I did another loft, attached and detached the surfaces.
Chugging along then the last attachments of the top cylinder to the lofted surfaces, the gaps did not match. I had cut too much from one end that needed to be attached. Also not sure how the cv’s needed to match up. There were some gaps in the model and some of the edges had a triangular gap.
SUBDIVISION SURFACES for MODELLING
With subD’s the mesh can be subdivided to add detail only where it is needed. Always create UV texture before converting polygons to subDs. Keep the polygon mesh as simple aspkossible, can convert three-sided or n-sided polygons though will get better results with four-sided polygons and non manifold geometry is not supported. Subdivision surfaces are largely unsupported in pipelines outside of Maya.