Over 3 years ago, I started work on Skirmish Line. You can and probably should check it out at this link to the Steam store page. In this post-mortem, I will be using what I call "The Good, the Bad, and the Unexpected" format for portions of the postmortem. Game development is complex and filled with uncertainty. There are good and bad aspects from various decisions but also sometimes unexpected outcomes, so I feel this format helps highlight the process more clearly. My goal is to help other indie developers by detailing my own journey, and I sincerely hope this helps others who are starting out.
Starting Out
Your Starting Skills
When I started working on Skirmish Line, I had virtually zero programming training. I understood basic concepts such as loops from taking an introductory course in programming, but little else. I also knew that I am a terrible artist. These were my weakpoints starting out.
As a modder for a competitive Company of Heroes mod and an applied math major, I had a strong numerical background. I am able to study a game and break it down into a basic mathematical model. I had also been playing games for most of my life. These skills gave me a strong understanding of designing games, in particular how mechanics interact on a mathematical level.
Money
I have always been a very frugal person. During my university days, I did not own a car, preferring to ride the bus. I had two part time jobs and frequently shopped for discounted items at the grocery store. Along with some savings from before, I managed to accumulate a small sum of money as a student. From there, I took advantage of the stock market, moving the majority of my savings into indexed stocks but later into more risky options. This last idea came from my background as an Economics major. All of this combined yielded about $25,000, which would form the seed money for developing games. To be clear, I am not advocating that you do what I did, but the general idea should be considered. Save money, and if you think you are able to handle it, invest. The end goal is to have enough money to sustain yourself and with enough leftover for business expenses.
First Lesson
There will be a distinct lack of pictures in this article. Unfortunately, it had slipped my mind to take any screenshots of earlier builds of the game. When you embark on your own projects, be sure to consider taking a few pictures here and there. Game development is a lengthy process with a lot to learn. You'll probably make some silly mistakes like that I did by forgetting to take screenshots. Don't be too hard on yourself. This is all a learning process.
The Idea
The entire idea behind Skirmish Line was originally to develop from the grounds up, an improved version of the popular Mud and Blood 2 Flash game. This is not a terribly original idea, but it is an idea with other valuable properties.
First off, Mud and Blood 2 is literally my favorite Flash game of all time. This might seem like a redundant point, but you need to be interested in your project if you are going to work on it for a long period of time. If you're an indie developer and you're working on games you don't even care that much about, what are you doing with yourself? I do mean it. Ask yourself that.
There is a potential audience. Mud and Blood 2 was a major hit, with millions of plays across various websites. Released back in 2007, the game has aged enough that players would want something newer. This plays into nostalgia without infringing on the work itself, as Urb, the developer for Mud and Blood 2, had moved on from the project years ago.
Mud and Blood 2 is a mechanically driven game with a lot of math going on in the background. This fits with my strengths. The game doesn't need 3D modeling and the art can be fairly simplistic that at the time, I thought I could even do myself.
The final property that I considered was the flexibility of the idea. Having played Mud and Blood 2 extensively, I knew what I didn't like about the game but also what else I thought could be incorporated into it. Thinking on another game, Rising Storm, a FPS game set in the Pacific Theater during WW2, I considered the idea of implementing Banzai charges into Mud and Blood 2. I recognized that much of Mud and Blood 2 involved ranged combat and wanted to implement systems that would enable more melee focused situations. From there, Skirmish Line, which was tentatively titled: Mud and Blood: Pacific, was chosen as my first project.
Art
Single Sprite Expression
Skirmish Line uses single sprite expression for most of the in-game sprites. This means there is a single sprite for each a state an object might be in. In Skirmish Line, the individual soldiers have a separate sprite for the different state they are in, e.g. idle, pinned, aiming/firing, dead.
The Good
Single sprite expression cuts down on the number of sprites needed to be drawn. This can be a huge money saver when you scale up the project.
The Bad
The most obvious outcome is that the game won't look as good when you're inspecting an individual element, as there really isn't much going on when you zoom in. For a low budget product, we found this trade-off acceptable.
The Unexpected
When taken as a whole, it actually looks pretty good. A huge melee with multiple units fighting looks and feels good despite most of the fighting involving only 2-3 sprites on the part of each individual unit.
Minimalist UI
For Skirmish Line, I went for a fairly bare bones UI. All unit menus are hidden until accessed. One major departure from the original Mud and Blood 2 game is that there are no UI buttons on the screen itself.
The Good
With such a minimal UI, players get a nice view of the game with little obstruction from the UI. This helps a lot when for players taking screenshots. There is a built in function to hide the screenshot (similar to many other games), but players rarely use such a function when playing.
The Bad
Some players may find the UI lacking, especially new players who are expecting more detailed stats. This comes along with one of the game's core design philosophy, being a minimalist game where players are expected to figure out the game through repeated plays.
The Unexpected
Surprisingly, some players prefer having UI buttons on screen. I ended up implementing a couple of buttons for such players, but this highlights how some players are very GUI oriented, at least in this genre of games.
Design
The 33 Rule
Borrowing Firaxis' 33 Rule, the goal with Skirmish Line was to make a game with the basis of Mud and Blood 2 while changing and adding enough to the game to create a new experience. In essence, a third of the game would remain the same, a third would be changed, and a final third would incorporate new mechanics.
The Good
Being a first project, being able to build the foundations of the game after an existing game was a big help. The original prototype of the game was simply an attempt to copy all of the mechanics of Mud and Blood 2. For a beginner, this cuts out a lot of unknowns and gives a core goal for everyone to work towards.
By working with a game in mind, we can also make conscious decisions to change elements that didn't work well. For Skirmish Line, we abandon the "6 man rule" from Mud and Blood 2, a behind the scene rule (known only by the most hardcore fans of the game) that would induce artillery strikes on the player should their number of non-hidden units exceed 6 on the field. This change radically alters how the game would play out, but we found this made the game more accessible for players.
The Bad
Sticking to the rule means refusing a lot of changes. For Skirmish Line, this means keeping things like single unit selection, the top down view, and also things like the general inaccuracy of weapons, which we felt were a core part of the game's systems. We received a fair amount of negative feedback for this, with some players claiming the game is a clone of the original game with no improvements but also quite a lot of players praising the changes and new additions. Ultimately, sticking to the 33 Rule means some players are going to think the game isn't different enough.
The Unexpected
In the attempt to copy the core mechanics of the game during the prototyping phase, the mechanics actually turned out to be different in nuanced ways. Since we don't have access to the source code of Mud and Blood 2, our own interpretations of mechanics often resulted in very slight differences that would have notable gameplay effects.
In removing certain mechanics, as the aforementioned 6 man rule, we found new problems. The 6 man rule acted as a cap/limiter on the number of units for the player and helped prevent the player from hitting a critical mass of units To solve this issue, we had to get creative, first implementing an alternative, soft version of the 6 man rule in which the enemy boss wave spawn chance are increased as the players have more units on field. When this wasn't enough, we implemented new ways for enemy units to tackle the players' defenses. We added a smoke screen that would deploy in front of a boss wave. Certain enemy units would throw smoke grenades to conceal the movement of fresh enemy spawns. Special waves that adjust accordingly based on the number of units the player has on field.
It took several months of playtesting and experimentation to fix the gap in the design left open by the removal of one mechanic. When we simply tried to implement the 6 man rule itself and variations of, players felt frustrated by the seeming randomness of artillery strikes. Our eventual solution resulted in a variety of new systems that would further distinguish Skirmish Line from Mud and Blood 2.
In-Game Achievement System
A hallmark of the Mud and Blood series itself is the presence of an in-game achievement system called Medals and Ribbons, which require the player to accomplish certain feats in order to unlock various bonuses.
The Good
An in-game achievement system helps add a lot of replayability. While many players already naturally attempt to experiment and try new playstyles, an in-game achievement system can help push the players towards new areas that they might normally miss or simply add new challenges for the player. All in all, this is a fairly low cost mechanism for replayability and can really help extend playtime.
The Bad
The biggest cost of in-game achievement goals is that it is often difficult to design interesting challenges. For starters, goals have to be challenging enough to be meaningful but also feasible enough for players to achieve. When you take into account varying player skills, an achievement system may discourage players. A mixture of easy and hard challenges is an important consideration for player retention.
The Unexpected
Another major decision was to reduce on the number of grindy medals and ribbons. For an example, Mud and Blood 2 had a medal that would require the player to bypass 10000 waves within their first 99 games, requiring the player to average over 100 waves a game. This is further compounded by the requirement that this be achieved within the first 99 games on a player's profile. Players would have to restart their profile if they failed the requirements.
The decision was made to cut the number of games on similar medals in Skirmish Line and to have the requirement be based on a rolling score of the player's last set of games. While many players appreciated this change, a few diehard fans actually want those grindy achievements, perhaps conflating grind with difficulty and challenge. Nonetheless, we made a conscious decision to not push our players into what seem to be an arbitrary skinner box.
Mutator Systems
Skirmish Line incorporated a mutator system we called "challenges", not to be confused with the in-game achievement system. The challenge system is a set of mutators that modify existing rules within the game. For instance, the "Just a Flesh Wound" challenge would reduce the healing rate to 10% of its normal value while providing the player with 2 additional points of resources each wave.
The Good
A mutator system was another very effective and low cost feature. Requiring no new art assets or much in the way of programming, the system added new ways for players to enjoy the game. For what was a very minor system, a lot of players praised the feature, citing it as an interesting addition to the game.
The Bad
Mutator systems run the risk of confining the game space. One of the challenges, "Jack in the Box" became very popular with players. At some point, the mechanism behind this challenge actually came in conflict with a later planned system. Since the challenge was popular with players, we couldn't just remove it. We had to modify the planned system instead.
The Unexpected
For what would be a well liked feature, the challenge system was a random idea that I felt like incorporating on a random whim. From there, I began to take the mindset of actively looking for low hanging fruits that can be implemented.
Campaign Mode
Skirmish Line has an extra feature where players can purchase units beforehand and go on the offensive. Units would be persistent across missions, which themselves can be completed unlike the endless defense missions of the core game. On paper, this all sounds pretty cool.
The Good
Campaign Mode was initially coded in 2 days after having watched a series of tutorial on serialization (how to save your game). Campaign Mode is also an extra feature that could be used to help market Skirmish Line.
The Bad
Building a feature in 2 days on a whim and expecting it to just work is a bad idea. Building a feature in 2 days because you are feeling pressured to make the game work on mobile devices because the core structure of your base game modes don't allow for easy serialization is an even worse idea. Campaign Mode was a feature that was outside of the scope of the project, and the biggest example of a feature that needed to be cut from the game early on.
At the time, I was under pressure to find a way to make the game mobile compatible. Since much of the game's systems were built using coroutines, it was very difficult to serialize the game. This wasn't a major problem on PC since games weren't expected to take longer than 2 hours, with most being under 1 hour realistically, but mobile gamers would expect to be able to play in very brief intervals, thus necessitating either shorter missions or an ability to save the game's progress.
I did a lot of tweaking of Campaign Mode, but it never was quite as fun as the two endless defense modes. Ultimately, it was a feature that probably should have just been dropped, and the development time on it spent elsewhere.
The Unexpected
Steam actually cares about what you put on your store page, at least when it comes to potential false advertising. When we submitted the game for review, we had advertised 3 game modes. One of these game modes was Campaign Mode, which we had disabled on the menu at the time being because we wanted our players' attention to be focused on the two endless defense modes for the launch period. Subsequently, we had to re-enabled Campaign Mode in order to pass Steam's review check before the game could be released on Steam Early Access.
Distinction Mechanics
One of the most important thing when it comes to your designing your game is to consider how to make your game distinct. A general rule of thumb that I have gathered is that there are two primary ways to make your game stand out: 1) Make your game the most polished of its kind and 2) Make your game different enough that nothing else quite matches it. As an indie developer, I can't match the graphical fidelity of most other RTS games. So for Skirmish Line, this means abandoning a lot of what would be considered normal for a RTS game in order to embrace something different.
1) No group unit selection
