Essential Guide to 3D Rendering

3D rendering is everywhere: real estate, online shopping, gaming, movies and more. Learn about the art and science of 3D visualizations from artists and experts in the field.

In this article, you’ll find:

What Does 3D Rendering Mean?

3D rendering is a computer graphics process that uses three-dimensional data and models. The goal is to create a lifelike or non-photorealistic image. 3D models are a digital file of an object created using software or through 3D scanning.

3D rendering is also a form of virtual photography. The staging and lighting of scenes are fundamental to the generation and capture of images, whether photorealistic or intentionally non-realistic.

Ben Rubey, 3D Art Lead for Marxent, explains, “3D rendering is the act of creating a 2D image from a 3D scene. Compare it to taking a photo with a camera. In 3D rendering, you take all of the 3D data and turn it into a snapshot of the scene.”

Two Types of Rendering: 3D Real-Time and 3D Post-Process Rendering

3D real-time rendering produces and analyzes images using graphics software, usually to create the illusion of motion from 20 to 120 frames per second. 3D post-processing rendering is done on a 3D render once it reaches a stage acceptable to the artist. Post-processing fixes minor errors and adds details for additional realism, usually with editing software.

3D Rendering vs. 3D Visualization

3D visualization is the system of multidisciplinary procedures that create a convincing image that looks like it exists in a real space, from concept to the final representation. 3D rendering is one of the final steps in 3D visualization.

3D Modeling vs. 3D Rendering

3D modeling is the process of developing a mathematical representation of an object or surface as it would appear by width, breadth, and depth dimensions. 3D rendering transforms 3D modeling into high-quality, detailed, and lifelike images.

3D modeling and 3D rendering are two separate steps within computer-generated imagery (CGI) creation. 3D modeling precedes 3D rendering in the 3D visualization process, and modeling services are often purchased. Learn more about the process and how to outsource modeling services.

What Is 3D Product Rendering?

3D product rendering generates 2D images from models. 3D product renderings create photorealistic images that show how an object will look after manufacturing. The product is usually rendered to show multiple angles.

Many industries benefit from 3D product rendering before the manufacture of products. For example, 3D product renderings can help test a product’s appeal to customers before it goes to market, reveal design defects, and save development costs.

How Does 3D Rendering Work?

3D rendering is a multistep process to render a product or complete scene into a 2D representation. Rendering can take milliseconds or many days for a single image or frame with the method used for video or feature films.

Steps in the 3D Rendering Process

The 3D rendering process begins with a consultation and a resulting vision. Next, there is analysis and design, which is the basis for modeling. 3D rendering comes after, followed by refinement processes. Once the render is approved, it is delivered.

Rendering steps may differ depending on the project, the type of software used, and the desired outcomes.

Pre-Rendering Steps

Before rendering begins, consider these three steps, which are separate and foundational to the process:

Vision: Before any work begins, hold an initial consultation to understand the goals of the project: the company, its market, appearance, and intended use of the image. Based on that input, it is easier to determine what the final deliverable will be. The client or creative director then approves the vision.

Analysis and Design: With the approved vision in mind, project analysis begins, and decisions about object rendering are determined. Decide on any features it should contain in the finished product, such as color, texture, camera angles, lighting, and environment.

Modeling: 3D modeling produces a 3D digital representation of a surface or object. Using the software, the artist manipulates points in virtual pace (called vertices) to form a mesh: a collection of vertices forms an object or a solid. The solids generated are geometric shapes, usually polygons (also known as primitives). The polygons are manually or automatically generated by manipulating vertices. If the desired outcome is special effects or character animation, the digital object can be animated.

“3D modeling is about the creation of objects, like a chair,” Rubey notes. “In 3D, a chair can exist as a geometry, the shape of the object, but it is invisible until the camera captures it, renders it, and adds the materials, lighting, color, and texture.”

3D Rendering Steps

After modeling, the 3D artist begins his work to bring the scene to life. “The best way to make sense of 3D is to compare 3D objects to objects in the real world,” Rubey explains. “Let’s say I want to render a spoon that’s sitting in my kitchen. First, I need to draw or capture the shape or geometry of the spoon in 3D. Then I add the material I want: clear plastic, opaque plastic, wood, or stainless steel that either has a shiny or matte finish. Then bring in the lighting to add dimensionality. This final stage is what makes the object look real.”

“Finally, you decide the camera’s position and take the pictures. We can put a camera above, below, and facing–just as in real life. Then you can take one image or an animation that is a series of images, as is the case in cinema or film. When you take a picture in real life, the lens opens to take in the light. In 3D, it’s the same, but the computer does the mathematical calculations of the light quality and angle. The more elements, the more lights, the longer it takes to create an image.”

1. Rendering: Materials and Texture
An accurate depiction of the object’s material is essential to realism. The artist changes the material settings and appearance, like glossy plastic or matte linen, to get a realistic visual representation. Other parameters are changed, such as the surface or even the hardware used to install it.

2. Rendering: Lighting
Light is everything, according to Rubey. “A good lighting person in 3D understands the physics of light and reflection. Lighting creates shadows; shadows make objects appear real. Without convincing lighting, the products look fake and unnatural. People don’t necessarily understand why they think something looks fake, but it largely has to do with a lack of realistic lighting, reflections, and shadows.”

3. Rendering: Details
After texturing and lighting, the 3D artist will continue to sculpt and add details to complete the concept, whether the goal is to make the form as close to lifelike as possible.

4. Rendering: Feedback and Refinement
The client or art director’s feedback is collected to make any refinements or changes. The artist incorporates the input, makes any changes, and submits the image for final approval.

5. Delivery
The final image is provided to the client or stored for use in a more extensive image sequence. The resolution and format of the pictures depend on the ultimate use: print, web, video, or film.

The 3D Rendering Artist

3D rendering artists are unique craftspeople since they are both creative and appreciative of technology. Many 3D artists have experience in the arts or industrial design and convert their skills into digital form. In industrial design, 2D markers and created highlights to create products like cars, which is also called rendering.

Julian de PumaJulian de Puma is a fine artist and a 3D artist with more than 25 years of gaming and engineering visualization experience. de Puma says that flexibility was an advantage in the past but that employers today often look for specialists. “For example, mechanical and industrial design clients use software tools that are high-end and more expensive because they require precision. Those artists tend to have more of an engineering brain. Organic rendering is more like working in clay, creating dragons, monsters, people, soft stuff, and more akin to traditional drawing or painting. I can do both but prefer the more organic subjects.”

No matter what type of work the 3D artist does, continuous learning of new programs is part of today’s profession. While technologies are coming at an ever-faster pace, de Puma notes, “They do tend to make things easier and faster, which is a good thing.”

Different 3D Rendering Techniques

Photorealism, or the illusion of reality in non-realistic images, is one of the main goals in 3D rendering. Most techniques focus on creating believable perspective, lighting, and detail.

Types of 3D Rendering

  • Real-Time or Interactive Rendering: Real-time rendering is used primarily in interactive and gaming graphics, where images process from 3D information at high speed. The dedicated graphics hardware has improved the performance of real-time rendering, ensuring rapid image processing. “The best example of real-time rendering is a video game,” Rubey explains. “It’s happening right now with renders moving at 60 frames per second. Marxent has a product that shows rendering in real-time: 3D Room Planner. When you want to render in high quality, the computer calculates how the natural shadow looks. It takes a few minutes to figure out the more realistic scene.”
  • Non-Real-Time or Offline Pre-Rendering: Typically used in situations where the need for processing speed is lower, this method is employed when photorealism needs to be at the highest standard possible for visual effects. Unlike real-time rendering, there is no unpredictability in the process. “A Pixar animated movie takes an hour to render a single frame,” notes Rubey.
  • Multi-Pass Rendering: This post-production process divides an image into separate layers. Each layer is tweaked to optimize the image overall. The technique adjusts color and lighting intensity to preserve details. Video games, computer-generated movies, and special effects use this technique to create more realistic scenes.
  • Multiple passes usually happen in films to improve the final image. At Marxent, we render one frame. In 3D, we separate the rendering into passes: one pass just of the shadows, one pass just of the reflections, another pass just for colors. We take these passes and put them into compositing software, layer them, and change each aspect independently of the other, making the shadows lighter or darker. Many different passes provide better results with more control – like Photoshop, but for animation.”
  • Perspective Projection: This technique makes distant objects appear smaller relative to those closer to the viewer’s eye; the software program will create perspective projections by multiplying a “dilation constant” to set objects into scenes appropriately. A dilation constant of one means no perspective, while a high dilation constant can cause image distortion or a “fisheye” effect. Orthographic projection, which views objects along parallel lines perpendicular to the drawing, is used for scientific modeling that requires precise measurement and third-dimension preservation.
  • Radiosity: This technique simulates how surfaces act as indirect light sources for other surfaces when illuminated. Radiosity produces realistic shading that mimics the way light diffuses in real-world scenes. The diffused light from a specific point on a particular surface is reflected in a broad spectrum and lights the virtually rendered space.
  • Rasterization: With this technique, the “classic” for 3D rendering, objects generate from a mesh of polygons or virtual triangles or polygons to create 3D models. In this virtual mesh, the corners (vertices) of each triangle intersect with the vertices of triangles of different shapes and sizes. Data is associated with each vertex, including its spatial position, texture, and color.
  • de Puma explains rasterization in gaming: “Low-polygon modeling keeps old or weak processors from bogging down. That way, you can do real-time animation on older systems. Or you can run lots of characters in a scene. Use low-poly modeling in mobile gaming, where high-resolution characters and objects aren’t necessary. In modern games, running on modern systems, high-resolution characters are made with varying levels of detail (LOD); as characters get farther from the camera, their detail drops. They shed polygons. Their texture resolution drops as well.”
  • Ray Casting: This is a fast technique that detects visible surfaces. The 3D artist allots the location and sets the point of view, which usually incorporates a 60-degree field of vision. Within the virtual space, the artist positions light sources. Light rays trace individually, and ray intersections are determined. Based on these intersections, what is visible based on the POV is determined.
  • Ray Tracing: By tracing light paths as pixels in an image plane, this technique simulates how it meets with virtual objects. Ray tracing is slower than ray casting.
  • Resolution Optimization: 3D rendering image resolution depends on the number of pixels used to create the image. The higher and denser the number of pixels in the image, or pixels per inch, the sharper and clearer the final image will be. The resolution depends on how realistic the image needs to be.
  • Scanline Rendering / Wireframe: This is an algorithm for visible surface determination. Rather than scanning on a pixel-by-pixel or polygon-by-polygon basis, it scans an object row by row.
  • Shading: Shading is a rendering process that computes the color of objects in a scene from a given viewpoint. An example of shading is texture mapping.
  • Texture Mapping: Texture mapping defines surface texture, color or high-frequency detail. It vastly reduces the number of polygons and lighting calculations when constructing a photorealistic scene in real-time.
  • Transport: This technique displays how light in a scene moves from one area to another. Visibility is the main factor in light transport.
  • Z-buffering: Also known as depth buffering, z-buffering helps determine whether a complete object or part of an object is visible in a scene. It is used in software or hardware to improve rendering efficiency.

How to Composite a 3D Render

A composite is a post-3D rendering step. The process assembles render passes and layers. Along with adding realism, it is a time- and money-saving step since it adjusts images faster than rendering.

Compositing Examples

This sequence illustrates how different layers are composited from the raw render through a final composite.

 

Here is a sequence from the video game Grand Theft Auto that shows how data sets can be manipulated to enhance photorealism:

Courtesy of Enhancing Photorealism Enhancement, Stephan R. Richter, Hassan Abu AlHaija, and Vladlen Koltun

Where and How 3D Rendering Is Used

Architects and interior designers were the first to popularize the use of 3D rendering in the 1980s. Today, every industry, from advertising to scientific research, uses 3D rendering to persuade, entertain and educate audiences.

Gregoire Olivero de RubianaGrégoire Olivero de Rubiana is a Managing Partner and Cofounder of The Full Room, a French-based agency that creates 3D visualizations and CGI for living and home retailers. “Our clients use 3D renderings instead of photos,” he explains, “3D provides more product images that are also ready for future use, faster and more economically. The process creates customer intimacy, and in turn, accelerates sales conversion.”

de Rubiana also points out that the ability to use 2D images and enhance them enables even greater realism and a desire to make the object or environment their own. “The point in either 100 percent 3D or a mix of re-engineered 2D and 3D is to invite the customer to travel into a beautiful interior or spectacular exterior for inspiration,” de Rubiana offers. “Often the exterior creations are so successful, clients ask, Where is that place? and actually, it’s nowhere and anywhere!”

3D Rendering Examples

Many industries such as architecture, retail, and medical use 3D rendering to visualize realistic objects, sell products, entertain, teach or engage. 3D rendering also generates believable people, places, actions, and objects that can only exist in imaginary worlds created in movies and video games.

See a variety of ways organizations use this technology in our 3D rendering examples article.

3D Rendering Best Practices and Level of Detail (LOD)

Best practices to optimize scenes focus on how to speed up rendering while still making objects appear realistic to get the highest-quality images at the fastest speed.

“The dead giveaway of novices in the 3D world are those who can’t quite create a photorealistic scene is how they work with lighting and detail.” explains Benoît Ferrier, CG Artist and Sofa Manager of The Full Room Studio.

Ferrier offers lighting and detail tips for those new to the field include:

  • Moderation in Lighting: For 3D interiors, lighting is an essential factor in creating realism. Overloading a scene with irrelevant light sources can ruin a composition. We often make the most of natural outdoor daylight when we shoot products. For night scenes and studio pack-shots (still or moving product shots), we rely on the standard three-point lighting setup (key light, fill light, and back light) just as in photography. For a more natural look, moderation is the key.
  • Keep It Soft: Symmetry, hard angles, and straight lines are a core feature of ‘counterfeit’ imagery. Unevenness and jagged corners or any kind of patina is the way to emulate the real world.
  • Know Your Tools and Be Observant: Just because there are effects available doesn’t mean you should use all of them. Misuse of effects or filters as grain, depth of field, and chromatic aberration to mimic photography are beginners ‘tells.’ Look at spaces and objects in the world around you, and based on distance, be careful with the depth of detail in modeling and texture.

Rubey provides more tips to save computer time and render faster:

  • Lower Polygon Count: Work on getting the model’s geometry to be less expensive in terms of the total number of polygons in a scene. For example, if there are parts of the model that you will not see in the render because of the angle of the camera or if there are parts or products that are farther away from the camera, you can hide them or use low LOD. LOD is when you build the same product in several formats: low details or high details to render from a closeup.
  • Use LOD for Textures: Another option is to use the LOD for textures. Just like in computer games, the exact product could show a high-quality texture map if rendered from closeup, or low/small texture maps if rendered from afar.
  • Lower the Number of Everything: The more objects you have, the more special effects and lights you have and the more computing time it takes to create a single frame. One box with one thing is calculated quickly as opposed to a forest with many trees or characters. The exact size of the image will take much more time to render. You can optimize the models themselves.
  • Stick with a Midlevel Polygon Count: When you make a model in 3D, it is made from small triangles. Let’s say you’re creating a ball. With a 40,000 polycount, it will look like a disco ball — lots of facets. With a million-polycount, you do not see any facets. So, you impact the level of smoothness and realism by the polycount. But it can take a long time to render a million-polygon object. A midlevel polycount is used for objects for efficiency. A best practice is to have an object made up of no more than 60,000 polygons. The key is finding the balance between speed and realism.

Ferrier and his group of artists use state-of-the-art systems that are constantly evolving, just like the industry. Ferrier’s team generates in-house innovations to make the entire process faster and the results more realistic. “The toolkit may change, but not the artisan approach and cutting-edge creativity we apply to every project,” Ferrier stresses.

History of 3D Rendering

Before there were computers, manually drawn 3D renderings were the standard in the arts, engineering and sciences to communicate dimensional reality. Thanks to the trailblazers in 3D visualization, we have made significant advances in every period since the 1800s.

History of 3D Rendering

  • The 1800s: Industrial Revolution: The machinery that changed the world through industrialization were all rendered in 3D before they were put into production — for example, the dimensional engineering drawings of James Watt. The inventions created from dimensional drawings include the power loom, the steam engine, electric generators and the incandescent lamp.
  • The 1900s: Matrix Mathematics: Arthur Cayley developed the algebraic aspect of matrices in two papers in the 1850s. In computer graphics, matrices are fundamental to manipulating 3D models and their projection onto the 2D screen.
  • The 1920s: Bauhaus: The art school founded by Walter Gropius changed three-dimensional spaces representation. Even laypeople could understand how space was used in proposed buildings and public spaces, although they still created those images by hand.
  • The 1950s: First Digital Image: Russell Kirsch and his team developed the first programmable computer, the Standards Eastern Automatic Computer (SEAC). SEAC consisted of a drum scanner and data program to feed images into the computer. A photo of Kirsch’s three-month-old son Walden was the first image scanned in 1957.
  • The 1960s: Computer-Aided Design (CAD) Systems: Patrick Hanratty is known as the father of CAD, which he developed while he was with General Electric. CAD uses computer systems to create, modify and analyze designs. Many other graphical systems followed, including Ivan Sutherlands Sketchpad system to model 3D objects.
  • The 1970s: 3D Solid Modeling Software: Rendering took off when Martin Newell used 3D visualization and rendering to create the “Utah Teapot,” the icon of 3D rendering.
  • The 1980s: Binary Space Partitioning: Binary Space Partition and Binary Space Partition Trees are the brainchildren of Henry Fuchs, Zvi Kedem and Bruce F. Naylor in the 1980s at the University of Texas. The structure of a BSP tree efficiently provides information about space and objects in a scene. Other BSP applications include ray tracing, collision detection in 3D video games and other complex spatial scene applications.
  • The 1990s: Modern Modeling/3D Printing. The 1990s saw rendering technology take off thanks to better software and increased computing power and speed. Toy Story, the first film presented wholly in 3D rendered graphics, caused a revolution in Hollywood. Video games advanced quickly, too, from pixel art to full 3D renderings.
  • The 2000s: Augmented and Virtual Reality: In the new millennium, 3D graphics became ubiquitous in advertising, entertainment, science and online shopping. The giant leap forward was in augmented, virtual and mixed reality visualizations, allowing the viewer to enter a visualized experience more fully.
    “As for the future,” states Rubey, “if you consider that the first Pixar movie released in 1995, we’re already creating real-time games that look better in many ways. As technology and real-time rendering get better, Pixar and post-production studios are also pushing their boundaries. I don’t know how fast it will happen, but the gap is getting closer, and it’s becoming difficult to see the differences in many areas in real-time vs. offline rendering.”

Benefits of 3D Rendering

3D rendering has many benefits: quality visual communication, the ability to show multiple viewpoints, precise lighting and exact specifications, and the opportunity to explore and design at a low cost.

  • Fast Concepting: 3D renderings provide a level of detail and accuracy of scale over a physical or two-dimensional model. 3D rendering provides a sense of realistic perspective and scale for spaces, products or experiences.
  • Quality Visual Communication: Clear visual representations for buyers or clients help sell your concept and lower returns if you sell a product.
  • Show Multiple Viewpoints: The ability to see an object in multiple positions and perspectives allows the viewer to experience the rendering as it would appear in real life from every angle.
  • Precise Lighting: You can control the outdoor and indoor lighting cast upon your product in real life.
  • Accurate Measurements and Specifications: When customers know an object’s dimensions, they are better equipped to buy products or create or plan in virtual spaces — one of the best uses of 3D rendering.
  • Explore and Design at Low Cost: Customers can generate ideas and explore the outer limits of imagination through the power and flexibility of 3D rendering.

Challenges of 3D Rendering

The challenge of 3D rendering are creating convincing realism in a reasonable amount of time. The main issues to overcome: the model itself, texture and materials, and lighting.

Challenges include:

  • Model: The model must look realistic in terms of proportions, size, and details.
  • Textures and Materials: If the textures and materials are not high quality and realistic, it will not matter how accurate the model is; it will lose the realism.
  • Light: This is usually the most neglected factor, as many are not aware of its importance. Since we can usually see when a model is not correct or when a texture is not realistic, most of us notice that something is off, but it isn’t easy to understand that it’s due to the light. People think that in 3D, all you need to do is add light, and once you see the product, that is a common mistake.
    Achieving good lighting requires an experienced artist who is a professional at lighting and who knows how to bring out the correct details of the render. The artist should know how to create an environment and how to add a feeling and a story to the scene by using and adjusting the correct lights.

How Long Does 3D Rendering Take?

Simple images can be rendered quickly in 3D, while an action sequence for an animated film might take weeks to produce. Factors that can affect rendering time include hardware, technique, scene complexity, artists’ skill, and the final output requirements.

How Much Does 3D Rendering Cost?

The 3D rendering depends on the scale of the project and the level of detail. Prices can begin in the low hundreds for a single simple conceptualization and up to many thousands for large-scale projects for big corporations.

How to Streamline 3D Rendering

The most important factors for achieving good renders are pre-production planning and robust modeling. If you are skilled enough and have appropriate hardware and software, the process runs smoothly. You can also use professional services to expedite the process.

Efficient 3D rendering is all about keeping the vision in mind, having a straightforward process, and understanding the end goal. But there are limits to what good planning can do.

Depending on the size of your project, do you have enough staff, computing capacity and the right software? Does it make sense to make additional investments in people and technology? If you have a single workstation or even many of them, it may not be enough to deliver your render if time is of the essence.

Cloud-based services can provide more computing power, so you won’t have to worry about file size or using external drives, and design expertise can save you hassle and time.

Reusable Rapid 3D Renderings for E-commerce

Marxent can help you rapidly render your products with the 3D Cloud platform. Design with mid-poly models that render into high-quality 3D renders. Marxent product visualizations and 3D asset management for eCommerce deliver results fast and efficiently. Our 3D Room Planner spaces render with no human intervention. We don’t just offer speed; we also provide quality.

Ready to get started with a 3D project?

For a complete guide to launching a 3D project, visit our 3D Project Planning Resource Center or contact us to request a 3D project consultation.