A lot of this week was spent on attempting to find ways to accomplish our desired look and feel for the Neon City – we had a spoken checklist of things we’d like to include in the city to make it feel more active and alive. However, after the last four days of working on test environments, I’ve come to the conclusion that working with blueprints is less kind than I originally thought it was. While I understand C# and there are things I’d be able to code without much effort in Unity, using blueprints in UE4 didn’t yield the results I entirely hoped for.
I spent a good two days attempting to lerp a material on an object from one to a second. We wanted to do this to have the dry roads/surfaces lerp to the wetness/puddle vertex material. I quickly discovered that there is a large difference between textures and materials in UE4; people handle them differently, however sometimes the use the terms interchangeably, which is the most frustrating part of my learning process. I feel like this is something that should be achievable, but because of the way UE4 seems to handle textures and materials, I just couldn’t find a way to do it through blueprints. Maybe my answer is somewhere out on the internet, but I had to go to the team and we needed to make a final decision.
However, the things that have now been accomplished:
- Rain particle system – although, captured in the video it looks a lot brighter than I intended. I think I may need to up the opacity to make it appear less intrusive in the scene.
- Day/night cycle – this is timed in a 1 minute cycle, so each side lasts for 30 seconds. We can adjust this however we all see fit, which is nice and will allow us to showcase all the pieces of the scene at the best time of day.
- Lights coming on at night – this is timed rather easily via blueprints, however it can be memory intensive if we have a large light system throughout the whole scene.
- Vertex puddle/wetness painting – I think this does look fantastic (although, again, it looks less high-quality in the text video than it does in the real time test).
- Camera work – this was pretty easy to put together and the camera angles will need to be decided on by the group once the whole scene has been built. Our main goal will be to show off as many of the hero models by the team in the best way possible.
There are now a few new things I’m going to redirect my anger/frustrations to after the failed attempt of lerp’ing the materials.
- If we end up having a MVP with rain all the time – which is our current solution to my material problem – then we’re going to incorporate a lighting/thunder storm and a change between light/heavy rain throughout the day.
- I would like to look further into the particle systems to include steam (which will come from grates and and exhaust vents from the buildings), perhaps some fire (from a steel trash can down an alley), and sparks (which will come off the neon signs due to heat during the heavy rain).
- We’re going to have to look into ambient lighting and fixing any inconsistent shadows/lighting during the day/night cycle. I did notice that some of the darkness still feels too artificial, so perhaps there might be something better out there than using the built-in skybox.
- Of course, there are the animations of the various objects we’ll be importing into the scene. We’ll tackle those well into the production phase.
I’m still happy with our testing; it seems like it’s at a point where I’ll be able to focus on my technical model without feeling like I’m ignoring my programming/game design duties for the team. I think the next step will be to draw out some concepts of a street layout we’ll be able to follow, which I’ll do on graph paper and show to the group for their opinion.