An extensive interview with a seasoned software developer
Steam Next Fest events are always a good way to discover and interact with in development games from studios of any size. This is true for flight games as well. At Steam Next Fest February 2024, Blue Sky Aces by Curious Dog Games immediately caught my attention. World War One era flight games / simulators are uncommon these days. My first impression of the public demo for the game was good and I find myself returning to it somewhat frequently as work by its developer continues. The demo has had some notable changes since I last wrote about it. I strongly suggest trying the demo yourself if you haven’t since February.
I recently joined the Curious Dog Games Discord server while researching this project and reached out to the developer for an interview. They were kind enough to provide us with screenshots of a new build of the game due to come out shortly after this interview.
Hello! Thanks for joining me in this interview. I’m interested in learning more about you since there is not a lot of information about you out there on the internet.
Firstly thanks for giving me this opportunity. My name’s Brent, and I’ve been making games under the Curious Dog Games name for a little over 3 years now. By day I’m a husband, and father to two great kids, working as a software developer. By night, I’m a part time game developer working on what will be my first Steam game.
Nice to meet you. Starting at the beginning, is game development something you had planned on trying when you were younger?
Yeah it’s something that’s been a dream for a really long time, but it’s only really become possible in the last 5 years after I discovered Unity by accident while installing Visual Studio. Like a lot of people, it was something that I first got seriously into while studying, but the realities of life quickly put an end to it, and I settled into a normal software development job for almost 20 years.
What inspired you to try your hand at game development?
My family’s first computer came with two large manuals, one was MS Dos which I knew of, and the other was something called GW Basic, which I had no idea about. One day curiosity got the better of me and I read the GW Basic book and quickly realized what it was, but while I tried making a few basic games, I didn’t get very far as it was a time before the internet, You Tube, Google, and all those resources that we take for granted now. It did however lay the foundation which would come in handy later in life.
After some initial research about Curious Dog Games I was taken over to your Itch.io page and saw that you have a few games completed. Has game development become more of a focus for you in recent years?
Yeah probably only the last 5 years. I’d got a bit disillusioned with the job I was in, and wanted to try something different, but at the same time I didn’t want to start again from the bottom as I have a family to support. I really didn’t know what that something was, but I figured that doing nothing wasn’t going to change anything. At this time I discovered Unity by accident, and I instantly knew what I was looking at, this was the thing that I’d tried to build so many times while I was in my early 20’s. Back then I just didn’t have the maturity or experience to realize the dream, which was only made harder by the limited tools available at the time.
Around 3 years ago I decided it was time to leave the company that I’d worked at for the previous 19 years, and by pure chance a Unity development job came up locally, so I figured I had nothing to lose from applying. Fortunately I succeeded in getting the job, and now I work on a 3D training simulator for earthmoving machines during the day, and make a flight simulator in the evenings. This combination has been really beneficial for both, as what I learn in one can usually be applied to the other.
How did you start creating the concept for Blue Sky Aces?
I made a WW1 game called Mech’s take the Somme for a game jam back in 2021, and as part of it I modeled a Fokker Dr1 triplane as one of the enemies. After this project I decided to have a go modeling a Sopwith Camel, with the idea that if I made a few, then I could create an asset pack to put on the Unity asset store. Somewhere throughout this process I got distracted from this asset pack idea, as I wanted to see if I could get them flying within a game. That led to seeing if I could create AI to fly the planes too, and from there the game was born.
Looking back at other indie flight game development projects over the last four years, many of them focus on modern air combat with fourth generation combat aircraft or fictional future settings with fantastic super fighters. Why create a World War One era flight game?
I grew up when fourth generation aircraft were at their peak, and I love all the cool technology, and the raw power, but the most fun flight simulators I played as a kid were the WW2 ones. While guided missiles that can take out an enemy from dozens of miles away are amazing, I much prefer the feeling of getting up really close to an enemy in a proper dogfight. The reason I chose a WW1 era game was largely pragmatic, the period has many of the same attributes that I like, but the market isn’t as saturated. There’s also something I like about the simplicity of the aircraft of this period, where someone with a basic understanding of aerodynamics, some good woodworking skills, and a lot of perseverance, could probably build one in their garage. That said I wouldn’t recommend anyone try it, building a game is way less problematic for a marriage than an actual plane, and a whole lot cheaper too.
That is a good point about this specific type of genre not being saturated. What are some considerations that you need to take into account when creating a World War One era title? Is adhering to the technological restrictions of the era a challenge when trying to maintain high quality game play?
I do find getting reliable technical information for the period challenging, partially due to the time that has elapsed, and the limited science of the day, and also because the second world war tends to overshadow the first, so there’s a lot more information on it. That’s not to say it isn’t out there, but it just feels much harder to find. We’re trying to take a pragmatic approach with balancing technological restrictions and game play. If we have evidence of a technology being trialed during the period, then we’ll include it if it enhances the game play, even if in practice it was just a prototype, or saw limited use. Part of the justification for this was that there seemed to have been a lot of experimentation during the period, even by individual pilots, so it seems fitting to give players the ability to experiment with these technologies as well.
Is this where the idea for the upgrade system introduced in May 2024 comes from?
It did play a part, but the main driver for the upgrades was from player feedback, people wanted more depth, and something to work towards. It was also the realization that people play games like this very differently, they’re just motivated by different things. Some like the history and technology of the period, other’s just like to collect stuff, and some just want a sandbox to play with. The upgrade system just seemed to offer something to all of these groups, as well as adding some replayability.
The way aircraft and missions are unlocked is very interesting. I cannot think of another flight game that does it the way Blue Sky Aces does it. As players complete missions, the in-game date moves forward and that unlocks more content?
Yeah, the theory is that as you progress through the war more aircraft and missions will become available, as well as the types of threats that you’ll face. This is currently quite limited in the demo as I didn’t want people to have to grind away on the same few missions just to unlock a couple more, but it’ll happen slower in the full game. Like with many wars, technology progresses so quickly, and the first world war was no exception, so it doesn’t make much sense to put late war aircraft like the Fokker DVII against early war models like the Airco DH2. That said, there’s nothing stopping players from trying to play the later missions with earlier aircraft once they’ve unlocked them.
I admit that at first I was unsure how this type of progression would fit, but during my time with the demo I can say it is working out so far. How do you plan on introducing missions going forward? Will there be a set single player campaign or will all missions be standalone like they are now?
The current plan is to split them into different groups, training or tutorials, single missions, dynamic campaigns, and mission builders. The first two are similar to what you see in the demo now, but with just more of them. These are handcrafted by me, and have global leaderboards that players can compete with each other.
The dynamic campaigns is one I’m really looking forward to implementing, but at this stage it’s just a concept. The idea is that each map will have its own campaign, and that the goal will be to either create a localized gap in the enemies defenses, or to wear down their combat effectiveness over time, until the point where friendly forces can breach the enemies defenses. Each campaign will consist of multiple dynamic missions, which will vary depending on the results of the previous missions. There will be a strategic element to it, where you’ll have a limited number of aircraft available to you, as well as a limited rate of replacements. The enemy will have the same limitations, so you’ll be able to decide whether you want to focus on specific objectives like destroying key infrastructure, or broad goals like gaining air superiority.
I’ll be the first to admit that there’s a lot of unknowns to work through to realize the vision, but I believe it’ll be worth it for replayability alone.
The mission builders are going to consist of a few different types such as fighter sweep, ground attack, and air combat. These will allow the player to customize the mission by choosing aircraft types and numbers, selecting targets, etc. They’ll be available for all of the maps, providing players the ability to just jump in and start flying, without needing to worry about tactics or strategies.
Some of the other Skyward Flight Media staff members always think I’m a little odd for saying this, but I think one of the top features of the game is the inclusion of infantry in combat. It adds a rather “human element” that I think is intentionally forgone in many combat flight games. I’d like to know your thoughts on how infantry is presented in the game.
Despite all the technological advancements that happen in war, and all of the amazing aircraft, ships, and tanks involved, you still need boots on the ground to win. This was especially the case in the first world war, where men and horses probably bore the majority of the workload. Then there’s this weird thing where people are reasonably okay with the idea of a game where you can shoot soldiers, but at the same time will not tolerate shooting animals perceived to be pets. The end result is you get soldiers and trucks, but no horses, because I don't want to find videos of planes machine gunning horses! 😄
That said, the main reason for the soldiers is that I wanted to create a world with life and movement on the ground. Most flight simulators put a huge amount of processing power into making the ground look realistic, and it usually looks great from a 1000 feet up, but fly down low and you quickly find out their worlds are pretty empty, and the textures quickly start to pixelate. I’ve tried to do the opposite, I want you to be dogfighting in the treetops, in fact I’ve deliberately limited the altitude to around 3000 feet (1000m) to encourage it.
Another factor is that armies of the period weren’t mechanized, so I can’t have large numbers of vehicles roving around the battlefield providing that sense of life on the ground, so it only seems fitting to have soldiers given that most of the war involved men facing off against each other in trenches on the front lines.
The first World War was certainly defined by the massive amount of infantry engaged in combat. I did not consider it from this point of view most likely because I am so accustomed to other games asking players to pick out armored vehicles as the main representation of ground combat. What are some of the hardest things you have solved development wise in Blue Sky Aces?
That’s a difficult question as there’s a lot of time that’s elapsed since I started, and so there’s been many challenges over the last few years. Probably the first big challenge was the scale of the world and the performance challenges that come with it. Modern hardware and game engines make it rather easy to build small and detailed worlds without much need for optimization, however the scale of terrain required for flight simulators doesn’t typically work out of the box without a lot of work. This specific challenge was one of the key factors that steered me towards the game’s art style, as I wanted to get on with building a game rather than spending years learning advanced algorithms for rendering large scale terrains.
Another big one was the aircraft AI, as I’d only ever done character and vehicle AI in the past. This was definitely one of the most satisfying parts of the project, and the point where I knew I could actually turn what was then a prototype into a game. This was one of those tasks that had so many layers of complexity to it, from the basics of getting the AI to fly a plane, to teaching it to dog fight, and probably the most challenging which was teaching it how not to crash into each other or the ground. That last one is still a work in progress.
The vehicle system that controls the trucks and trains driving around the world was another big challenge. I’m pretty sure I’ve spent countless hours watching them drive around test tracks, keenly looking for any glitches or problems. It’s rather hypnotic watching large numbers of them, and it provides a lot of insight into real world traffic issues that we all experience on our everyday commutes. Performance was something that was really important for this task, and probably took up the bulk of the time. It’s a real balancing act, as you want their movement to be good enough, but no more, as every millisecond of CPU time is precious.
The other big challenge is actually a lot of little ones, anything related to the physics system is often a real hair pulling experience, whether it be aircraft movement, soldiers moving around, or water collisions. I just have this love-hate relationship with the physics system that has cost me many hours of sleep.
The way you approach updates about the game is very transparent. Besides your YouTube dev log videos that started in September 2023, you also have a text channel development blog in the official Discord server. It is not all sunshine and quick updates either. You openly talk about the difficulties you have run into as well. Why be so transparent?
I’ve been a professional developer for over 20 years, and I’ve worked on all manner of enterprise projects, and I can honestly say that while game development is the most rewarding, it’s also the most difficult type of development. Part of that is because you’re not just a developer, you're wearing many hats, and you’re doing it at the same time. There’s a lot of people out there literally trying to sell the game development dream to people, “do these 5 things to make 6 figures guaranteed” kind of stuff. I don’t want to deceive people into thinking this is all roses, it is difficult, and it takes an extraordinary amount of time and perseverance to keep going, but it can also be extremely rewarding.
Flight model is always important, but even more so when the scenario has some historic ties. How would you describe the flight model to someone that has yet to play Blue Sky Aces?
Believable but approachable is the goal. I want the player to feel like it’s realistic enough that they don’t question it, but without any of the weird nuances that aircraft of the period had such as propeller torque making turning one way easier than the other. One of the key drivers for this is that I want people to be able to play the game without any special hardware.
That said, I do have plans for a more realistic model as well, which would be best suited for players with HOTAS setups, where it’s much easier to make continuous small adjustments that you just can’t do very well with a mouse or controller. This model is likely to have the ability to perform maneuvers that you can’t do currently such as stall turns and spins, as well as additional forces such as propeller torque and wind.
On that note, how is controller/device support looking? Are there any restrictions?
Generally speaking controllers, joysticks, throttles, rudder pedals, and other similar devices are all supported, however you will need to manually setup all of your control bindings. The exceptions to that are XBox style controllers which should just work, but of course you can always customize your controls. Unfortunately I’m really limited in what testing I can do with hardware as I simply can’t afford to go out and buy one of everything. That all said there’s still some work to do to improve the usability of the controls screen to make it easier to configure HOTAS style devices.
Do you have an idea of where your game needs to be development wise before preparing for a full release or early access release on Steam?
That’s a tricky one as my own opinion of what done looks like is constantly evolving. Part of that comes down to the feedback from players, and the expectations that it places on you. Initially you get very little feedback and you think that you’ll add a couple of dozen missions and call it done, but then players start engaging with you and you realize that the whole thing is starting to snowball. At this point in the process it’s quite clear that I need more depth and breadth of content to live up to my own expectations, let alone the community’s. In practical terms that means more aircraft, and the ability to customize them, a wider variety of mission types, and a campaign mode. The last one is probably the key decider for me, as it feels like the big feature that’s still completely missing, whereas the others will arguably never be 100% done, and can always be incrementally improved post release.
Thank you so much for this interview. I’m glad we could get deep into a few topics and really get your thoughts on game development. I’ll be writing about the next big update to Blue Sky Aces for sure. Good luck to you!
Thanks again for the opportunity and support, I really appreciate the chance to show off something that’s been a big part of my life for almost three years now. I’d also like to take this opportunity to thank the other half of the team. Earlier this year I got an offer from someone who saw one of my development logs to model the aircraft. This has taken a huge load off my shoulders and has massively improved the overall look of the game - you only have to look at the early development logs to see the difference. To add to that he’s also extremely knowledgeable about aircraft of the period, which saves me a massive amount of research time. The Discord members have also been a huge help, whether it be offering suggestions, reporting bugs, or just providing reassurance, it’s all been invaluable to the process.
About the Interviewer
Co-founder of Skyward Flight Media. After founding Electrosphere.info, the first English Ace Combat database, he has been involved in creating flight game-related websites, communities, and events since 2005. He explores past and present flight games and simulators with his extensive collection of game consoles and computers. Read Staff Profile.
.png)