Oneiros shared an insight into how they worked on the recent project in Unity using HDRP (High Definition Render Pipeline).
Introduction
One of our latest projects has been made in collaboration with Unity with the aim to showcase their HDRP (High Definition Render Pipeline) through a virtual walkthrough of the newly opened London office, designed and built by M Moser Associates!
With this aim, we worked on the creation of a software usable on PC where it was possible to move freely between the various rooms using mouse and keyboard and where the users are able to interact with some aspects, as switch between a day/night mode, changing floor materials, turning on / off TVs and lamps and have access to BIM information related to the project.
Although the main focus was related to the implementation of a real-time application, another advantage of using Unity in the Architectural Visualization industry is that you can get different types of output from the same scene.
For example, it is possible to make a video in cinematic mode:
As well as a video in Virtual Reality mode:
Static rendering images and of course, as mentioned, the interactive PC build (usable with keyboard and mouse) can be downloaded from here!
Scene Production
In the case of this project, everything was already defined: we have received from the architectural studio the whole 3D file of the office and furniture and also some 360 images that helped us to replicate the feeling and the atmosphere of the office.
One of the first things to do is to divide the process into two different parts. The first one called “single assets” that includes all the furniture and props and the second one called “environment” that includes the office building itself. This is because the two parts require different workflows.
The creation of single assets is generally a standard process. Starting from the high poly 3D models supplied by the architects, we have evaluated each topology and starting from their high poly version, we have built the low poly assets.
Once the low poly is made, we bring both 3D models (high and low poly) in Substance to work on the textures and get the PBR materials.
As for the environment modeling, a good workflow implies the first draft in 3ds Max which includes the main elements such as walls and floors. Then, once imported as .fbx in Unity, we begin the first lighting setup looking for potential lightmap problems.
When verified that everything works correctly, it is possible to go back and forth between 3ds Max and Unity adding geometry details and checking eventual lightmap problems.
Meanwhile, we start to work on the UV maps of the geometry and proceed to the realization of the materials in Substance Designer (unlike with single assets where we usually use Substance Painter).
Talking about the shaders, it is necessary to keep in mind that in a project like this one – a real-time PC build for AEC, – in post-production, you can only have camera effects. It means that almost everything must be done in 3D. So the workflow for creating materials needs to be adapted to this necessity.
In case of a single asset like an armchair, once we create the high and low poly versions and unwrap the last one, everything is imported in Substance Painter where we add details and bake all the maps needed for a PBR material.
HDRP
With the HDRP, there are several new features in Unity that I would define as essential in the AEC industry to make the qualitative leap in terms of graphics rendering.
For example, the planar reflections are fundamental to create the right feeling of reflective materials, while the new lights give a more realistic result and it is possible to have great control over the temperature.
In the HDRP, the “old” standard shader was replaced by the “Lit Shader”. Below, you can find an example of the Lit Shader with the textures produced in Substance Designer.
Lighting in Unity
The lighting setup in Unity requires to differentiate the use of real-time lights and baked lights first. The first ones will be calculated each frame, the second will create a special map called lightmap.
Once the base lighting setup is made, it is important to set the reflection probe and the light probe.
Here an example of with/without a reflection probe:
As mentioned, the idea was to give to the user the possibility to switch from a day scene to the night scene, so it was necessary to create a second lighting setup.
As mentioned, the idea was to give to the user the possibility to switch from a day scene to the night scene, so it was necessary to create a second lighting setup.
When the whole lighting setup is finished, the post-processing camera is the final step. Post-processing is fundamental to simulate a physical camera and have better control over color grading, exposure, etc.
Afterword
Unity has always been a very flexible graphics engine adaptable to different uses.
It is not easy at all to carry out so many features for sectors that may require completely different developments. Our first step, for example, was to divide the HDRP from the LWRP.
On our side, we will continue to push the limits of photorealism in the architecture, engineering and construction (AEC) improving and implementing new workflows and features.
Links
Oneiros, Virtual Reality & Real-Time CG Solutions
Interview conducted by Arti Sergeev
Get down with Quixel’s new groundbreaking surfaces: