Looking at our last roadmap, I’m happy to report that we’ve achieved the three main goals we set. Our goals were: 1) Create at least 6 unique and viable character loadouts, 2) Implement minion stat scaling and improve their AI behavior, and 3) Smooth out visual glitches in the cel-shader. In this developer blog, I’ll break down how we achieved each of these goals. We’re also writing a new road map to share as soon as its ready.
Abilities and Loadouts
Revn now has 18 character abilities for players to craft. With these 18 abilities in the game, there are now at least 6 unique unique for players to equip (a character can have 3 abilities, so 18 / 3 = 6). Our design team is still working on balancing the abilities, while the art team works to polish the visuals and create art for each ability. We designed the abilities to fill different roles, such as damage, crowd control, mobility, support, or utility. For example, the Dash ability is great for quickly escaping a fight (mobility), while Ghost Cloaking turns a character invisible (utility).
Minions
Minion stat scaling was easy to implement, but a little trickier to balance. We decided to give minions an increasing amount of health and damage with each wave that spawns. Finding the right amount to increase by, however, required a lot of testing. We also started utilizing more spreadsheets, to reduce the problem to a solvable math function. I’ll share more about these spreadsheets in another blog.
Cel Shading
Our third objective was to improve the cel shading post processing effects we use in Revn. This proved to be the hardest out of our three goals. I rewrote the shader at least once a week, always hoping I’d find the magic combination that would result in crisp shadows, smooth gradients, and well lit characters and scenery. I started showing screenshots of each iteration to Kyle, our lead concept artist. We’d sit down and play the game for a few minutes, and ask, “Does this look better?”. The answer was usually “Yes, but this other thing is now rendering in a bad way”. Kyle would ask what I was doing differently in each shader variation, but I told him I couldn’t tell him.
I needed someone to look at the results who didn’t know how I got there because I was starting to have a suspicion about how to get the results we wanted. I began to realize that we actually wanted realistic lighting and shading. So I turned off the cel shader, but left the outline-effect enabled. I showed it to Kyle, and he asked “What did you do? That looks pretty good”. And that’s when I told him. So, we aren’t actually using cel shading anymore. We’re still using an outline effect, which results in a stylized look for Revn, without sacrificing realistic lighting, shading, or reflections.
What’s next?
As I mentioned in the beginning of this post, we are writing a new road map. Our main focus right now is improving Revn’s art and in-game visuals, but we have a number of important technical reworks on the list, too. One the main things we need to do this week is upgrade from Unreal Engine 4.19 to Unreal Engine 4.22, to take advantage of all the new engine features we’ve been missing. Once this is done, we’ll have a better idea of what’s next for Revn.
Thank you for your continued support for Revn. If you haven’t already registered for the closed alpha, click here. Or invite a friend. In a few weeks, we will require a large number of players to help us record gameplay footage for our upcoming trailer, and we’d really appreciate any help we can get for this major event.
