Hello Riftbreakers!
The game development industry is a fast-paced environment. It combines the elements of software engineering, art, and entertainment. All three of those fields are constantly on the lookout for innovation and new creative solutions. As a result, the game industry changes rapidly, and keeping up with all the latest developments is almost impossible. Still, we do our best to keep up with the news and are constantly looking for new tech to put in our games to improve your experience. In this article, we will show you what we do to stay up to date with our ever-changing industry.
The media in this article will serve as examples of techniques we decided to add on the basis of what we learned over time - either by playing or listening to what others have to say. Remember when rocks obscured the vision of your mech during gameplay?
The most obvious and the most important thing is playing new games. EXOR Studios is a team of fifteen people with wildly different tastes. There are no two people in our office who are exactly alike when it comes to gaming habits. Some of us have 8000 hours in Counter-Strike, while others prefer spending 200 hours grinding 100% in Starfield. Most of us have a dedicated gaming rig at home but only have the time to game on the Steam Deck. Some people jump on an opportunity to play the latest AAA titles as early as possible, while others wait for a few patches to come out first. The point is that our experiences vary wildly, allowing us to learn about new gaming trends from various sources.
We sought a solution in other hack'n'slash games. After all - Diablo and similar have been dealing with it for ages. The answer to our problems was a camera object culler that 'cuts out a hole' in props that you can see through. Most action RPGs use a similar technique!
While we obviously enjoy playing games, we try to learn as much as we can while playing. Naturally, we tend to focus on our area of expertise within the studio. Designers analyze the ins and outs of economy, combat, and other game systems. Programmers try to notice possible optimizations and techniques used by other developers that could work for EXOR games. Artists stop and analyze props, the composition of the scenes and try to understand what makes animations tick. By observing the work of others, we can learn what works and what doesn’t. We try to understand why developers make their choices and how they affect the overall experience.
When creating the destructible rocks of the Crystal Caverns biome, we took inspiration from the tried and tested voxel-based games that allow you to dig in the ground freely, like Minecraft. We divided the destructible walls into chunks. Then, we iterated on the size of those chunks and the speed of digging to allow a smooth experience. What you see above is a very early version.
Recently, we have been looking into changing the loot distribution algorithm in The Riftbreaker. We spent countless hours playing various parts of Diablo, Path of Exile, and Grim Dawn, among many other hack’n’slash games, to figure out what we could do better. We analyzed how often a player gets a chance to loot something valuable. We compared the crafting-oriented systems with trading-oriented systems, as well as those where no trading or crafting is present.
After hours of playtests and design meetings that totally did not involve any shouting and contravening examples from various games, we arrived at this version of rock-digging. Much better!
We’ve drawn a couple of conclusions from these observations. First - we need to give you as many chances to find something good as possible. For that reason, we’ve designed additional, smaller loot containers to work in tandem with our Bioanomalies. This will encourage you to explore the world since you will get plenty of rewards for doing so. Second - we probably shouldn’t worry about real-world logic as much as we do. If a rat can drop a Mighty Greatsword of Slaying +14 in some games, why shouldn’t you be able to find an awesome weapon in a bioanomaly? You will see the results of our work when we launch World Expansion III - we believe you will like it.
At times we work on cutting-edge techniques that do not have many examples to follow. When we introduced ray-traced shadows in 2020 there weren't many other titles on the market with this feature. Luckily, ray-tracing is not a new technique and we could learn from good, old analog books.
Sometimes, observation alone is not enough. Luckily, a lot of talented developers share their knowledge on the internet in the form of articles on various websites, their own blogs, or YouTube videos. While many sites aggregate such pieces, such as gamedeveloper.com or various subreddits, it is not always easy to find something interesting to you. That is why we often have to take a shot in the dark and simply google phrases such as “loot system in game X” or “raytracing implementation in Y.” You would be surprised how often it is possible to find a complex analysis of such topics on some obscure sites that see two visits per year. We treasure them!
Using what we learned, we not only developed ray-traced shadows, but also highly detailed, pixel-perfect raytraced ambient occlusion. It utilizes the same acceleration structure we prepare for RT shadows, so the additional rendering cost is negligible.
One recent example is our Volumetric Lighting implementation. While researching the topic, we supplied our own observations and past knowledge with publications found online. The first one was an amazingly comprehensive presentation by Bartłomiej Wroński from Siggraph 2014 entitled “Volumetric Fog: Unified compute shader based solution to atmospheric scattering.” Another useful presentation came from Sebastien Hillaire: “Physically Based and Unified Volumetric Rendering in Frostbite.”
Volumetric Lighting is exactly what we needed to 'glue' the world together. It allows us to simulate the way the light interacts with aerosols, giving the scene a much more natural look.
The authors of these presentations detailed how they achieved realistic Volumetric Lighting in the AAA projects they worked on. Thanks to the data and observations from these documents, we avoided many pitfalls and achieved great results. For example, we learned what kind of resolution we should aim for when creating the light scattering map, and that the quality increase is negligible above a certain point. We also learned efficient ways of getting rid of the aliasing artifacts using jitter - a solution that’s very simple but not easy to come across!
The new lighting method opened up a whole new world for us when it comes to map construction.
Doing your own research will only take you so far - not all knowledge can be shared as an article or a video essay. However, you can fill in the gaps during developer conferences. They are (usually) multi-day events during which people from various studios give talks about their games, the technology they use, or even the organization of work within their company. Listening to talks from both veterans and newcomers to the industry can be very inspiring and allows you to draw conclusions you can later implement in your own work.
One of the new, subtle improvements we're introducing for the new biome is the behavior of liquids and the simulations of waves on their surfaces. It's still work in progress, but you can see a nice wake behind Mr. Riggs that displaces the algae on the surface of this small lake.
Still, the most important part of attending industry events is networking. You can meet your old friends and trash-talk them for not having dynamic shadows in their game (cheers, Robert!). You can also make new friends and learn that they are facing the exact same issues as you - crucial software crashing, illogical technical requirements, trouble with implementing network play (wink, wink), etc. Getting to know these people gives you a chance to learn something you haven’t discovered yet and perhaps brainstorm a solution to your problems together. Such relations are truly invaluable.
Our new system takes into account individual vertices coming into contact with water, which means that each of Fungor's tentacles creates its own ripple and wake as it moves around.
If you are a developer or would like to start your career in game development, here’s the TL;DR:
- Talk to other people in the industry, learn what they do, and keep an open mind. You don’t have to go to conferences to do that - Reddit, Discord servers, and various forums can help you with that.
- Read technical articles and watch technical breakdowns of your favorite games. Not what you like, and challenge yourself to try new things in your work.
- Play—a lot. Nothing beats a hands-on experience with games, both old and new. Try to understand why they did things the way they did.
The waves are not physically simulated but still do a good job in creating an impression of movement on the surface. Here, there are dozens of Stickrid legs interacting with the surface and causing ripples and waves to appear.
It’s impossible to always be the first to know all the novelties in the game development world. However, you can always try to learn as much as possible about what seems interesting to you and hone your skills. It will pay off in the long run. You can always talk to us and ask questions about gamedev on our Discord - www.discord.gg/exorstudios - we’re always happy to share what we have learned so far on our 15+ years-long journey.
EXOR Studios
PS. We started streaming our tests of the multiplayer mode together with our viewers! Join us on www.twitch.tv/exorstudios every Tuesday and Thursday at 3 PM CET to watch developers being beaten at their own game!