Hideout: Start of the Project
When I start a new scene, the first thing I do is gather a lot of references. I gather reference for all the props in the scene as well as inspiration for the mood, lighting, and colors.
For the Hideout environment, I had a concept to start from, which presented me with all the things I needed to include in the scene in terms of props. This helped a lot when blocking in the scene and when I needed to look for references.
In the beginning, I followed the concept closely, which not only helped to figure out what props to put in the scene but also gave me a clear direction in where I would take the lighting, what color scheme to follow and what perspective the scene would be presented from.
I then started my blockout and only worked from one camera angle in order to follow the concept as much as possible. I worked from that angle the majority of the time.
It was not until very late in the process that I decided I wanted several shots from more than one camera angle.
In order for it to work, I had to populate the scene with more props as some areas were very empty, which was very noticeable from the new camera angles.
So I did a second blockout of my scene and filled up the empty spaces with more props.
Throughout the entire process, from reference gathering to the final touches, I would always have in mind what story I would tell in the scene. This helped me when gathering references, modeling and texturing props, as well as set dressing.
Hideouta Props: close-ups
The way I approached different assets for the environment heavily depended on the type of asset.
For the assets with a lot of detail, I would bake down the detail from a high to low poly model.
For the assets without a lot of detail that needed to be baked down, I would use Face Weighted Normals in order to avoid doing high and low poly versions.
Face Weighted Normals is simply editing the normals on the model to make it look smooth and higher res.
This technique allowed me to work faster and saved time since I had so many different assets to complete.
Below I have listed some resources on the subject if you want to look into it:
The cloth, pillows, and carpet in the scene were a way for me to try a different workflow and test what would work best and fastest for me.
At first, I tried to use Maya’s cloth simulation. I didn’t get the results that I was hoping for and having to redo the cloth simulation took a lot of time.
After trying out Maya’s cloth sim and having heard so much about Marvelous Designer I decided to give the software a go. It was very easy to use and could produce very good results in a matter of minutes, which sped up my cloth workflow quite a lot.
By getting the shape and folds I wanted in Marvelous Designer, I simply took it into ZBrush for cleanup and voila, good looking pillows in a matter of minutes.
When I set up my shots in UE4, I always use the Cinematic Camera, since it has a lot of features and you have more control than the standard camera in UE4.
I tend to use and change Focal Length, Aperture, and Focus Settings.
For this scene, I varied the Focal Length from 30 to 50, depending on the shots. It can also be a good idea to go below 20 for a very wide-angle shot, however, I didn’t find it useful for this particular scene. The 30-50 range gave me a lot of room to work with and created some nice, comfortable and intimate shots.
Changing the Aperture and Focus settings was how I achieved the depth of field in the scene.
For shots where I wanted to get a large portion of the scene to be blurred out, a lower aperture was set. I didn’t want any parts blurred out, the aperture was higher.
The Focus Settings allow me to manually set a focus point in the scene and choose what parts to be blurred out and what should be in focus.
There are also debug tools in the camera settings that give you a clear idea of what is in focus.
Lastly, there are also post-processing settings on the cameras themselves if you want to tweak specific shots, however, I tend not to use those as I use a post-processing volume instead and keep the post-processing global.
I used Substance Designer for the wood material on the floor, walls, and ceiling in the scene, the rest was all done in Painter.
My main focus while working on the wood in Designer was on color variation and shape.
I wanted it to feel used and old so that it could fit in an old attic.
The main challenge in working with this wood was to make it feel used not to the extent that it would seem very uncomfortable to walk on.
As this was my first time creating wood in Designer, I tried a couple of different methods for making the wood grain. I found the best one was warping a Gradient Linear with Anisotropic Noise, which was a tip I got from the texture artist Neil Houari.
From there on I warped the grain to match my planks. I then added noises, tweaked the roughness, added color variation and finally ended up with the result you now see.
When working with so many different assets in a scene I had to be a bit faster when working with textures, which meant I had to both create and utilize Materials and Smart Materials.
Instead of creating custom alphas in either Photoshop or Substance Designer, I relied on information I could find from Materials, Smart Materials and alphas/grunges already in Painter.
By doing this, I can use all the information provided with the materials, especially height and normal information. From that, I can then create my own materials and tweak them to fit my prop and environment. From that, I could also make my own Smart Materials and reuse them on different props.
When using Smart Materials I would never advise to utilize them out of the box. I would always change them to my specific needs and make them unique.
Finally, I would like to thank my mentors Sean Marino and Jeremy Estrellado, as well as the DiNusty Empire, for all the critique and feedback on this piece. I learned a lot from all of them, which helped me achieve my final result.