Feature: Displays


defi ned, followed by wireframing. Once wireframes are completed, feedback should be collected and then a new fl ow built. T is is an iterative process that must continue until the interactivity with these low-fi delity fl ows feels right. From there, designers and artists can start improving the look and feel, with the help of tools such as Photoshop, Sketch or other 3D soſt ware. Each team has its own preferred tools,


so it is becoming increasingly important to be tool-agnostic when working with 3D content, but also to be able to adapt to other platforms quickly. Easily switching between programs like Maya, Cinema 4D, Blender or 3ds Max allows more fl exibility for the project, but also improves collaboration with other team members during the project. In the end, what matters most is that the 3D content’s geometry and UV maps can be manipulated before being imported to the embedded pipeline, where they are further optimised by shaders for real-time rendering. Altia’s customers oſt en choose Maya


and 3ds Max, since these allow complete control over the meshes and geometry, enabling precise control of each vertex. Other tools may off er diff erent advantages, such as better renders, but those features are not useful when it comes to embedded targets. T e fi nal look and feel of the scene are highly dependent on the shaders. In tools like Altia Design, custom materials and shaders can be used to set and tweak the visual output of the scene. A balance between diff erent anti- aliasing methods and post eff ects can be set to make sure everything looks great while keeping the frame rate as high as possible. During design, 3D content can make


the embedded code and display go awry. To ensure that the interaction and display are correct, the testing needs will grow exponentially, hence it’s advisable to use an automated test system as early as possible. Such a system will ensure the GUI performs as expected, even when parts of the interface might dramatically change. T e strength of automated testing is that once set up it is very cheap to re- run tests frequently.


EV charging station screen example


real-time rendering, on the other hand, would have fewer than 100,000 polygons and will heavily rely on textures; some elements take a lot of time to render in real time, and some would be impossible to render at all without a workaround. Creating a low polygon (“low poly”) model requires tremendous skill, and is normally done from scratch, rather than use a high polygon (“high poly”) model as a starting point. Good-looking low-poly models


sometimes can’t be optimised by just reducing the number of polygons of a high-poly version.


Polygons When developing real-time 3D, it is important to remember that the embedded hardware of today is typically ten years behind the silicon found in most desktops and game consoles. T at means that, even though designers might spend days polishing a 3D animation, signifi cant optimisation must happen for that GUI to run effi ciently on current hardware. Simplicity is key when every pixel must be calculated in 16ms, so it is recommended that designers hold off on heavy polishing until the application achieves a frame rate of 30 frames per second (FPS) or greater. Creating 3D renders varies tremen-


dously. On one end is a real-time graphics application where each frame must be rendered in a fraction of a second, on the other there are physically accurate scenes which take hours to render. T e fi rst approach requires a lot of power, whereas the second needs a lot of memory. On a 3D system, product development teams have neither – so creativity is a must. A 3D model for photo-realistic


rendering will have millions of polygons and use very little texturing; here, rendering is done by calculating the real-world parameters and light bounces, known as “ray casting”. 3D models for


www.electronicsworld.co.uk April 2023 27


Going 3D 3D for embedded targets is usually on the lower end, where everything needs to be optimised as much as possible to allow good performance and frame rates. T e performance on the target depends mainly on the scene size, number of polygons, complexity of shaders and usage of additional eff ects like post-processing. While the number of polygons is usually considered the main indicator of scene complexity, this assumption can be misleading. It is not true that a scene with ten times more polygons would render ten times slower. T e shaders play a very signifi cant role in performance. Knowing how to write custom shaders and optimise them is a very important skill when it comes to 3D for embedded targets.


Consumer electronics device monitor


Page 1  |  Page 2  |  Page 3  |  Page 4  |  Page 5  |  Page 6  |  Page 7  |  Page 8  |  Page 9  |  Page 10  |  Page 11  |  Page 12  |  Page 13  |  Page 14  |  Page 15  |  Page 16  |  Page 17  |  Page 18  |  Page 19  |  Page 20  |  Page 21  |  Page 22  |  Page 23  |  Page 24  |  Page 25  |  Page 26  |  Page 27  |  Page 28  |  Page 29  |  Page 30  |  Page 31  |  Page 32  |  Page 33  |  Page 34  |  Page 35  |  Page 36  |  Page 37  |  Page 38  |  Page 39  |  Page 40  |  Page 41  |  Page 42  |  Page 43  |  Page 44